Welfare and open theism
April 22, 2007 — 7:37

Author: Alexander Pruss  Category: Open Theism  Comments: 21

The following propositions are incompatible:

  1. Metaphysically necessarily, there are no facts of the matter about future free actions.
  2. Metaphysically necessarily, one is better off insofar as one's morally good plans are fulfilled and worse off insofar as one's morally good plans are not fulfilled.
  3. Metaphysically possibly, there was a person x who permanently ceased to exist and who made a morally good plan whose fulfillment required a free action after x's permanent death.
  4. Metaphysically necessarily, one is neither well nor badly off when one does not exist.

I conclude that (1) is false, but others might draw other conclusions.

There is a difficulty about the permanent cessation of existence. As Jon Kvanvig has pointed out, if there is an open future, there are some difficulties in making sense of any logically contingent facts about the future, even ones that are predictable by means of the laws of nature. However, if an at all orthodox open theism (not that I think an open theism can be very orthodox) is true, then God has to be able to guarantee claims about the respective eternal destinations of the just and the unjust. If God can do that in some way (say, by making binding promises), he should be able to guarantee that x permanently ceases to exist. Alternately, if one thinks that temporally gappy existence of persons is logically impossible, one can replace "permanently ceased to" with "ceased to exist". There will also be theological difficulties for those who think that annihilation, which is after all worse than eternal damnation, is incompatible with divine goodness. Open theists convinced of these difficulties will be able to shrug off my argument.


  • Alex,
    How does rejecting (1) help? You seem to think that if there are facts about the future then the fact that S’s plans will be fulfilled in the future actually increases S’s well-being now. I know of no view on well-being on which that is true. Can you imagine me having a legitimate moral complaint against you now because it is true that you will, in 2011, steal my car? It is certainly not the truth of some future contingent proposition P that improves or diminishes my welfare. It is the obtaining of the future state of affairs described in P that does so.

    April 22, 2007 — 9:51
  • Philip

    I like this one, Mike seems to be rejecting (2), but it sure seems like the results of my life which occur after I die increase the quality of my life. Would Martin Luther’s life have been as good if the Reformation got snuffed out shortly after his death?
    I wonder if an open theist who is a growing block theorist could say that a person’s life does become better when it becomes true that a certain good result of his life occurs. Since on this view his life would still exist.

    April 22, 2007 — 18:28
  • It should be obvious that I did not deny (or affirm) any of Alex’s premises. I’m not sure how it could be gathered that I did. I simply asked a question about how denying (1) would help; that is hardly an endorsement of (1) or a rejection of any other. I am not averse to the idea that some informed preference-theoretic account of value might be true according to which the satisfaction of S’s preferences post mortem increases S’s overall welfare. That is not an indefensible position. On the other hand, welfare theories on which there is no diminishment in S’s overall welfare without a perceptible change in S mental states are also defensible. I’ve taken no position on either of these here.

    April 22, 2007 — 20:46
  • Philip

    Mike, very sorry, I now realize you weren’t taking a position.

    April 22, 2007 — 21:24
  • Mike:
    Claim (2) was not meant to assert that the welfare due to the future fulfillment of a present plan is to be attributed now to the agent. I meant claim (2) to be compatible with the idea, for instance, that welfare cannot be attributed to an agent at one time but to a life as a whole. All that (2) claims is that a life where one’s good plans succeed is a better life than one where one’s good plans fail, even if the success or failure is in the future of the life. Whether the difference is welfare is located when the plans are first made, or over some interval of time, or over the lifetime as a whole is left open by (2).
    However, all that said, I am inclined to think it makes a difference to the welfare at t what results later. Whether a struggle to get out of a swamp is futile or not depends on what comes of the struggle. But it is plausible that Jones is now the worse off for engaging in a futile struggle, i.e., a struggle that will in fact be ineffective.
    Maybe the best move for the adherent of OT who wants to remain an adherent of OT would be to deny (4) and accept growing block as per Philip’s suggestion?
    Another interesting option is to say that what is true at t, for some fixed t, changes with time. It is, let us say, not true today that there will be a sea battle tomorrow. Likewise, it is not true today that there will not be a sea battle tomorrow. Suppose tomorrow comes and a sea battle occurs. Then tomorrow one will be able to correctly assert: “Yesterday it was true that there would be a sea battle in a day.” This implies that the past really changes, which is weird. But if we accept this weirdness, then we can say that posthumously at t2 it becomes the case that at the time t1 when a plan was made the planner was well off with regard to the plan’s success, even though when it was t1, the planner was neither well nor badly off with regard to the plan’s success, there not yet having been a fact of the matter about it.
    This has the odd consequence that how well off I am on April 23, 2007 not only depends on the future but actually changes with the flow of time.
    This would be a rather strong rendering of Solon’s famous maxim.

