A puzzle about eternal life
November 14, 2007 — 11:30

Author: Alexander Pruss  Category: Afterlife  Comments: 45

Suppose Jane and Bob live alone on their planet, hundreds of lightyears from any other people.  They are 20 years old, and they find themselves with a deadly disease that will suddenly and painlessly kill both of them in a year.  They pray for divine aid, and an angel comes to them with the following offer: While they will die in a year, their mental and bodily functioning, as well as that of their environment, will be sped up by a factor of seventy, so that while they will die in a year, during that year they will have lived the equivalent of seventy ordinary years of life. Since their internal clock is sped up, it will feel to them as if they lived through seventy years. Moreover, they will forget that the angel had visited them, and so they will not know that each subjective minute is only a seventieth of a minute.

Question 1: Prudentially, should Jane and Bob take up this offer?  (My intuition: Yes.)  Note that it is not just a matter of it feeling like they get seventy years of life.  They really do get seventy years' worth of learning, interacting, stewarding their environment, praying, growing emotionally, and so on.

Question 2: Are Jane and Bob as well off in this scenario as they would be in a scenario on which their disease is cured and they live for another seventy years?  (I go back and forth. My initial inclination is to say "Yes" or "Almost".)

Suppose we answer Question 2 in the affirmative.  Now modify the case.  Jane and Bob pray for eternal life.  An angel comes to them with an offer: Instead of eternal life, God will do the following for them.  During the next six months, their and their environment's speed of functioning will be increased up by a factor of two, so they will feel like they are living the equivalent of a year.  During the next three months, their and their environment's speed of functioning will be sped up another factor of two, so it'll feel like they are living the equivalent of another year during those three months.  During the next 1.5 months, we get another speedup, so it will feel like those 1.5 months were a year of life.  And so on.  And then at the end of the year they will die.  But it will have felt to them like they were living forever.

Question 3: Is this just as good as eternal life?

Question 4: Do answers to any of these questions depend on the nature of time?

Comments:
  • Christian Lee

    Alex.
    Question 4: Do answers to any of these questions depend on the nature of time?
    I think they all do. I wonder whether they are getting their full seventy years or not. Answering this question seems to me to depend upon the nature of time, and in particular, whether the rate of passage is fixed by this or that.
    The two scenarios may be indistinguishable to them, but “I” don’t want no phony.

    November 14, 2007 — 19:04
  • Alex,
    As to question 4, would it depend on whether time is dense or continuous? If it is dense, then they might come to the point of time that is right before a full year, and then they just wouldn’t be able to speed up any more.

    November 14, 2007 — 20:38
  • Alex,
    I’m not sure that the proposition set expressed at the end of the second question is coherent. After all, what can it mean for me for it to ‘feel like’ I am living forever if at some point I will objectively die? It seems to me that if the latter is true then it will not seem to me that I am living forever, because there will be some slice of time at which i will to be. Conversely, I think that the only way for it to feel like I am living forever is for me to go on living interminably. Otherwise ‘forever’ is just a non-narrow descriptive term, something ‘a innumerably long but still fineite period’.

    November 14, 2007 — 20:44
  • A.P. Taylor

    Correction: that should read “there will be some time slice at which I will cease to be.” Sorry.

    November 14, 2007 — 20:46
  • Christian:
    Here’s another case. They get the speedup for a year, so that they do 70 years’ of activity during the year, and then they get to spend the next 69 years unconscious in a coma. It’s hard to say why this scenario is less valuable than the scenario where they just live for 70 years normally. After all, on both scenarios they live for 70 years, and in both their lives are equally rich in action, experience, prayer, etc. (You may worry that in one case they are wrong about the flow of time. But being wrong about the flow of time is a very minor evil–time is so weird, that most people are wrong about it in some way.) But while I think 69 years of unconscious life has some value, I do not think it adds very much value. Hence, the scenario where they get the speedup but don’t get the 69 years of unconscious life is only somewhat less valuable than the speedup plus 69 years of unconsciousness, and the latter is equally valuable to the 70 years of normal life. If this is right, then it seems to follow that the year of speedup is somewhat less valuable than 70 years of normal life, but not very much less valuable.
    I guess what these cases get it is the question whether what is valuable is only the contents of our lives, or whether the sheer length matters, too.
    Andrew:
    On the scenario as I paint it, there is a time at which I no longer exist, but at any time at which I still exist, I feel just as someone living forever does.

    November 14, 2007 — 21:43
  • Alex,
    I presume that last comment was meant for me (since it was not pertinent to Andrew’s remark about dense/continuous time). So, I’ll respond to your reponse as follows.
    Your reponse seems to be saying that there is something it ‘feels like’ phenomenally to live forever. Now, I am thiry-two years old. It seems to me that there is something phenomenally that it feels like to have lived thirty-two years. Furthermore, there is something about what it is like to be in one’s thirties that one does not know, and presumably cannot know, in one’s twenties (responses to Jackson’s Knowledge Argument notwithstanding). Thus it seems plausible to me that their is something it is like to live forever, but I would deny that any being of objectively finite lifespan could have access to the relevant phenomenal states. Only a being which has in fact lived interminably could have such access.
    Perhaps you have a different phenomenology in mind, but I cannot see what it could be. If the subjective ‘feeling’ of living forever differs only in scope and not in any characteristic phenomenal content, then I think there will be no meaningful difference between the feeling of having lived thirty-two years and having lived forever.

    November 14, 2007 — 23:05
  • Ryan Byerly

    Two comments:
    First, we might want to distinguish between two different phenomenal experiences. On the one hand, there is the experience where it seems that I WILL live forever. Call this Experience 1. The other experience is where it seems that I HAVE lived forever. Call this Experience 2.
    Experience 1 is roughly equivalent to the experience where it seems to someone that she is in the midst of living forever. I don’t think that this experience is equivalent to the experience of eternal life. For, this is actually a rather common experience. Consider the case of your average undergraduate student, to whom it indeed seems that they will live forever or that they are in the midst of living forever. If they in fact end up not living forever, then Experience 1 is not equivalent to eternal life.
    What about Experience 2? I don’t think that the people in your scenario have this experience. This would presumably be an experience had at the END of an eternal existence. But none of the people in your scenario ever reach this end. Indeed, it seems that this kind of end is unreachable.
    Secondly, I find it doubtful that experiencing living forever and experiencing eternal life are the same thing, at least if we conceive of eternal life in terms of the full meaning which this phrase has in the Christian tradition. Suppose for the sake of argument that the biblical writers indeed did think that persons sentenced to hell experience punishment forever. We can infer that these people live forever. But surely we would not say that they experience eternal life. At least, the biblical authors would not stand for this.

