Saturday, May 16, 2009

Confusing Welfare and Happiness

Larry Solum "highly recommended" the paper 'Welfare as Happiness' by legal scholars Bronsteen, Buccafusco and Masur. I'm not sure why. As Solum himself explains in a follow-up post, the paper's main argument (that hedonism must be true because our lives are constituted by subjective experiences, and hence nothing can affect our life unless it affects our experiences) rests on an equivocation. Much of the paper continues in this vein. For example:
What it means for something to be a bad experience for an individual is for it to make the individual feel bad. Negative feelings define bad life experiences, whereas positive feelings define good ones.

It could be argued that whether a moment is positive or negative for a person does not depend solely on that person's physiological experience of it. Indeed, disagreeing with our thesis may well require such an argument. But what else could make it positive or negative?

This passage neglects the fact that there are two different things we might mean in talking of 'bad life experiences'. We might mean to talk of subjectively unpleasant experiences, as the authors do. (But then their claims are tautological, when surely they had hoped to establish a substantive conclusion.) Alternatively, we might mean to talk of a life experience (or event) that was bad for you, i.e. undesirable for your sake. For a possible example of the latter, consider the experience of being subtly mocked (without realizing it). Though the undesirable feature of this circumstance is not subjectively transparent to you, nonetheless you (and others who care about you) generally have reason to hope that no such event actually befall you -- or so one might plausibly claim. This is the substantive question of welfare, which the authors fail to address or - it seems - even recognize.

In Solum's follow-up post, the authors set out their core argument as follows:
each of us has a veil of experience, and anything that happens outside that veil of experience and never affects it (even indirectly) has no effect on our lives.

Here they appear to confuse the metaphysical question whether an event "affects" you, in the sense of altering your intrinsic properties, with the normative question of welfare: whether it "affects" you in the sense of being an event that you have self-interested reasons to care about. (In fairness, I think even Shelly Kagan may have once made a similar mistake.) These are wildly different questions, and if one thinks that they are linked in any way, this would require substantial argument. It is far from obvious that welfare supervenes on our intrinsic properties. (Presumably only hedonists will believe this.) Sometimes people talk of one's "life" or "life story" in a broader sense that includes all of the external events and relations that could conceivably be relevant to assessing one's life along any dimension (one's success, popularity, and - of course - welfare). It would then follow trivially that one's welfare supervenes on the intrinsic properties of one's "life" in this broad sense. But this is compatible with any of the main theories of welfare, since one's "life" in this broad sense includes all manner of external events and relations.

I was also struck by the authors' self-defeating arguments against objectivism. Consider the following:
According to the objectivist view, not only don't the individuals know what's good for them, but their view of what's good for themselves doesn't determine what's good for them--no matter how considered or accurate (in terms of happiness) a view it is.

But the authors' own view is no different in this regard. They are hedonists, not preferentists. Even if I care about truth more than happiness, the authors will paternalistically insist that happiness is in fact what's good for me. For them, individuals' own views of "what's good for themselves" are strictly irrelevant -- no matter how "considered" and well-informed our self-regarding preferences might be. (Of course, even preferentists think that people might have false beliefs about welfare: some people believe the competing theories, after all! But there's no avoiding that.)

Their main objection to preferentism is that people may have preferences about distant events that are intuitively irrelevant to their welfare. (They mention 'Sheila the environmentalist', who passively hopes that a rare foreign squirrel avoids extinction.) But this merely shows that preferentists need to restrict which desires count in determining one's welfare, say to those that concern one's own "life story". One might wish to critically examine whether various such proposed restrictions are sufficiently well-defined and non-circular to do the job. But the only such view that the authors consider is the explicitly circular view according to which only "self-interested" preferences count for determining your self-interest. They then respond:
A self-interested, restricted theory of welfare demands that the individual actually receive some benefit before one can say that her welfare has increased; this conception of "benefit" is rendered meaningless unless the individual actually experiences the benefit. To claim otherwise--to argue that an individual's welfare can improve without that improvement registering subjectively--is to welcome Sheila and her Sri Lankan squirrel back into the fold.

But this is effectively just to assert that no independent restriction can be found. There's no argument here, and they appear completely oblivious to the fact that philosophers like Derek Parfit (with his "success theory") have actually proposed some prima facie plausible candidates for this role -- proposals that they're really obliged to engage with. It's extremely frustrating.

[Compare 'psychologists mangling philosophy'. As a general rule, if you're going to attempt scholarly work that's beyond your disciplinary expertise, it would seem wise to consult with a colleague from the relevant discipline, so that they might flag these sorts of problems.]

17 comments:

  1. Richard,

    Thanks for taking an interest in our work.