    April 23, 2007 — 10:28
  • Luke Gelinas

    Might one not say that, while the overall value of one’s actions through t is capable of being affected by some state of affairs S after t, this doesn’t entail that one’s welfare through t is capable of being increased or diminished by S in the same way?
    I guess this is my intuition. If I die before my plans and strivings to save Venice issue in success, but someone else’s subsequent actions connect with mine in the right way to save the city, my intuition is that the value of my endeavor is increased. But I don’t see why my welfare should be increased as well. And I just don’t share the intuition that my welfare now is tied to future states of affairs. So deny (2) as stated.

    April 24, 2007 — 7:57
  • Luke:
    That is certainly a way out.
    But suppose that someone thwarts a good plan dear to my heart. Initially, suppose all this is happening while I am alive. Am I not the worse off?
    If I find out about the thwarting, clearly I am the worse off for being thwarted. I would feel as if a bad thing happened to me, and I would be right. But I think it is not just my finding out about that makes me worse off. This is testified to by the phenomenology of the feeling that a bad thing has happened to me. It is not the finding out that I feel is the main bad thing–as if it would have been much better had I not found out–but the thwarting.
    So I will be worse off for my good present plans having been thwarted in my lifetime, even if I am unaware of this. But why the “in my lifetime” restriction? There are plans of mine where it matters comparatible little to me whether they are achieved in my lifetime or afterwards. In such cases, once we realize that things we are unaware of can affect our welfare, we should likewise hold that our welfare is affected by things after our death, even if (contrary to fact) there is no life after death.
    A different tack. While morally speaking being an attempted murderer and being a successful murderer are on par, one is better off being an attempted murderer than being a successful murderer, since an attempted murderer is responsible for less evil, and being responsible for evils is a part of our illfare. But whether George is an attempted murderer or a successful murderer can depend on people’s free actions after George’s death (e.g., George dies shortly after stabbing Fred; Fred is still alive then, but may die of his wounds, depending on how conscientious the physicians are; whether Fred dies of his wounds determines whether George was a successful murderer).

    April 24, 2007 — 23:08
  • Luke Gelinas

    If we restrict (2) so that it doesn’t allow future states of affairs to increase well-being in the present, I find it more plausible. It seems to me that some action A after t might increase the value of my life taken as a whole; it might even retroactively increase the value of my life at t; but I find it odd to say that, at t, the future occurrence of A increases my welfare.
    I suspect the intuition here is the same one that inclines people toward openism. A can’t increasing the value of my life at t simply because A hasn’t happened yet; there’s nothing there to connect with my present state to do the increasing. So the same considerations driving (1) might also drive the rejection of this interpretation of(2).
    Couldn’t the openist also reject (4) on purely theological grounds? Suppose you hold that at death the person doesn’t cease to exist, but enters some unconscious intermediate state until the eschaton, at which point she regains consciousness and spends the rest of eternity either in the presence of God or not. Or suppose you think one goes straight from earthly existence to one’s ‘eternal home’. Either way you can say that one doesn’t cease to exist, and that one’s welfare is increased whenever the relevant actions take place (not before).

    April 25, 2007 — 13:55
  • Luke Gelinas

    Sorry, that wouldn’t be to deny (4). But it would let the openist affirm (1)-(4) on a modified version of (2) like the one I previously suggested.

    April 25, 2007 — 17:39
  • Luke:
    1. Suppose George has a good plan, and it is fulfilled one second before his permanent end, without his being aware of it. Then it seems plausible that his life was the better for it. But what does it matter if the plan is fulfilled a second before his permanent end or a second later, given that in neither case is he aware of it.
    2. The theological grounds for thinking there is an afterlife will only be relevant, I think, if they are grounds for thinking that it is metaphysically necessary that there be an afterlife. I think that we will live forever, and probably all persons will live forever. But I am not sure it is metaphysically necessary that all persons will live forever.