    November 14, 2007 — 23:53
  • Christian Lee

    Alex: “I guess what these cases get at is the question whether what is valuable is only the contents of our lives, or whether the sheer length matters, too.”
    Agreed. Here is a worry. If the value of some experience is a function of its intensity and objective duration, then the life with the speed up has much less value in it than the normal life.
    But you want to characterize these lives as indistinguishable in their value relevant features. And there is something to be said for the claim that if experience e is indistinguishable from e’, then the value e contributes to one’s welfare is the same as the value e’ contributes to one’s welfare. We then get an inconsistent triad. And I have no idea how to solve it.

    November 15, 2007 — 3:27
  • As to question 4, would it depend on whether time is dense or continuous? If it is dense, then they might come to the point of time that is right before a full year, and then they just wouldn’t be able to speed up any more.
    Andrew, I’m not following. If time is dense, then there any two temporal points will always have a point between them. So no matter how far I’ve gone short of the endpoint, there will always be a point between where I am and the endpoint. If time is atomistic, what you’ve said follows, but if it’s dense or continuous it doesn’t.

    November 15, 2007 — 7:09
  • Alexander – like Ryan, I have doubts about the ‘feeling of eternal life’.
    In fact, it might be easier to focus on the ‘feeling of mortality’, with which most of us are acquainted, that sense that our lives on this world are limited, that we will run out of options, that if you don’t do it now, perhaps you never will. To feel a sense of immortality would have to include not having this sense of mortality.
    Of course, we weren’t born with a sense of mortality, and so memories of the time before we were aware of our mortality can give us a sense of what it is like to have a sense of living forever. Perhaps we can achieve a release from the sense of mortality by learning somehow to ‘forget’ or better, to disregard the fact of our death, focussing only on the present moment.
    (I’m hoping that most readers will have had such an experience, and so will understand what I’m saying. I’ll also mention that Margaret Donaldson’s excellent book ‘Human Minds’ includes a good discussion of the way perception of time changes as we develop.)
    But of course, what we disregard doesn’t necessarily go away. It is all very well to sit under a tree and attain a sense of eternity, but if I allow myself to think I’m immortal when I cross a busy road, I have fallen victim to a comforting but dangerous illusion.
    What of Jane and Bob? The interesting question is whether they are entitled to a sense that their ‘earthly’ life will never end. (What is the name of their planet? Substitute appropriate word for ‘earthly’). In order to make sense of the question, we certainly need to know about the nature of time. In fact, I’d go further; I want to know what theory about the nature of time allows us to make sense of the question.
    Here is my worry. I’m being told that, from my point of view, they die after one year. So I reach the point where I experience them dying. And if I experience them dying, then what I experience is their death happening to them.
    But it seems what we are saying to them is ‘You will never reach the point where your death happens to you.’
    But then haven’t Bob and Jane been doubled somehow? There is Bob and Jane whose death I experience, and Bob and Jane who will never reach that point. How can this be?
    I do have some thoughts about how different theories of the nature of time might handle this: it’s a question of what it really means to ‘speed up’ the experience of time for an individual. I’m wondering whether some theories about time would impose an automatic speed-limit.
    But, (as you probably guessed) the limits on my own experience of time mean I have to break off here for the moment.

    November 15, 2007 — 9:52
  • I’m deservedly taking a lot of flack for my talk of what it feels like to have eternal life.
    I was using “eternal life” as just a synonym for “living forever”. As Ryan points out, that was a bad idea, because biblically eternal life is much more than that.
    Ryan’s distinction between the experience of having lived forever and of having eternity ahead of one is very helpful, and I agree that I shouldn’t be talking about either one here.
    Rather, my claim that it feels the same to have the infinite-future-compressed-into-one-year as to have an infinite future has to be made as follows. For any time t within one of the two scenarios, there is a corresponding time t* within the latter scenario such that the phenomenal state that I am in at t in the one scenario is the same phenomenal state that I am in at t* in the second scenario. Moreover, any phenomenal state that I feel in one scenario to have lasted N seconds feels like it lasted N seconds in the other scenario.

    November 15, 2007 — 9:53
  • Ben:
    Good points and good questions.
    I think the idea of speeding the functioning of two individuals and their planet makes some sense on every major theory of time, as long as (a) time is dense, and (b) the rest of the universe goes on at the old steady pace. On some theories of time, condition (b) is not needed.

    November 15, 2007 — 10:13
  • Greg

    So I’m not sure what got Alex thinking about this topic, but I thought that some of you might be interested to know that Roy Sorenson uses similar cases in his paper “The cheated god: death and personal time” to argue against a certain account of the badness of death. The paper is in the April 2005 issue of Analysis.

    November 15, 2007 — 12:15
  • Jeremy,
    Ah, okay, I must have the terminology wrong. I should’ve said that if time is atomistic, blah blah (what I already said above). But if you agree that what I said follows, then the answer to question 4 is that the other questions do depend on whether time is atomistic or not.
    For example, the answer to question 3 is that it would not be just as good as eternal life. Why? Because they cease to exist! Once they get to the moment of time right before the end, it will be impossible to speed things up, and then they will die. I assume that it is better to actually live forever than not. Alex, you pressed the point that for any time at which they do exist, they feel as if they are going to live forever, but feeling just amounts to a false belief. It still remains the case that in one situation, they live forever, and in the other, they cease to exist. Furthermore, just as it’s better to have true beliefs about the external world (that it exists, that my friends exist, etc.) than to have false beliefs, so it is better to have true beliefs that you are going to live forever than to have the false impression (or feeling) that you are going to live forever.

    November 15, 2007 — 14:05
  • oops, I just realized that most of my comment may be obsolete… I read only the comments in response to me and quickly responded without carefully looking at the other comments! mea culpa…

    November 15, 2007 — 14:09
  • Greg,
    The reason this interests me is because on my favorite theory of time (if one can call it that–I don’t like the word “time”) it would make no ontological difference if everything were sped up (i.e., the hypothesis that everything is sped up makes no sense). So if there is an important difference between living 70 years normally and living 70 years in a sped up way, then this is evidence against my theory.
    And, yes, there is the issue of the badness of death. If what is bad about death is the interruption of plans, then the “eternity fitted into a year” scenario is just as good as not dying. If what is bad about death is the alleged future non-existence (which is somehow generally taken to be worse than the fact of past non-existence (which I happen to find pretty terrifying)), then there is a big difference between the two options.