    My co-authors and I understand the view that a life experience could be bad for your sake without affecting your subjective experience, but we argue against that view. If you are subtly mocked but don’t realize it, and if the mocking never harms you directly or indirectly in terms of your subjective experience, then we believe it does not decrease your well-being — how well your life is going for you. We think that it’s hard to make sense of the words “for you” without reference to your subjective experience. There’s more on this in our forthcoming response to Larry Solum’s thoughtful comments, as well as in our paper. “For you” could refer to the satisfaction of your preferences, but both restricted and unrestricted preference-satisfaction theories face problems we consider decisive, as discussed below.

    Along the same lines, we don’t believe that you have self-interested reasons to care about anything other than your subjective experience. We find it hard to conceptualize how something benefits you if it doesn’t make you feel good. Having a lot of money, or even good health, or fame and success, are all means to the end of positive subjective experience. In what way are those things good for you if they don’t make your personal experience of life any better?

    In your own post about “stakes and sakes,” you seem to acknowledge at least the intuition that something done after a person’s death does not affect her well-being. The most natural explanation for that intuition is that well-being is captured conceptually by subjective experience. This also explains the intuition that the fulfillment of distant preferences, unknown to the person who holds those preferences, does not affect that person’s well-being.

    You’re right that in the sentence you quoted (“According to the objectivist view …”), we arguably use some loose language by referring at one point to preferences rather than to subjective experience. However, several things are worth pointing out: (i) because preferences often overlap with subjective experience, conflating the two as shorthand while discussing something else isn’t self-defeating; (ii) we couldn’t be clearer in the paper about the contrast between our position and the preference-based view; perhaps the most important part of the paper is devoted to exploring that contrast; and (iii) even in the sentence you quoted, we distinguish ourselves from preferentists by using the word “accurate,” a word you omit in favor of “well-informed” when describing the sentence.

    You suggest that we are attacking a strawman when we discuss the version of preferentism wherein “only ‘self-interested’ preferences count for determining your self-interest.” But it’s natural to assume that whatever makes a preference not “self-interested” also makes it not count for determining one’s self-interest. We considered this version of preferentism as well as unrestricted preferentism. Although we did not discuss Parfit’s success theory, our arguments against it are plain from our discussion of preferentist and objective-list theories, as well as from our dialogue with Solum. Specifically, we think it arbitrary for the success theory to distinguish between two categories of desire, neither of which has any effect on one’s experience of life, on the ground that one category is deemed to be truly characteristic of one’s success or well-being whereas the other is not.

    Finally, we agree with you that it’s a good idea to consult a colleague from the relevant discipline, and we consulted more than one philosopher about this paper.

    Once again, we appreciate your interest in the paper and aim to be entirely open in explaining our position, so as not to leave readers frustrated as your post suggested you felt.

    Best,

    John Bronsteen

    ReplyDelete
  2. Hi John, thanks for your reply. Let me take another shot at expressing some of my concerns.

    "Having a lot of money, or even good health, or fame and success, are all means to the end of positive subjective experience."

    Do you really mean to espouse psychological hedonism here? Even setting aside moral motivations, I have many ultimate ends besides "positive subjective experience". Granted, such paradigm "merely instrumental goods" as money and fame are not among them. Better examples would be social concerns (e.g. for the good opinion of my friends) and success in my personal projects (e.g. doing genuinely good philosophy).

    Now, it simply isn't true that I only care about what my friends think of me, or about doing good philosophy, as a means to "positive subjective experience". These things are instead among my ultimate ends, on a par with happiness itself. I would genuinely prefer to live a moderately happy life with genuine friendships and doing genuinely good philosophy, than to live a slightly happier life of farce. These are all genuine non-instrumental concerns of mine, and none has lexical priority over the others. It should also be fairly clear that none of them are purely moral/altruistic concerns. It's not as though I care about what my friends think of me for anyone else's sake. Why would it matter to them (except insofar as they have a sympathetic concern for me)? The point is that these are my concerns.

    Perhaps you didn't really mean to deny this plain psychological fact, but instead claim that I'm irrational to have such concerns. I'm not sure what reason we have to think this, though. It doesn't seem remotely plausible to me. I think this comes out especially strongly in third-person cases. Imagine a loving mother who cares about her son's welfare above all else. She'd do anything for him, to give him the best life she can. Is she thereby committed to plugging him in to the experience machine? That strikes me as insane. Even bracketing moral concerns, and just focusing on what's best for the child himself, it seems to me intuitively obvious that it would be far more desirable to raise the child so that he can actually achieve something meaningful with his life. (At least, I know I'd want that for any child of mine, just as I want it for myself.) Doping out on the experience machine strikes me as a pathetic life: bad, not just for the world, but for the person doing it.