    April 25, 2007 — 17:55
  • Luke Gelinas

    Hi Alex,
    1. If George dies at t, and his good plan is fulfilled by someone else’s free decision at
    t + 1, I’m fine with saying that, at t + 1, the overall value of George’s life taken as a whole, and George’s welfare, if he’s in existence, is increased. I just deny that George’s well-being at t can be influenced by something there is, on openism, no fact of the matter about at t. Again, the openist will, I think, want to know exactly what it is, at t, that’s supposed to be connecting with the rest of George’s life to increase it’s value.
    2. Right. I haven’t thought about this much, but what do you think about the following argument for the conclusion that, if you believe that human persons don’t go out of existence, you should think it metaphysically necessary that human persons don’t go out of existence. It takes the form of a trilemma:
    (P1) Necessarily, either human persons don’t go out of existence, or temporally gappy existence is true, or annihiliationism is true.
    (P2) The reasons for thinking temporally gappy existence false are reasons for thinking it metaphysically impossible.
    (P3) The reasons for thinking annihiliationism false are reasons for thinking it metaphysically impossible.
    (P4) So, if you have reason for thinking that human beings don’t go out of existence, you have reason for thinking that temporally gappy existence and annihilationism are metaphysically impossible, since, on (P1), the truth of the first disjunct entails the falsity of the second and third; but if we have reason to think the second and third disjunct false, we have reason, by (P2) and (P3), to think them metaphysically impossible.
    (C) So, if you think human beings don’t go out of existence, you should think it metaphysically necessary that human beings don’t go out of existence.
    If we grant the trilemma in (P1), I think the brunt of the weight falls on (P3). The justification I have in mind is that whether or not God annihilates human beings is, I assume, a moral decision, one that flows from God’s essentially morally perfect nature. Thus, if God’s morally perfect nature requires him to annihilate humans in one world, it requires him to do the same in all worlds. If annihilation is the morally best course of action for God to take, he takes it in all worlds.

    April 26, 2007 — 8:02
  • Luke:
    Thanks for these suggestions.
    ad 1: Your suggestion is in essence that (4) is false, and hence a challenge that I come up with an argument for (4). Here’s one. (i) x’s welfare is a property of x. (ii) Nothing has properties when it does not exist. Ergo, etc.
    This argument will work best given OT+presentism, since it is plausible that given growing block entities can change in respect of extrinsic properties (e.g., being thought about) and hence can have extrinsic properties after they have ceased to exist.
    Still, even given growing block, I think there is a lot of plausibility to the idea that only x’s Cambridge properties can change after x has ceased to exist and that welfare is not a Cambridge property.
    ad 2: I think you’re right that (P3) is the main question. Here I think you’re assuming that God always takes the morally best course of action. This is a problematic assumption for several reasons. First of all, it assumes that there is a best course of action, whereas there might be incommensurables or an infinite regress of better and better. Second, it seems to take away from the gratuitousness of grace. Third, it seems to damage omnipotence as it implies that God can only strongly actualize one state of affairs.
    There is a Jewish tradition that God every day creates a new choir of angels who sing a new song for that day and fade into non-existence.
    Consider the following argument. Take any possible world w. Let w* be a world just like w but with an additional angel who sings a new song for a day and fades into non-existence satisfied to have lived at all. There is some plausibility to thinking w* is better than w, isn’t there? A worry, though, is that for this to be relevant to my argument, that angel would have to have plans that extend beyond that day, and it is not clear whether a perfectly good God would create an angel of finite lifespan that has plans extending beyond his cessation of existence.

    April 26, 2007 — 8:49
  • Alexander, this is an interesting topic. There have been previous discussions.
    The Nicomachean Ethics, Book I, 10 (1100a 10-30) is relevant here. Aristotle considers whether the good or bad fortune of my descendents can have an effect on my post-mortem state of eudaimonia. Aristotle doesn’t say much in this passage beyond posing the problem, but commentaries on this passage might well contain relevant material. Aquinas, for example, states in his commentary that if fuisse is true, it must be because esse was true – i.e. no ‘was’ without a prior ‘is’. But eudaimonia (equivalent to your ‘welfare’) must be a secure state: I cannot truly say ‘He has eudaimonia’ if what he has is something that he might lose. However, even after I die, there are some things that I might lose – e.g. reputation. Aquinas’ conclusion is that losable properties may be ‘good fortune’ (Litzinger’s translation) but cannot be eudaimonia. It might be interesting to check out the history of commentaries on this passage – I’m sure I’ve read some interesting modern commentaries, but don’t recall which they were.
    Another source that I don’t, alas, recall, was an article in an Irish journal of philosophy. I read this article in the mid 90’s, and the journal was already several years old by then. Could I only return to the very library where I read it (Oscott College), I could easily locate it. However, tracking it down with the Philosopher’s Index has not been fruitful. The author argued that by bringing the project of a dead person to a successful conclusion, we are changing the past.
    I’m planning to follow this up with another post where I’ll actually say something about the issue, but first I should grade some papers.