    November 15, 2007 — 16:48
  • Let me suggest a change to the thought experiment. We separate Bob and Mary. Bob still gets the ‘infinity in a year’ option, but Mary gets the ‘true infinity’ option: she goes to a planet where she will live forever.
    Each of them seeks solace by bouncing a rubber ball. Are their situations equivalent?
    Well, Bob knows that there will be a time when Mary knows he is dead, and Mary knows this too. This will be a source of sorrow for them both, and constitutes an asymmetry.
    But, and here is my conceptual problem, how many times will each of them bounce that ball? In Mary’s case, she will go on bouncing that ball forever. Surely, she will bounce it an infinite number of times. If Bob really has ‘infinity in a year’, and he will use that infinity to bounce the ball, then he too must bounce the ball an infinite number of times.
    But wait. Whenever I visit Mary, there is always a finite number that constitutes the number of times she has bounced that ball. In fact, it becomes a family tradition that, even when I’m dead, my descendents will visit Mary’s planet, and when they ask how many times she has bounced the ball, the answer is always a finite number.
    With Bob, on the other hand, after one year, he is no longer bouncing the ball. He is dead, and he was bouncing the ball when he died. But when he was bouncing the ball, what number bounce was he on? If the answer is not a finite number, then there is an asymmetry between his situation and that of Mary: he passed the bounds of the infinite, whereas she never will.
    I’m not sure this is coherent, and so if some view of time implies this is coherent, I’d see it as a reason to reject that view of time.
    Of course, the problem does not come simply from speeding time up. I’m happy with the idea of speeding time up for Bob. I don’t see any incoherence in allowing Bob any finite number of bounces: fit however many billions of bounces into his year. But just because it makes sense to speed time up to the nth degree for any finite n, it doesn’t make sense to suppose that one can speed time up to the point of infinity.
    I note in passing that my argument here is basically a rehash of ideas presented in J. F. Thomson’s ‘Tasks and Supertasks’, Analysis, Vol 15, No.1, 1954, 1-13. Dummett exploits this as an argument against the ‘classical model of time’ in his Is Time A Continuum of Instants (Philosophy, 75, October, 2000). The Dummett/Thomson connection was pointed out by Ulrich Meyer, ‘Dummett on the Time-Continuum’, Philosophy, January 2005.

    November 15, 2007 — 18:53
  • Although they live alone on their planet, hundreds of lightyears from any other people, they might as well (more or less) live next door if they live forever (rather than the original 70 years), because of the possibilities of space-travel. Consequently even if the bad thing about death is the interruption of plans, then this might not be as good as living for eternity.
    Were they more effectively in a world of their own, would there then be a difference? Then I think not, but I also think that the problem of future non-existence would not arise for them then. When the world around us continues without us, still full of human life, that thought makes us sad. But their world continues as a rocky desert. And do we really care about not existing in prehistoric times, or on Mars now?

    November 15, 2007 — 19:43
  • Ben:
    I am afraid I don’t see the difficulty. Now that he’s dead, Bob bounced the ball an infinite number of times. But at each time at which he was alive he had bounced only a finite number of bounces. When he died, he had done an infinite number of bounces. Whether the ball was still bouncing at the moment of death (in this case the first moment at which he was not alive) is underdetermined by the description.

    November 15, 2007 — 20:59
  • Dear Alexander, if you don’t see the problem, it might be hard to explain it, but I’ll try.
    I place a video-camera within Bob’s special time-stream. After he is dead, I play the film backwards. There he is, lying dead on the ground. He rises up, and the ball rolls towards him into his hand. I see him perform the last action he performed before dying, bouncing the ball. But which bounce was that?
    To make this easy, I asked Bob to count as he was bouncing, so with each bounce, he shouts out a number. What number is he shouting out on this last bounce?
    Do I see him shout ‘Bounce infinity’ then drop dead? And do I next see him shout ‘Infinity -1’? But he didn’t begin counting as infinity-anything. He began counting at 1, and went on counting forwards. So what number precedes infinity?
    Of course, if he shouts out a number with each bounce, the bounces will take longer, but then with infinite time to bounce the ball, he should never reach a number too long to count.
    I do see a problem with the camera: how much film did it have? As the year progresses, Bob is moving faster and faster, packing more and more bounces into (what appears to me) to be just a second. If the camera keeps up (as it should, since it is in his time-stream), then it will appear to me to move faster and faster, until it runs out of tape.
    I can think of some possible solutions:
    (1) I set the timer so that the camera will start filming at a very high speed just before Bob’s death – I know when his death will take place.
    Now, with regard to (1), I suppose that you will say that with only a finite amount of film, then either the film will go so fast that it runs out even in the ‘tiny’ amount of time left, or Bob’s infinite bounces will outspeed the film, so that I’ll have a couple of frames where it is blurred, before the final frame shows him dead.
    The atomic view of time would say that, at a certain point, the film goes so fast that every frame of the film corresponds to an indivisible instant of time, and nothing can happen faster than such indivisible instants.
    I know that isn’t your view, but the challenge is to show that this view is wrong.
    (2) The camera has infinite tape. If I can speed Bob up, can’t I also provide infinite tape? And when I play it backwards, I should discover the number of his final bounce.
    or a variant on this:
    (3) The camera has finite tape, but in a loop. Old images are erased and replaced by new ones. So what should be left at the end of the experiment is the most recent images, i.e. the death of Bob, and his final bounce.
    (4) Bill, unlike Bob, has three years to live. He enters Bob’s time-stream, speeding up when Bob does, but after Bob’s death, he slows down to join the rest of us, and we ask him to describe Bob’s final bounce.
    Now I think you will find it hard to describe Bob’s final bounce, or his final and penultimate bounces. If he calls out a finite number, what was it? Jane will eventually do more bounces than that. If he is counting backwards from infinity, when did he change the direction of counting?
    Maybe you can find more reasons why we cannot have any record of Bob’s last bounce, and his death. Perhaps Bill and the camera, if they enter Bob’s time-stream with him, will never be able to re-enter our time-stream in such a way as to give us the required information. But in that case, how can we have Bob’s corpse? The presence of the corpse at least tells us ‘Bob has died’, and it is conceivable that Bob could somehow have recorded information on his own body about the last bounce. (In fact, sticking more closely to Thomson’s paper, the information I ask might not the the number of his final bounce, but simply whether it was an odd or even number. He only had to shout ‘odd’ ‘even’; I don’t see how infinity could be either of those.) But if we don’t have a record of his death, then surely the same reasons would mean we couldn’t have his corpse, in which case does it make sense any longer to say that after a year of our time, Bob is dead? Couldn’t we better describe his situation by saying he will disappear from our time-stream? But I thought that part of the downside of the scenario, for Bob, was the thought of there being a future time when people would say ‘Bob is dead.’