    You ask: "In what way are those things good for you if they don’t make your personal experience of life any better?"

    I find this an odd question. Whether "personally experienced" or not, succeeding in the projects I care about makes my life better. It means that my efforts were not in vain. You may ask, "How is it bad for you to have wasted your efforts and failed to accomplish anything that you had hoped to in life?" But I am not sure how to answer this, since I cannot imagine what else to say that could be more obvious. It is as if I were to ask you, "In what way is pain bad for you, if it doesn't get in the way of achieving your goals?" Such a question would betray a kind of monomania: letting one aspect of welfare blind me to all others.

    ReplyDelete
  3. "we arguably use some loose language by referring at one point to preferences rather than to subjective experience"

    My objection there was not to your "language", but to the substantive argument. You seemed to be rejecting objectivism on the 'anti-paternalist' grounds that each individual is the ultimate authority in determining their own good. But this is a consideration that equally justifies rejecting hedonism in favour of preferentism. After all, I might (non-morally) care more about learning the truth than being happy. If this is so, then by what right do you impose someone else's (hedonistic) conception of the good life on me? How can it be better "for me" to live the life I less prefer?

    Think of it this way: hedonism is a kind of objective list theory, just with a much shorter list. You're still forcing a single conception of the good life on all people, regardless of their own individual concerns or preferences. (Imagine telling Morpheus that you will shortly wipe his memory and plug him back into the Matrix to live a life of blissful ignorance under the yoke of the machines. He would be appalled, not just on moral grounds, but for his own sake. Now, you may just insist that this is "for his own good", whether he likes the idea or not. But it's clearly no less of a paternalistic and alien intrusion than if Morpheus had unilaterally decided to "rescue" people out of the Matrix whether or not they wanted to live in his scummy reality.)

    "we think it arbitrary for the success theory to distinguish between two categories of desire, neither of which has any effect on one’s experience of life, on the ground that one category is deemed to be truly characteristic of one’s success or well-being whereas the other is not."

    Parfit's view is that we can distinguish desires that are about our own lives from others that aren't. I am not sure what you think is "arbitrary" about this. Parfit offers a couple of examples that clarify the contours of the distinction he has in mind, but it's intuitive enough to easily grasp, and so not "arbitrary" in the sense of unnatural or gerrymandered. It also doesn't seem normatively arbitrary. Intuitively, the reason why Sheila's 'squirrel' preference isn't relevant to her welfare is just because whether the squirrel survives or not has nothing to do with her or her life. On the other hand, such facts as whether I do good work, and whether I'm secretly mocked by my apparent friends, are intuitively very relevant to me and my life. So if you want to deny this strong intuition then I think the burden is on you to provide a real argument. Merely asserting that any non-hedonistic view must be "arbitrary" doesn't strike me as the least bit convincing.

    In sum: your argument against preferentism rested heavily on the squirrel case, and the unsupported assertion that any restriction that excludes the squirrel will collapse into hedonism. I offer Parfit's view as an obvious counterexample. His view excludes the squirrel without collapsing into hedonism. So I don't see that you've offered any arguments that would apply against his view.

    ReplyDelete
  4. We think that caring about whether you do good work and caring about whether you’re mocked by your friends are in the same category as caring about the squirrels’ survival — rational preferences for things other than one’s own well-being. That these things do not relate to your well-being is far less intuitive with respect to the work and the mocking than with respect to the squirrel, but only because in real life those things would usually affect subjective well-being.

    We certainly understand the proposed distinction: that the first two preferences are “about your own life” whereas the third is about something else. But we view the distinction as irrelevant to well-being, and also perhaps as more apparent than real. Does Sheila’s preference for the squirrels’ survival become “about her own life” if she has donated money to organizations that help the squirrels? What if she has gone to Sri Lanka to help the squirrels herself? What if she has nurtured and protected the squirrels as though they were her family? My understanding is that at some point along that spectrum, Parfit would say that Sheila’s preference for the squirrel’s survival becomes sufficiently “about Sheila” that the squirrel’s survival affects Sheila’s well-being.

    That line of reasoning suffers from this problem: How much effort Sheila puts into saving the squirrel is irrelevant to the reason people intuit that the squirrel’s survival doesn’t affect Sheila’s well-being. The reason for the intuition is that it’s obvious that Sheila’s subjective experience is unaffected by something she doesn’t know about that’s taking place so far away and that doesn’t even involve other people — in a way that it can’t be obvious that one’s subjective experience will be unaffected by mocking or low work quality (even when those are stipulated not to affect subjective experience).