    April 26, 2007 — 11:22
  • I suppose, in an Aristotelian vein, I might argue that bringing about intended states of affairs is one of our perfections. But whether I bring about an intended state of affairs does depend on the future.
    The stability criterion is interesting. But maybe we can say that our plans become less and less relevant as time goes on. We only plan so far into the future. So within, say, fifty years of our death most of our welfare becomes settled.
    (Of course, I’m a B-theorist, so this stuff about whether the fulfillment of the action is in the future doesn’t bother me at all.)

    April 26, 2007 — 11:42
  • Here’s another thought. Suppose a project was dear to the heart of a friend, and now the friend is no longer existent. The project is good, and it will naturally come to fruition, but I can thwart it. Does the fact that the project was dear to the heart of a friend give me additional reason not to thwart the project, over and beyond the reason that I get from the value of the states of affairs the project is aiming at?
    An affirmative answer seems very plausible. A good explanation of why an affirmative action is right would be that the fulfillment of the project is a part of my friend’s welfare.
    On the other hand, one might argue that we have x-regarding reasons that do not derive from x’s welfare. For instance, I may have x-regarding reasons to honor x with an elegant funeral, but perhaps it does not follow that the elegant funeral is a good thing for x.
    I have a strong intuition that we have x-regarding reasons to bury x well, and that x’s welfare is unaffected unless x had a good plan to be buried well. But my intuitions may just be inconsistent here.

    April 26, 2007 — 11:54
  • Well, that was some relatively painless grading – for me and the students. I think they’ll all be happy with their grades: but what about their welfare?
    Alexander, I think OT+moving block is the most plausible form of OT, so I’ll restrict myself to that. As you say, it is plausible to suppose that there will be some properties of me that change after my death, and that these should be classified as Cambridge properties. So then we need to ask, as you say, whether welfare could be a Cambridge property.
    Here is where Aquinas’ work might be useful. Obviously, someone could just try to dismiss the whole issue by saying ‘Well, it depends what you mean by “welfare”‘, and, within the OT+MB framework, we could have Welfare1, which is fixed at time of death, and Welfare2, which varie after death. Who is to say which is the ‘real’ welfare.
    We can probably agree that analysis of ordinary language is not going to settle this: I doubt that the evidence of ordinary language points decisively to one answer and, even if it did, why would we be obliged to let ordinary language have the last word?
    Your own comments about the phenomenology of thwarting could be used to support the case for Cambridge Welfare. (The article in the Irish journal makes a point along these lines). The argument goes like this: when I discover that my plans have turned out well/badly, the discovery makes me feel good/bad, but this feeling is simply an appropriate response to the objective fact that my plan’s success/failure was good/bad for me, a fact that would be true whether I knew about it or not. My extinction prevents me from knowing whether my plans succeed/fail, but the objective goodness/badness to which my feelings would be an appropriate response is still there.
    Of course, the posthumous success of my plan constitutes a Cambridge change in my properties. However, what is so inferior about ‘Cambridge change’? Many of our deepest desires concern things that are outside of us, and to suppose that the only change that really affects the value of my life is a change internal to me is to condemn me to solipsism. The solution, although it goes deeply against the grain, is to learn to talk of Cambridge properties and Cambridge changes without automatically sneering at their merely Cantabrigian status.
    The Thomist argument would be that when we ask what ‘welfare’ is, or what ‘eudaimonia’ is, the true object of our search is that which constitutes the end of human nature, and to constitute the end of human nature would be to satisfy fundamental human cravings. However, that which is ephemeral, which we have but which might disappear, cannot satisfy human cravings – you cannot have true eudaimonia if your continuance in that state is in doubt. (Perhaps then ‘welfare’ isn’t the best term to adopt, because in many states, welfare is intended to be only a temporary support).
    However, this does not exclude Cambridge properties on the grounds of their being Cantabrigian. Internal and external properties might be necessary for eudaimonia, the point is that if their continued presence is not guaranteed, they are not sufficient for eudaimonia. So true welfare could include posthumous success, but there would have to be a guarantee of permanence.
    A naturalist might respond that, if all this is so, eudaimonia is something we can never achieve: nature has given us deep cravings that can never be satisfied. We had a long discussion about this topic a while ago on Prosblogion.
    A Buddhist might respond that the desire for permanence cannot be satisfied, and that this is indeed a deep and natural human craving, but the real goal of human nature is achieved when we let go of the desire for a state of happiness that is bound to continue. The same evidence that supports the Thomist theory that our true goal is a permanent state could also be used to support the Second Noble Truth of Buddhism that the origin of Dukkha (suffering) is Tanha (thirst, particularly for a permanent presence).
    I’d guess that an Open Theist could allow us to be extinct but, if all our best desires are fulfilled, permanently, say that we have achieved true eudaimonia, even if we aren’t around to experience it in person – just as in some tribal societies, the fact that the tribe will continue and prosper means that people do not worry so much about their individual fate after death. This also reminds me of the ‘objective’ view of life after death put forward by Process Theism. I particularly remember this view as explained by Keith Ward: “This is called the objective view of life after death, and it basically means you don’t really live after death at all.”