    November 15, 2007 — 23:07
  • re Question 4, it’s not so much the nature of time (assuming it’s not discrete, but presumably that’s up to God) as the nature of infinity that is being presupposed, as Ben says. The eternal life being lived in one year requires an actual infinity, whereas the eternal life being lived endlessly requires only a potential infinity. The problems with the former are subtle but many. E.g. presumably their souls persist and go to Heaven after a year if they were good, just like all souls do. There need be no time (subjective or objective) between the two, so there could be a first instant in Heaven: what do they recall? They recall all their eternal life, say, i.e. each moment was an infinity of (subjective) time ago, and yet the totality of those moments take them up to their present time.

    November 16, 2007 — 4:10
  • …oops, time could be discrete, for your scenarios to be possible. The only requirement is that actually infinitely many moments can be put into one year, of course (my excuse: infinity confuses; also there is a requirement for no intrinsic limitation to how brief a moment of time could be though, and most coherent discrete models of time assume do such a minimum).
    re Ben’s scenario, there was of course no last action Bob performed before dying, no last bounce. There have been many arguments against an actual infinity of physical things (or events etc.), but I don’t know of any that are still standing. But there is still the question, so that is still an assumption about time (and space etc.).
    It can seem that if there is a God then there could be actual infinities of bounces (and moments etc.), but really it is just that, if there is a God, then God will know the true nature of infinity, whatever that is. And there are conceptual problems with counting to infinity (if no decisive ones known at present).
    Although there is no last bounce, Bob would have counted all the way up to infinity (he would not have left out any finite number). But each number was infinitely short of infinity, none of them were anywhere near infinity. So it is a pretty big assumption that you are making (for all that standard maths makes that assumption).

    November 16, 2007 — 4:53
  • Martin – certainly, Jane’s infinity is potential whereas Bob’s is actual, and actual infinities are problematic. (I’ve been carelessly calling Jane ‘Mary’). But the problems I’ve been talking about with Bob are not merely based on his infinity being actual, but on it’s being a natural infinity that terminates in death.
    The problems I’ve been pointing out would not arise with an infinitely long piece of string that simply has no end, nor would I object if someone said that, from God’s perspective, Jane achieves an actual infinity of bounces: that might be a problem, but it isn’t the problem I’m talking about.
    The problem is that in the scenario described, Bob dies and Mary doesn’t, and this is a significant difference.
    Death involves a last breath, it involves last words and so on, and all the problems raised by a last bounce could be raised by either of these: if Bob didn’t have a last breath or last words, did he really experience death?
    Now experiencing death could be seen as a good thing or a bad thing: we might argue that Jane loses out because she will never encounter God, because she is denied life’s last great adventure. Or, if we fear the experience of death, we could argue that Bob loses out because he has an unpleasant experience that Jane avoids.
    Is what Bob has as good as eternal life, and does that depend on the nature of time? Well, I think it does depend on the nature of time, because I need to know whether Bob has a last moment, a final breath and so on. Does it make sense for him, during his infinite experience, to pray ‘Jesus, Mary and Joseph assist me in my last agony’ or will there be no last agony for him? I want the model of time to answer these questions. But when I consider what is required of a model of time that fits his infinite life into the space of a year, I find myself uncertain that such a model of time could also incorporate the experience of death – yet both infinite time within a year and eventual death were part of the hypothesis. I get the fact that after a year, there is no more Bob, but did he die?
    My worries are very much in the spirit of Dummett’s recent work.
    Abstract mathematical models alone cannot tell us how things are, unless we explain how to apply the model, which involves describing the possible experiences – i.e. how it would be to observers (central claim of Thought and Reality). One mathematical model we find intuitively attractive is to suppose that time is like the line of real numbers in classical arithmetic – a dense continuum. But there are things that, a priori, we know are impossible, but this model of time fails to reveal their impossibility ‘Is Time a Continuum of Instants’, Philosophy 2000. This is why I’m anxious to have a description of Bob’s death.

    November 16, 2007 — 7:49
  • John Alexander

    Alexander
    What would your intuitions be if the angle simply gave them each 70 years worth of memories without the corresponsing experiences?

    November 16, 2007 — 8:12
  • Ben, I don’t think that can be right; e.g. I only said that “the eternal life being lived endlessly requires only a potential infinity,” not that you were assuming that it was a potential infinity.
    And if death does involve a last breath, then that just means that instead of Jane and Bob dying at the end of the year, as Alexander said they did, they would just be no longer living, whereas before that time they were alive (and subjectively living endlessly). Alternatively we might choose to count that as dying, and then death might not involve a last breath; but either way Alexander’s questions seem to arise (for all that we would express them differently).
    Jane and Bob would seem to avoid the experience of dying, since their end would always, for them (whilst they were living), be infinitely remote; but not necessarily the experience of having just died, which is the one that depends greatly upon how good one has been (and which is therefore the important one, arguably).

    November 16, 2007 — 8:23
  • Alex writes, “If what is bad about death is the alleged future non-existence (which is somehow generally taken to be worse than the fact of past non-existence (which I happen to find pretty terrifying)), then there is a big difference between the two options.”
    That’s interesting–why do you you find it terrifying to contemplate the infinite stretch of your past non-existence?
    (I ask out of genuine curiosity, not as a way of introducing some objection. I’ve taught Lucretius’ symmetry argument a lot, and as far as I can tell, almost all of my students buy Lucretius’ assumption that there is nothing terrifying about my past non-existence. They either reject the alleged symmetry between past and future infinite non-existence, or they accept the conclusion that my future eternal non-existence isn’t as such bad.)

    November 16, 2007 — 11:15
  • Ben:
    “I see him perform the last action he performed before dying, bouncing the ball. But which bounce was that?”
    Why are you assuming that there is a last action that he performed before dying? After every action he performed, there was another. When you play it backwards, before every action, there was another. There was no last bounce.
    If the atomic view of time is a necessarily truth, and if the number of atoms in any given interval is necessarily finite, then my second thought experiment makes no sense. But I see no reason to suppose that the atomic view of time is a necessary truth.
    Bob didn’t have an experience of death, but he ceased to exist, and hence died (cessation of existence is sufficient for death but not necessary, since one can die either through ceasing to exist or through ceasing to be embodied).
    The Dummett piece is interesting–thanks for the reference. I think the main argument presupposes an intuition about physical necessity in a counternomic situation. The argument is that if you’ve got something swinging faster and faster, there has to be a fact of the matter as to where the item has to be at the end of the period of faster and faster oscillation. But this is really weird. The scenario is counternomic–eventually the item will be swinging faster than the speed of light. So our intuitions about the determinism of processes of swinging are beside the point.
    The only positive argument here is based on an intuition that is alien to me…
    I kind of like the idea of a fuzzy model of time, but not Dummett’s, because Dummett’s fuzzy model has intervals with non-fuzzy endpoints (at least on my quick skim).
    I should say that I think it is logically possible for time to be discrete, and Dummett’s fuzzy realist model is the right model of time in some possible world. But I do not have that much reason to think time actually is discrete. (The only argument that impresses me at all is the argument about the difficulties of causation through infinitely many intermediate causes.)