    You could dispute our claim about the source of the intuition, but the claim is bolstered by the related problem for the success theory of post-death events. As Parfit points out, it would be indefensible for a success theorist to claim that such events have no effect on well-being, because post-death events are indistinguishable in any relevant way from pre-death events that do not affect subjective experience. But counting post-death events toward well-being is deeply counterintuitive to many people, who don’t see how you can be affected by anything once there is no longer a “you” to be affected. Introducing the word “retroactive” does nothing to ameliorate this problem.

    A success theorist has her examples (preferences against being secretly mocked, and preferences for doing high-quality work), and we have ours (post-death events and squirrels). But there’s a big difference. The success theorist’s examples problematically pump intuitions in a way that ours do not. In real life, being secretly mocked seems almost certain to redound to one’s subjective detriment: one will learn of the mocking, be treated less well by the mockers, miss out on happiness-enhancing opportunities as a result of being truly disliked, etc. And doing high-quality work is hard to disentangle, not conceptually but in terms of intuition, from feeling like one is doing high-quality work (and perhaps also from the sense of explicitly other-regarding value that we may attach to such work).

    Of course, a committed success theorist will say that she has considered and rejected such considerations as the true source of her intuitions. But they are very plausible candidates to be the source of most people’s intuitions. By contrast, what besides the subjective theory can explain people’s intuition that post-death events do not affect a person’s well-being?

    Again, thanks for engaging in this exchange with us. We think it’s helped clarify and sharpen the distinction between the competing conceptions.

    ReplyDelete
  5. Hi John, could you say a bit more about how it can be rational, on your view, to prefer to not be mocked? For whose sake would this outcome be good? I pointed out that (unlike the squirrel case) it cannot plausibly be interpreted as an altruistic concern. If it is good for anyone, then it must be good for the agent herself. But you insist that it cannot be good for her. So it looks as though this outcome must not be good (desirable) at all, on your view. (Unless you want to say that such things are desirable for the sake of the world as a whole, independently of any individual's welfare. But, again, it seems very unnatural in these cases to 'moralize' the self-regarding desire in this way.)

    Regarding desires that are 'about your own life', I agree that this is a graded concept. But the mere existence of borderline cases is no reason to think that a distinction is "more apparent than real", so long as we can also identify clear cases on either side of the divide. Success theory can naturally accommodate the graded nature of the concept by weighting desires in proportion to how central they are to our lives, rather than trying to force the artificial precision of a binary distinction. So I don't see any problem here.

    "How much effort Sheila puts into saving the squirrel is irrelevant to the reason people intuit that the squirrel’s survival doesn’t affect Sheila’s well-being"

    As with your earlier claim of psychological hedonism, I suspect that this is empirically false. I suspect that if you ask people to imagine that Sheila dedicated much of her life's work to safeguarding these squirrels, before leaving the country and never hearing about them again, most people will no longer share your intuition that whether the squirrels survive (and hence whether Sheila's efforts were in vain) is irrelevant to her welfare. But I guess the plausibility of the latter claim is just something that each reader has to judge for themselves.

    "post-death events are indistinguishable in any relevant way from pre-death events that do not affect subjective experience"

    This is controversial. I'm inclined to agree with Parfit that ultimately the best worked out view will have us conclude that posthumous harms and benefits are possible after all. But one might deny this, e.g. on the grounds that only presently existing things can be harmed or benefited. So I think you are underestimating the diversity of defensible views in this vicinity.

    "what besides the subjective theory can explain people’s intuition that post-death events do not affect a person’s well-being?"

    Again, the most straightforward explanation is nothing to do with hedonism. It is instead the simple principle that only presently existing things can be harmed or benefited. (You implicitly acknowledge this yourself in saying that "many people... don’t see how you can be affected by anything once there is no longer a “you” to be affected.")

    I should add that I think acceptance of this principle derives from confusing the metaphysical and normative questions that I distinguish in my main post. Once we understand this distinction, it no longer seems nearly so mysterious how we could, for your sake, have reason to want your projects to succeed even after you die. It seems to me quite commonsensical that ordinary sympathy for others could lead us to want their life's work to not be in vain.

    ReplyDelete
  6. Overall, it's worth noting that Success Theory captures overwhelmingly more of our intuitive case-by-case judgments about what things we (or others) have reason to care about for our sakes. The arguments in favour of hedonism, by contrast, tend to appeal much more to abstract, 'top-down' theoretical intuitions. (As you wrote in your first comment, "We find it hard to conceptualize how something benefits you if it doesn’t make you feel good.") As a general methodological rule, I think that intuitions about particular cases are generally more reliable than top-down theoretical intuitions, since it's easy for people to make abstract philosophical errors, or to fall into the grip of a theory that blocks their ability to conceptualize something that others have no difficulty at all in understanding. Further, I think we can debunk the specific theoretical intuitions you rely on in your paper, as deriving from various abstract philosophical errors as previously explained (e.g. conflating different senses of 'life', 'affects', etc.). So that is why I am inclined to judge that (something like) success theory is much the better account of welfare on the whole.