    April 26, 2007 — 13:22
  • Heath White

    Yes, it seems to me that making welfare a Cambridge property (or a soft fact) would render 1-4 consistent. I think then the major question would be why anyone would care about their welfare in that sense.

    April 26, 2007 — 13:32
  • Hi Alex,
    Sorry, I didn’t think I was denying (4)–though maybe I am implicitly w/o realizing it! I agree that if George’s good plans are fulfilled after George’s death, and if George has permanently ceased to exist at death, George is no better off as a result of his plan being fulfilled, since he’s permanently ceased to exist, and so can be neither well nor badly off.
    My strategy is to deny (2) as stated, insofar as (2) leaves open the possibility that a future free choice F might increase welfare in the present, before F comes to pass. If F comes to pass after George’s death (even one second after George’s death), and if George permanently ceases to exist at death, George is in no way better off as a result of F; he is, after all, no more. I have no problem biting this bullet, since I’m not convinced that it entails that the objective value of George’s endeavor isn’t increased as a result of F hooking up with it in the proper way, and I think this might satisfy the salient intuition driving your position. If, on the other hand, George doesn’t cease to exist at death, then F’s obtaining can increase his welfare when F in fact obtains—not before.
    Right, my own view is that a morally perfect being must always do the best. But I don’t want to saddle the argument with it, and I’m not sure I need such a strong a claim to adequately defend (P3). Strictly speaking, all I need is the claim that, when it comes to deciding the eternal destinies of human beings, God necessarily does what is best. This seems to me fairly plausible. My wife sometime has this worry, which she frames as an objection to Pascal’s wager, that God will act capriciously in deciding our eternal destiny. Assuming that there is a uniquely morally best action for God to take when it comes to deciding how to treat human beings after death, which seems likely—this doesn’t strike me as a case where incommensurable goods might be in play—we might need to appeal to something like the principle that God necessarily does what is best when determining eternal destinies to allay concerns about a capricious God.
    I like the Jewish tradition you mention; thanks for sharing it. But I’m not sure I see its relevance. This is no doubt just me being obtuse. Could you perhaps say another word or two about how it’s supposed to work?

    April 26, 2007 — 17:37
  • Ben:
    Thanks. That is very enlightening, and I will need to absorb it.
    “Strictly speaking, all I need is the claim that, when it comes to deciding the eternal destinies of human beings, God necessarily does what is best.” — Actually, my argument says nothing specifically about human beings. Maybe one could read “one” in (2) as referring only to human beings, but if that were the case, there would be no inconsistency.
    Claim (3) was that it is possible that there is a person x such that, etc. This person might be one of those one-day-only angels, if there are any such.

    April 26, 2007 — 18:30
  • Luke Gelinas

    Right, now I see. I don’t have a rejoinder ready to hand, but this has been an interesting discussion. Thanks.

    April 26, 2007 — 19:20
  • David Gordon

    If (3), then (3′): Metaphysically possible, there was a person x who permanently ceased to exist and who made a morally good plan whose fulfillment required an action after x’s permanent death.
    (2),(3′), and (4) are incompatible; (1) need not be added to produce an incompatible set. Thus, the difficulty raised here has nothing to do with free actions or whether there is a fact of the matter about future free actions. The difficulty is no more a problem for open theists than for those who reject this view.

    April 27, 2007 — 17:03