    November 16, 2007 — 11:28
  • Tim:
    I don’t know why I find my past non-existence (actually it’s only a finite stretch of time–less than 20 billion years) terrifying. But I have found the idea of the world existing with me not existing terrifying from the first time I thought about this. The first time I thought about it was when I was a little kid and my parents got me a book about in utero development, and it dawned on me that before this process started, I didn’t exist.
    Since it’s probably not a justified fear, I can only give an explanation, not a justification. It may be terrifying only because it is evidence of my contingency, which contingency in turn provides (defeasible) evidence for the claim that it is logically possible for me to cease to exist.
    It is not terrifying in the same way that future non-existence would be. However, it’s hard to compare and contrast, because the past non-existence is real while the future non-existence is merely counterfactual, which makes it hard to compare them.

    November 16, 2007 — 11:34
  • 70 years of false memories would be not much good at all. They surely rightly want 70 years’ worth of living, i.e., of acting, thinking, praying, talking, etc.
    70 years of false memories would be a form of serious deceit. But in the 70 years sped up to one year, there wouldn’t be much deceit. Note that we do not assume our subjective time to be reflective of objective reality. We talk of how the minutes dragged on as hours, say.
    Here’s a different thought experiment. The angel offers Bob and Jane 70 years of life. However, each time they go to sleep, they will sleep, mostly dreamlessly (except for maybe a couple of hours) for about 364 days and nights. So they will only have 70 days’ worth of activity. This seems to me to be a worse scenario than either the scenario on which they get 70 years of activity compressed into one, or the scenario on which they just get one year of normal life.

    November 16, 2007 — 11:42
  • re (3), arguably (see below) the angel’s modified offer could be better than eternal life, because at the end of those 70 years their souls could still exist, whereas if they actually lived forever then they would only be able to have those experiences, with no more time left for anything else.
    re (4), that answer depends upon time being such that it contains God, because if it does not contain God then God could presumably put their souls into another infinite stretch of time.

    November 16, 2007 — 17:14
  • Dear Alexander: I realize that, according to the scenario, there is no last bounce, but I’m trying to argue there must be one, and therefore the scenario is absurd.
    How about this: replace video-tape with cine film. The film is infinite. Perhaps it is a piece of film of finite length, but the second frame is half the size of the first, etc. I set it to start playing 1 second before midnight on Bob’s last night, and to finish slightly after midnight, by which time Bob is dead. I set it to run at Bob-speed.What does it show?
    It ends with Bob’s dead body, (or, if the body disappeared, the space where he lay). I presume this is preceded by his dead body, and his dead body infinitely many times – if I start with him dead and work backwards, I can go on forever, but not reach a frame where he is bouncing the ball rather than lying down dead. If I did so, I would have found the final bounce – but there was no final bounce.
    If I start with the first frame, Bob is bouncing, then bouncing, then bouncing. Working forwards, I can go on forever and never get to the last bounce – each bounce is followed by another.
    But what if I jump straight to the middle – approaching the strip of film side-ways. Or I set one person to start from each end, but give them the power to run as fast as Bob was able to bounce: capable of running through infinite frames in a finite time-period. by a dead frame, or vice versa. If they meet in the middle, then at least one of them has passed a bouncing frame followed – but of course, there couldn’t be such a frame if there was no last bounce.
    However, maybe that isn’t a decisive objection to the whole scenario. Martin, I like your suggestion that there is no moment of death, so to speak.
    In that case, let me return to Alexander’s original questions.
    If I have no car, and no car seats, I am not deprived of car seats. But if I have a car without car seats, I am car seat deprived.
    Jane, with her infinite life, does not die. Never will the world be Jane-less. She has no moment of death, but is not thereby deprived.
    There is a point where the world is Bob-less, so we can say ‘Bob is now dead’, but he has no moment of death, no chance to pronounce his last words, to focus his dying thoughts on God. Of course, many people do not experience the moment of their death (they are unconscious) or do not experience it as the moment of their death (they are taken by surprise), but Bob lacks the experience of the moment of death because there is, for him, no such moment. I think that Bob is moment-of-death deprived: he has a momentless death, and that is a reason (perhaps a defeasible reason) for him to reject the offer.
    Remember the original scenario. He has one year until death. He plans to bounce the ball as often as he can, while still leaving ten minutes to focus on God and make his peace. (He knows that if he starts the ‘focus on God’ activity too soon, his attention will drift before the end). He calculates that after the nth bounce, he had better begin focusing on God, and this final activity will take exactly ten minutes.
    If he chooses pseudo-infinity, he loses this option. At whatever point he focuses on God, he will spend more than ten minutes (according to his experience) focusing on God – perhaps his mind will wander. Also, whenever he chooses to start to focus on God, he is sacrificing the chance to bounce the ball an infinite number of times. His ambition under the one year scenario – bounce the ball as many times as possible, while still spending the last ten minutes making my peace with God – is one he cannot fulfill with the second option. Were there a theory of time where he could have his cake and eat, that would affect the decision, but there doesn’t seem to be such a theory (where he has time to confess his sins after having bounced the ball infinitely many times). Thanks for the discussion: it provokes thoughts about the value of death as well as the nature of time and infinity – having a last moment is valuable, because it allows some people the chance to give someone or something a special place in their life story.