    Having said that, I think that reasonable people can disagree on this matter. It certainly isn't my aim here to conclusively establish this view. I just wanted to explain why I didn't think the arguments in your paper did much to advance the dialectic, or offer reasons to anyone who doesn't already agree with you. Worse: I got the impression, when reading your paper, that you thought you were offering knock-down objections to the competing views, and this is manifestly not the case.

    ReplyDelete
  7. The question is this: what is the reason that only presently existing things can be harmed or benefited? The answer is that harm or benefit is conceived as referring to subjective experience.

    As you suggest, it’s possible that most of the arguments have been aired and that in the end, the choice between your conception of well-being and ours comes down to people’s intuitions or considered judgments. We would be interested in finding some way to design an empirical test that would be acceptable both to us and to the philosophers who espouse competing views.

    (So as not to be accused of ducking your question, we will give an abbreviated answer to the question whether it is rational to prefer not to be mocked. First, we are not convinced that a preference must be for anyone’s “sake.” Second, even if it turned out that our position commits us to deem that preference irrational, we would not have a problem with doing so. The intuitive resistance to such a claim stems from difficulty in accepting the stipulation that there will never be any subjective harm from the mocking.)

    ReplyDelete
    Replies
    1. You say :"The question is this: what is the reason that only presently existing things can be harmed or benefited?"

      I'm sorry, but this is obviously false. It's quite possible to harm and benefit dead people, and this is a common intuition that a large percentage of the human population shares. The intuition you have that you can't harm someone once they are dead isn't the knockdown argument you think it is. In fact, it's a weird intuition not shared by the majority of the human race.

      Haven't you ever heard someone justify their actions because "It's what [name of dead friend or relative] would have wanted."? Haven't you ever heard someone become angry if the memory of a dead person is shown disrespect. In fact, people generally get even angrier at you if you disrespect a dead person than they do if you disrespect a living one.

      The obvious reason people hold these beliefs is that the wishes of a person remain part of their wellbeing, even after they are dead. So if you do something because that's what a dead person would have wanted, or defend the reputation of someone who is dead, you are improving their wellbeing.

      Now, of course you might ask where we stop honoring the dead, since we obviously don't think a 7th century peasant's views should be taken into account when determining social policy. But the Success Theory that Richard has already mentioned does a good job of taking that into account.

      That being said, even if your arguments against Success Theory are valid (and I don't think they are), that still doesn't establish hedonism. If my only choice is to bite the bullet of hedonism, or bite the bullet that Sheila's life is affected by the distant squirrels, I'll bite the second bullet gladly. Believing that distant events can affect our well being is far more palatable than believing that it's okay to lie to people if they never find out, or that it's okay to force people to plug into experience machines.

      Delete
    2. Thanks for your input, Evan. Let me start with your main point and then move on to the other things you wrote.

      1. Can someone’s well-being be affected after she’s dead?

      You make one argument on this score, which is that people say things like “It’s what Jill would have wanted” and “Don’t disrespect the dead.” People certainly do say those things. The question is whether the fact that they say those things reveals that they believe Jill’s well-being can be increased or decreased even after she’s dead. You make no argument about that question, saying only that the affirmative answer is obvious.

      It’s not obvious. Let’s start with disrespecting the dead. One possible reason that Jack might object to someone’s disrespecting Jill after she’s dead is that Jack thinks that would diminish Jill’s well-being. Another possible reason is that Jack holds some deontic or even religious or other sort of belief about the wrongness of disrespecting the dead. Such deontic or religious or other beliefs need not have anything to do with Jill’s well-being or other consequences of the act of voicing disrespect.

      So, which explanation is more plausible — that Jack is concerned about Jill’s post-death well-being, or that Jack has some other sort of objection to speaking ill of the dead? Well, consider your own point that “people generally get even angrier at you if you disrespect a dead person than they do if you disrespect a living one.” No one disputes that living people have well-being, but it’s at minimum a lot less clear that dead people do. So the well-being explanation is arguably at odds with your own point that people tend to be more upset when one speaks ill of the dead than when one speaks ill of the living. If the reason people get upset when one speaks ill of the dead were that people think that diminishes the dead’s well-being, then why wouldn’t they get more upset when one speaks ill of the living, who more obviously possess well-being that can be diminished?