    November 16, 2007 — 20:14
  • Ben:
    Do the frames record instants? Which instants? Does the film speed up towards the time of Bob’s death and slack off afterwards? The frame-by-frame business doesn’t seem to me to make sense.
    How does Bob lose in the pseudo-infinity scenario vis-a-vis the infinity scenario? In neither scenario does he get to experience his death. In both he gets equal amounts of stuff done.
    And here’s a further thought to complicate matters. Take a scenario where Bob lives forever in the ordinary, non-accelerated way. But now embed the time-sequence in that world in a longer bigger, non-Archimedean time-sequence, but have Bob live only as long as the shorter infinite sequence endures. Now there is a future time (or hyper-time) at which Bob no longer exists. Of course, it’ll take infinitely long to get there. (But likewise there is a real sense according to which from Bob’s point of view, it’ll take infinitely long to get to death on the eternity-in-a-year scenario.) I don’t think Bob loses just because additional times get added on to the end of the infinite time sequence. If this is right, then the mere fact that there will be a future time at which one doesn’t any longer exist isn’t that bad.
    Here’s another argument to the same effect. Suppose that the universe has an infinite past. The angel makes Bob the following offer. Instead of Bob living for an infinite future time, Bob will first live for a year at a normal pace. Then, God will transport Bob back in time to the year 1000 BC, and put him on a planet where things are just like they were on his original planet after that year, so the transition will be seamless. Bob will live for a year, until 999 BC. Then God will transport Bob back in time to the year 2000 BC, and put him on a planet where things are just as he last saw them in 999 BC. He lives until 1999 BC there, and God transports him back another thousand years. And so on.
    Bob ends up living forever in personal time–I don’t think he loses anything by being shuffled around in time like that. Nonetheless, there is a time after which he doesn’t exist, namely a year from now.
    If you think an infinite past is impossible, but will allow me the possibility of a universe with infinitely many habitable planet, Bob can seamlessly live forever within a single year of external time, by being constantly transported back one year and put on yet another planet.

    November 16, 2007 — 21:02
  • John Alexander

    Alexander, et al.
    I have some of questions/comments:
    1) Is the word “good’ in #3 vague? Does it mean quantitatively the same or normatively the same?
    2) What does this thought experiment have to do with our understanding of eternal life, assuming that we have that option given certain normative conditions? Obviously, given the way the thought experiment is set up we could be either Bob or Jane, and if we assume the angel does not tell us what he is doing but is doing what it is that he is doing to spare us from emotional suffering, then we will never be aware of any of changes that take place within our experiential framework with its corresponding memories. Can deceit ever be morally justifiable or even obligatory?
    3) If you were Bob or Jane and the angel made you aware of the options why not ask simply for the 70 years without the tricks if the memories/experiences are going to be the same anyway? My reason for asking about simply getting the memories is that everything of importance will be intact regardless of the options re the memories Bob and Jane would have (recall Captain Picard’s ‘experiencing’ an lifetime on a planet only to find out that the memories were instilled in him by a computer when he awoke from a 30 minute ‘nap’).
    4) How are these thought experiments any different from Russell’s five-minute argument re the epistemic value of the memories? I have memories of ‘experiences I had yesterday even if I never had those experiences and these memories will track with memories that others have (my students in class with me) even though they never had the experiences. It seems the crucial epistemic factor is reducible to the memories I have and the normative value I place on them re how I live my life. I assume the eternal life, if that option exists, is available to me only if I live a certain life. Oops, I forgot about Hell so I guess eternal life is available for all, one track will just be better then the other.
    5) Do these thought experiments require us to reject a dualistic framework for understanding what we are as ontological beings? If we die in any of these situations does that mean that our souls die also, or if it is our souls that are eternal in some Platonic sense) are our physical bodies simply temporary residences of our souls.

    November 17, 2007 — 9:02
  • Alexander, in your original post you said:
    ‘And then at the end of the year they will die. But it will have felt to them like they were living forever.’
    So it is part of your original hypothesis that they die. Before the angel came, they knew that sudden painless death was going to strike. Originally, my understanding was that after the angel came, if they opt for the ‘infinity in a year’, I understood that the sudden and painless death would still strike, that, after one year, the suddenly dropping dead will take place. Well, that seems to be incorrect: unless I’m mistaken, the consensus is that with ‘infinity in a year’ there can be no ‘moment of death.’
    Martin suggested that, if we don’t call what happened to Bob dying, then ‘Alexander’s questions seem to arise (for all that we would express them differently).’ I can’t agree with that: the sentence quoted above suggests a very clear contrast between what will really happen, and what it will feel like to them. One thing that needs to be considered then is the importance of a gap between our experience and reality. If Alexander said ‘It will have felt to them like they were living forever, even though from our perspective there would come a time when they don’t exist’, that would be proposing something different for consideration.
    I’m dubious about the concept of ‘moment-less death’, but I’m willing to try to make sense of it.
    One thing I’d really like to know whether there is a corpse – if I can point to Bob’s dead body, I feel comfortable saying that he’s not only merely dead he’s really quite sincerely dead. However, if there is a corpse, that also suggests (though it needn’t prove) that Bob had a moment of death: if coroner can aver he’s thoroughly examined Bob, he might have some conclusions about the time and cause of death that involve a moment of death. Perhaps Bob’s dead body can be reconciled with a moment-less death, but I don’t know how.
    Here is where I tried another tactic to justify ‘Bob has died’: Bob was certainly once alive, but the universe now no longer contains Bob. After all, a body is proof of death but absence of a body is not a proof of life: it is possible to die and leave no corpse. You have convinced me that this was too big a concession: the person who keeps travelling back to the past never dies, yet they were alive once at a later point, they are absent from the universe. So a period of life followed by absence forever isn’t necessarily death. It doesn’t follow that Bob’s disappearance does not constitute death, but it does cast doubt on his death.
    This is why I want to know what would be observed by someone who watched Bob as the end of the year arrived, but did so in sufficient detail to see everything that Bob did. This could be a person, or a camera – let’s say the camera shoots sixteen frames per second, but that it stays in synch with Bob – sixteen frames to a Bob-second. It starts shooting at one second to midnight our time, and continues up until midnight our time, by which time our world is definitely Bob-less. Whatever was the maximum speed the camera attained while filming Bob, it retains that speed for whatever portion of the final second remains. Incidentally, I assume Bob has enough food to last him for infinity, that he leaves behind infinite toe-nail clippings etc, so infinite film shouldn’t be a problem. What might be a problem, is whether the camera, or the living eye-witness or the corpse, or the toe-nail clippings, are accessible to us afterwards. I suspect it’s an all-or-nothing deal. If we have Bob’s corpse, he’s really dead, but we can start to figure out the cause of death, implying his death is an event recorded on the camera. No corpse implies no camera, or at least no film, or at least the part of the film that shows what happened to Bob when, from our perspective, he disappeared.
    So does Bob leave a corpse or disappear? If he disappears, does his physical environment go with him – never to return? If the latter is true, then instead of saying Bob died, couldn’t we just as well say he entered another time stream – his reality no longer inter-acts with ours, so he is absent from our reality, but that doesn’t mean that he is under an illusion when he thinks of himself as immortal.

    November 17, 2007 — 10:07
  • Ben:
    You can have a corpse if you like.
    On the infinity-in-a-year scenario, I’m assuming that they’re alive for an interval open at the upper end, like [t1,t2). Now in English “moment of death” is a vague term. We need to make it precise. It could mean the last moment of life or the first moment of non-life. If it means the last moment of life, there is no such thing in this case. So, there is a moment of death in the sense of the first moment of non-life, namely t2. But that is a moment that the people don’t experience, because they’re dead at that moment.
    So there is a death, but one not experienced. (But that’s fine. The only way to experience death is to survive it, and in this scenario we’re talking counterfactually of cessation of existence rather than the kind of death we’ll face, which is separation of soul and body. And it is dubious that the cessation-of-existence kind of death can be experienced, since when you exist, it hasn’t happened yet.)
    I assume the video tape speeds up as one gets closer to the end of the year. It had better, because Bob’s activity speeds up, so pretty soon his activity will far outstrip the frame rate. But if it speeds up as one gets closer to the end of the year, then it will have used up an infinite length once it reaches the end of the year, and presumably it won’t be able to record any more. And we can’t play an infinite tape backwards, except by speeding it up so fast we can’t observe it. 🙂

    November 17, 2007 — 10:59
  • Dear Alexander: the hypothetical film may be infinite, but this thread probably won’t be. I sense that if I write much more on this, I’ll just be going around in circles (perhaps I already am), so I’ll try to make this my last comment.
    You are right that the film speeds up as midnight approaches: it always keeps pace with Bob, who is getting ever faster. We are left with an infinite strip of film, but I assume that we can watch segments of this film, frame by frame. If we can achieve the kind of speed that Bob eventually did, I don’t see why we couldn’t watch the whole film – that was why I had the example of two people running towards each other until they reach the middle. If they do, they should find Bob’s last bounce, or last breath, or whatever – but no such event should exist.
    Leaving that aside, let me repeat why I think Bob (with the psuedo-infinity) is deprived but Jane (with a real infinity) is not. It is because to be dead without having had a last moment of life is to be deprived, whereas to have had no last moment of life and not to be dead isn’t to be deprived. This is just a moral intuition that others might not share, but I can certainly point to similar examples: in England, it is customary for the Queen to send official congratulations to anyone who reaches their 100th birthday. If I lived in England, reached by 100th birthday and received no congratulations from the Queen, I would be deprived. If I fail to reach that 100th birthday, my lack of congratulations on reaching it is not a deprivation.
    I’ve variously said that what Bob is deprived of is the ‘experience of death’ a ‘moment of death’ etc. I have in mind something that I think is part of the normal human process of mortality, something that Bob lacks. I’ll focus on last words.
    If I choose my last words, I get to assign something a position of special honour in my life – e.g. Bob might choose to say ‘Jane’ as his last word. Of course, I might not know these words will be my last ones, or I might not choose carefully – perhaps I die saying something banal. I have the opportunity, but I waste it.
    Bob has no such opportunity. He cannot choose his last words well, or foolishly, since he never chooses a last word. To be precise, what he cannot do – and I think this is the best way to describe it – is choose, of his last possible word, that that word be ‘Jane’. He lacks this choice because no word of his is his last possible word (even though every word is possibly his last). That is how I think Bob is deprived. Lacking a last possible word/thought/deed etc. is not a lack if you are never dead. Jane and the time-traveller are never dead, but Bob is because we do have his corpse.
    Well, I think I’ve pushed this as far as I can, but thanks for the discussion.

    November 17, 2007 — 21:56
  • Ben:
    Thanks for all your patience. These discussions of infinity and supertasks seem extraordinarily fruitless. One either has the intuitions that Aristotelians (about time) like you, Dummett, Bill Craig or other defenders of the Kalaam argument have, or else one has the intuitions that various “moderns” like me, Adolf Gruenbaum, etc. who blithely want to model time on the continuum of reals have. What is particularly interesting about this case is that the intuitions on each side seem to have just about no traction with the other side. This is different from a case like free will where the typical compatibilist feels that there is something to the principle of alternate possibilities, and the typical incompatibilist feels that there is something to the idea that responsibility requires action to express one’s character, but the two sides assign different weights to the intuitions, or a case like the debate between utilitarians and deontologists where many will admit that they can feel something of the force of the arguments on the other side.
    Here, the “Aristotelian” intuitions carry just about zero weight with the “moderns” (I don’t even think there is anything strange with things like Hilbert’s Hotel or supertasks; the grim reaper paradox gives me a bit of pause, but just a bit), and the “moderns'” glibness about the possibility of modeling time on the reals carries no weight with the “Aristotelians”. It would be interesting to see how the discussion could be moved forward.
    I don’t know what to make of this meta-observation.

    November 17, 2007 — 22:17
  • By the way, I’ve made a related post on my personal blog.

    November 17, 2007 — 22:43
  • Steven Carr

    ‘During the next three months, their and their environment’s speed of functioning will be sped up another factor of two, so it’ll feel like they are living the equivalent of another year during those three months. During the next 1.5 months, we get another speedup, so it will feel like those 1.5 months were a year of life. And so on. And then at the end of the year they will die. But it will have felt to them like they were living forever.’
    Each halving is like one year of life?
    That would enable them to solve a puzzle that always puzzled me.
    At the end of the period , we simply ask them how old they are.
    They tell us how many years they think they have lived, and we will know if the highest number is odd or even.