      Similar arguments apply to Jack’s desire to do what Jill would have wanted. It could be that Jack thinks this will increase Jill’s well-being, even though she’s dead. But there are plenty of other explanations, too. For one thing, doing what Jill would have wanted could be a deontic or religious commitment of Jack’s. Or alternatively, it could be that the cultural mores of doing what the dead would have wanted affect the quality of life of the living: knowing that people will care about what you wanted after you’re dead affects how you live. This latter point is one of the principal justifications in law for honoring people’s testamentary documents (wills, trusts, etc.).

      I’m not saying it’s impossible that people think the dead have well-being, or that it’s impossible that such a belief explains why people make statements like the ones you cite, but you haven’t really advanced your ultimate claim by citing those statements because there are other explanations for the statements that seem at least equally plausible.

      Delete
    3. 2. Your other objections to hedonism

      At the end of your comment, you reject hedonism because it requires “believing that it's okay to lie to people if they never find out, or that it's okay to force people to plug into experience machines.”

      Let’s clarify what this discussion is about. The issue is simply, “What is well-being?” That’s what my co-authors and I wrote a paper about, and it’s what Richard and I went back and forth on in his post and comments. The issue is not, “What does morality require?” or anything like that. Someone could take the view that well-being is happiness, and that well-being is the only important thing. Or someone could take the view that well-being is happiness, but there are things other than well-being that are important, too. Or someone could take the view that well-being is preference-satisfaction, and that well-being is or is not the only important thing.

      My view is that well-being is happiness, but I don’t make any claims about morality. So it could certainly be the case — and, if so, would be entirely consistent with my position on well-being — that it’s not okay to lie to people if they never find out and that it’s not okay to force people to plug into experience machines. There are plenty of things that increase well-being that people widely believe are not okay to do. That’s not what we’re talking about here.

      3. The Success Theory

      The burden on someone who holds my view of well-being isn’t to show that hedonism is a perfect conception of well-being, immune from serious attacks. Rather, my burden is to show that hedonism is better than other conceptions of well-being. If it’s the best conception there is, then that’s enough.

      So if you want to attack what I’m saying about hedonism, it’s really valuable and maybe even essential to say what you think would be a better conception of well-being. Then we could compare the problems you can identify in hedonism with the problems I can identify in whatever conception you think is better.

      If your choice is the success theory, then you should say what you take that theory to be. Suppose, for example, that you believe the best conception of well-being is that it’s the satisfaction of preferences that are “about you,” wherein preferences count more heavily toward your well-being the more they are “about you.” To even get off the ground, such a conception needs to explain what it is about a preference that makes it “about you.” Is it merely the strength of the preference, or is it something else — and what is that something? For example, suppose that Jill wants two things: (1) she wants to make the cheerleading team, and (2) she wants there to be world peace 500 years after she dies. Is preference #1 more “about her” than preference #2? Most people might well intuit that it is. But why? Could it be that preference #1 affects her experience of life, whereas preference #2 does not? If so, to what degree is the success theory driven by hedonic considerations? And if not, then what is it that “about you” means? Unlike the hedonic theory, the success theory needs substantial explanation simply to say what the theory is. Only when we know what it is can its limitations be compared with those of hedonism. And I suspect its limitations will be legion; but if you disagree, then by all means tell me what makes one preference more “about you” than another preference.

      Delete
    4. 4. Overclaiming

      You wrote: “The intuition you have that you can’t harm someone once they are dead isn’t the knockdown argument you think it is.” When did I ever say I thought it was a knockdown argument?

      It is you, not I, who may be accused of overclaiming. You twice used the word “obvious” to describe claims of yours that are in fact highly controversial. You also wrote that “a large percentage of the human population shares” the intuition that dead people can be harmed and benefited. How in the world do you know that? You try to demonstrate it via the points I discussed in section #1 of this comment, but those points don’t come close to establishing it.

      Delete
    5. >The burden on someone who holds my view of well-being isn’t to show that hedonism is a perfect conception of well-being, immune from serious attacks.
      I'm not sure your arguments about the "veil of experience" necessarily support hedonism. Not if you define hedonism as the belief that the only thing that determines someone's welfare level is what emotions they feel. Imagine an entity which only cares about what happens in its veil of experience, but doesn't care what emotions it feels. For instance, this entity desires to perceive itself being President of the USA, even though being President would be a huge responsibility that would cause it much psychological torment. You offer to put this entity into an experience machine that will simulate it being President, and it gladly accepts. It lives a life of great torment, but that is what it wanted.
      Such an entity doesn't care about whether its presidency was real or an illusion. It only cares about its "veil of experience," and cares nothing for anything outside it. But I wouldn't call it a hedonist.

      Here's another good argument against hedonism: Imagine a young person is molested. At the time they are too young to understand what's going on, so they aren't upset. A few years later they get older, realize what happened, and are horrified. According to your theory, the thing that harmed them was realizing they were molested, not the molestation itself. If they had been killed in a car accident before they got older then the molestation wouldn't have harmed them at all.

      Delete
  8. > Unlike the hedonic theory, the success theory needs substantial explanation simply to say what the theory is.

    You seem to be arguing from an intuition that human values should be simple. I don't share this intuition. I think human values have high Kolmogorov complexity and it would probably take millions of words to define them. In my view, if a theory can sum up our values shortly, simply, and easily, then it is probably wrong.

    Still, if you want success theory defined a bit better:

    First, divide desires into desires to have certain experiences, desires that aren't directly experienced, but still involve you (ie, that your spouse doesn't cheat on you), and desires that do not directly involve you (ie world peace). And yes, the boundaries are fuzzy. That's a strength, not a weakness, of Success Theory. In real life things don't form sharply defined categories.

    The first two types of desires are always part of your wellbeing. The last type is related to your wellbeing in a way that is proportional to how much effort you put into making it come true. For instance, if Jill spends years working for world peace her life will have gone much worse if it doesn't happen. If she did nothing her life hasn't gotten worse at all.

    >When did I ever say I thought it was a knockdown argument?

    In addition to this blog, I also read your discussion with Larry Solum. You made great use of the idea that the dead cannot be harmed in that discussion, so I thought that you must consider it extremely important.

    >You also wrote that “a large percentage of the human population shares” ...How in the world do you know that?

    In addition to the widespread beliefs I mentioned before, many people also have a desire to leave "legacies" and be respected after they are dead. The most extreme example of this was the many ancient rulers who desired eternal fame. If they were successful, and we still remember them today, I think you can reasonably say they have greater welfare than those who have been forgotten.

    > If the reason people get upset when one speaks ill of the dead were that people think that diminishes the dead’s well-being, then why wouldn’t they get more upset when one speaks ill of the living

    I think it is probably due to an aversion to harming the helpless. People also get angrier when helpless people like the mentally disabled are mocked, because they cannot defend themselves.

    Also, it seems to me that mocking the dead is worse if the dead person in question cared a lot about their reputation when alive, which implies that it is about wellbeing.

    > So it could certainly be the case...that it’s not okay to lie to people if they never find out and that it’s not okay to force people to plug into experience machines.

    Granted. But even if it was still immoral, it seems wrong to me to say that it makes someone better off to plug them into an experience machine against their will, or that it doesn't make someone worse off to ruin their reputation if they never find out.

    ReplyDelete
    Replies
    1. 1. “You seem to be arguing from an intuition that human values should be simple. I don't share this intuition. I think human values have high Kolmogorov complexity and it would probably take millions of words to define them. In my view, if a theory can sum up our values shortly, simply, and easily, then it is probably wrong.”

      I think your view is widely rejected. The natural world is no less complicated than human values, yet no one argues for analyzing it via the reverse of Occam’s Razor.

      2. “First, divide desires into desires to have certain experiences, desires that aren't directly experienced, but still involve you (ie, that your spouse doesn't cheat on you), and desires that do not directly involve you (ie world peace). And yes, the boundaries are fuzzy. That's a strength, not a weakness, of Success Theory. In real life things don't form sharply defined categories.

The first two types of desires are always part of your wellbeing. The last type is related to your wellbeing in a way that is proportional to how much effort you put into making it come true. For instance, if Jill spends years working for world peace her life will have gone much worse if it doesn't happen. If she did nothing her life hasn't gotten worse at all.”

      First of all, you made no attempt to explain what the term “involve you” means. You gave an example of something that you said does involve you (spouse not cheating) and an example that you said does not directly involve you (world peace), but you didn’t say what makes the first one involve you but the second one not directly involve you. (And by the way, which is it: involve or directly involve? You start with the former and then switch to the latter in the very same sentence.) The point is not that the “boundaries are fuzzy.” The point is that if you have no way to explain what makes something “involve you,” and you want people to just rely on their intuitions for this, then that is a major weakness of your theory of well-being. The whole point of a theory is that it helps resolve intuition conflicts. If the theory itself depends upon intuitions merely to explain what the theory is, then its value is severely compromised. This is an additional reason, by the way, that your view about the value of complexity is so thoroughly unusual.