    November 18, 2007 — 4:47
  • Raymond W. Aldred

    Wow, I really like this puzzle. Perhaps we might be able to adjust this case to create a similar puzzle for subject sensitive invariantists, in epistemology. I’m actually taking a class on subject sensitive invariantism so any criticisms or replies would be very valuable to me. I’ll try to spell out the adjusted puzzle for you:
    The following four claims seem to be plausible and all have theoretical advantages, the first two have theoretical advantages in moral theories, the latter two have theoretical advantages, apparently, in epistemology:
    (MRA) We ought to want most of our actions to be rational or morally justifiable in some way.
    (MBA) If act A is the rational choice for you to make and you can do A, then you ought to do A.
    (KA) If you know p, then you are rational to act as if p.
    (KB) If you know act A is best (ie. Has the best consequences of the available acts), then you are rational to do A.
    (MRA) has the theoretical advantage of explaining why hypocrisy is wrong in some cases. If a chauvinistic boss hires a female employee to look good in the eyes of other females. In particular he wants to look good in the eyes of the employee he hired, so he can later take advantage of her.(MRA) would rule that this action is morally wrong: Although his action has some good consequences, his desire was not good.
    (MBA) seems to have the theoretical advantage in explaining why inaction is wrong in, some cases. Suppose I walk by a drowning person and I know that the rational or best thing to do is save him, and I can save him, if (MBA) is true then I ought to save him.
    (KA) and (KB) are both held by Subject Sensitive Invariantists like Stanley, McGrath, and Fantl. The theoretical advantage of (KA) and (KB) is that it seems to explain why some pragmatic concerns can act as defeators for knowledge. In short if it is not rational for me to act as if p, then I don’t know that p.
    Now suppose a person, call him Sam, is approached by an morally neutral brain scientist. Sam lives on the street. The scientist offers Sam the option to be wired into a brain in a vat, if Sam chooses to be wired into a brain in a vat then, the scientist promises his life will be happier than it is now. In addition to being happier, the scientist promises to erase all of Sam’s memories about his proposal, so he can go on believing and “feeling” like he’s living the high life and thus be happier. The cost, is that any of Sam‘s attitudes towards mundane propositions are not true, thus he loses most of his knowledge. Now suppose Sam is a subject sensitive invariantist, so he holds (KA) and (KB): at the same time Sam generally wants to do the right thing, so he holds (MRA) and (MBA).
    Should Sam choose to be a brain in a vat?
    If yes then it seems to lead to inconsistent results, since we ought to want most of our actions to be rational actions. But if we want most of our actions to be rational, then we ought to want to know our mundane propositions about the world. Thus Sam ought not to choose to be a brain in a vat since there will be less actions that are rational, and Sam should choose to be a brain in a vat since it has the best consequences.
    If no then it seems to be counter intuitive to (KB), since Sam might know that the best action for him is to be a brain in a vat, since it has the best consequences. But it is not rational for him to choose to be a brain in a vat since we ought to want most of our actions to be rational ones.
    So it seems in this case, we should abandon one of the principles, but which one should we abandon?

    November 18, 2007 — 16:06
  • Mr. Aldred:
    I think that whether one sees your puzzle as like mine depends on whether the “speedup” in my scenarios involves deception. I am not sure it does. On my preferred view of time, it does not essentially involve any deception, actually.
    Your puzzle is interesting, though. But I don’t see why a negative answer to your question conflicts with KB. First of all, KB does not say that the rational action is the one that is best for oneself. Second, the idea that one is better off in the vat depends on a dubious hedonism.
    Besides, KB is false, since it entails something very much like utilitarianism. It is rational to do what is best, but “what is best” cannot be identified with what produces the best consequences.

    November 18, 2007 — 23:36
  • John Alexander

    I have a question although it is probably beyond the time to bring it up, but if they die do they not remember living? Asuming that human beings have both a soul and a body if the body dies and the soul continues then there would be a continuity of memeories and ‘life’ would be going on after the body dies. If the soul dies with the body then there would be no memeories and therefore no point of view. You would have some sort of eternal life in the instance of the soul continuing, but not if the soul does not continue. The issue seems to be centered on the existence of the soul and if the soul does exist and continues after bodily death what is the issue?

    November 21, 2007 — 8:16
  • Let me throw another little wrench into the discussion. If the amoralist atheist wants to argue solely on the theist’s home-turf, using the theist’s own confidence that there is objective morality, shouldn’t the atheist allow the theist to make use of other beliefs of her own, such as those about the afterlife or about Christ’s redemption and the “felix culpa”, in responding?

    November 21, 2007 — 8:57
  • John Alexander

    Alexander
    I completely agree that anyone has the right to make use of their system of beliefs to support a position within that system. As an amoralist and an atheist (I do not beleive IN a theistic God) I hang around this blog because I think I have an obligation to enter into the discussion within the context of that discussion and not simply argue apart from it without input from those who disagree with me. Different perspective helps me to better understand the importance and complexity of the question. I think dialogue is extremely important in understanding the questions that we all face in our struggle to make sense of it all.
    I have spent many decades reflecting on the nature of God and the existence of suffering. I went to Mass a few weeks ago with my wife to meet her new priest. He told the story of why he became a priest after having had a successful life in business, etc. Without going into detail his story, which included working with the poor in the third world, moved me to realize that the problem of evil is not really a problem about why God would allow evil to exist, but why I continue to allow evil to exist in those areas that I can have a positive impact on. It hardly seems fair to be critical of God when I do so little to alleviate suffering around me. Since that day I have found the philosophical problem of evil to be much more personal and I think more deeply felt then before this ‘religious’ experience. Now, I do not think I will ever come to believe in a personal and loving God,(I am like Dr. Rieux in Camus’ outstanding treatment of how we face suffering, The Plague) but that should not stop me from facing suffering head on and trying to eliminate wherever I can. Who knows, maybe God is silent in his heaven because I am so silent (do so little) here in a world that is defined by suffering.

    November 21, 2007 — 10:09
  • Raymond Aldred

    Alex,
    Thanks for the reply, I kind of have the same intuitions about KB as you, it seems to be a strange principle. I can also understand why you might have your doubts about whether answering in the negative contradicts with KB. Answering in the negative, if interpreted as a utilitarian principle may be counter intuitive to KB. Why? Sam will be happier if he goes into a Brain in a Vat. I tried to construct the case where the consequent is the best if Sam chooses to be a Brain in a Vat.
    But KB seems true, if interpreted as a deontological principle. But if interpreted as a deontological principle, then Sam should not go into the brain in a vat. But then consequentialists might have issues with Sam’s decision because sam is ignoring the best consequences.
    One way the consequentialist could get around this problem though is to say that the best consequences would be for Sam not to go into the brain in a vat, because it will have the consequent of contradicting with MRA. Indeed this is my intuition about the case. But others might not have the same thoughts on the subject.
    Subject sensitive invariantists would say that KB is true whether it’s a deontological principle, or a consequentialist principle. But perhaps not both. Knowledge that A is the best action for S, gives one good reason for S to do A. “A is the best action” can have different interpretations, but it seems true with either interpretation you give.
    The problem is I have conflicting intuitions about this case, whether KB is true as a deontological principle, or KB is true as a consequentialist principle.
    The interesting thing about KB, I think, is that knowledge changes if the stakes are higher for you. If you can’t act as if p, and A is the best action based on p, then you don’t know that p. It’s an odd principle, and there are all of these cases where it seems counter intuitive to our fallibalist intuitions. But the principle itself seems so plausible that I find it hard to deny.
    It actually might be interesting to see whether this principle can be extended to Philosophy of Religion, like the problem of evil. But i’m just not sure how it can be.

    November 21, 2007 — 14:31