      Second, the theory you describe is radically incomplete in that it describes only whether something affects well-being and not how much it affects well-being, except with respect to the third category. So, consider your first category — desires to have certain experiences. Suppose two such desires get fulfilled. Does each affect the person’s well-being equally, and if not, then what determines which one affects her well-being more? Does it depend on how strong the desire is, how much effort she puts into it, how happy it makes her, or some combination of these or other things? And what if a desire from each of the first two categories is fulfilled? Does one of the categories outrank the other, or are they equal? The third category does rank how much desires affect Jill’s well-being, but how does that ranking interact with desires from the first two categories? For example, suppose Jill works hard for world peace, works a little bit for squirrel preservation, and doesn’t put any effort at all into making the cheerleading team (though she has a preference for making it). If world peace and squirrel preservation happen after Jill’s death, you’ve said that the former would count more than the latter toward her well-being; but you haven’t said how cheerleading would rank among the three things. Would it be first, second, or third? And what determines that?

      Delete
    2. Note that it’s not nearly good enough to say that these are all complications that would take “millions of words” to explain. For one thing, no theory of well-being (or any other concept) that I’m aware of uses that excuse. And for another thing, you’re not even being asked to defend the theory, but merely to say what it is. So far, you haven’t done two basic and essential things toward the end of saying what the theory is: explain what makes something “involve you” and explain how much different desire-fulfillments affect well-being relative to one another.

      This matters because my claim is that hedonism is a better theory of well-being than any other theory. To show that claim is wrong, one needs to propose an alternative theory that’s better. So far, the thing you’ve described is not even a theory, let alone a better one.

      I’ll also point out that the one and only part of the theory that’s made specific in your description seems hopelessly counterintuitive. Suppose Jill has two preferences in your third category, one for world peace and one for squirrel preservation. Suppose her preference for world peace is very, very strong: she cares about it and values it more than anything else in the world and would gladly lay down her life for it. But she never does anything to advance it because she doesn’t know how to. Squirrel preservation, by contrast, barely means anything to her, though she has a very mild preference for the squirrels to live rather than die. One time when she was seven years old, though, she wrote a letter to a Congressman on behalf of the squirrels as part of a class project. After Jill’s death, the squirrels survive (for reasons unaffected at all by her letter) and world peace is achieved. Had Jill foreseen these events and been asked about them, and had she believed (as I do not) that post-death events affect her well-being, then she would have said that world peace would improve her well-being much more than the squirrels’ survival. Yet you say it’s the opposite because all that matters for such preferences is “how much effort you put into making them come true.” I think that’s counterintuitive. (Of course, I don’t think any events after death affect well-being. But I suspect most of the people who do think so would find your approach counterintuitive.)

      I hope this provides at least an idea of how hard it is to come up with a good theory of well-being, or even a theory of well-being at all. Once hedonism’s problems are compared to those of an alternative, hedonism tends to look a lot better.

      Delete
    3. 3. “I think it is probably due to an aversion to harming the helpless. People also get angrier when helpless people like the mentally disabled are mocked, because they cannot defend themselves. 

Also, it seems to me that mocking the dead is worse if the dead person in question cared a lot about their reputation when alive, which implies that it is about wellbeing.”

      All of this seems just as consistent to me with a deontic view as it does with a welfarist one. Maybe people believe one shouldn’t speak ill of those who can’t defend themselves because it harms those people — especially if they’re dead, because then they really can’t defend themselves. Or maybe people believe one shouldn’t speak ill of the dead (or others who can’t defend themselves) because it’s just wrong, regardless whether anyone’s welfare is diminished by it.

      Again, due to the alternative explanation that’s at least equally plausible, I don’t think these examples advance the case for post-death events affecting welfare. And neither does the point about desiring eternal fame. People desire plenty of things that don’t count toward their welfare, even on your theory (i.e., things that don’t “directly involve them” and that they don’t put effort into making come true). People who care about their legacy may care about it for reasons other than their welfare, or they may mistakenly believe that it affects their welfare (perhaps, for example, because they can’t or don’t want to truly come to grips with their own mortality and its implications). I still think it’s highly intuitive to consider post-death events irrelevant to welfare, though I understand you disagree.

      4. “It seems wrong to me to say that it makes someone better off to plug them into an experience machine against their will, or that it doesn't make someone worse off to ruin their reputation if they never find out.”

      These are the most canonical arguments against hedonism. My co-authors and I address them in our article on which the Solum dialogue and this dialogue are based. In brief, we think those examples pump intuitions in unreliable ways because it’s hard to really believe that someone is never affected by their reputation being ruined, and hard to believe basically anything about the experience machine. (People’s intuitions about the machine could be polluted, for example, by their desire not to forego the opportunity to use their lives to improve the lives of others; or by their not being convinced the machine will actually work; or by their fear that while on the machine, they will be vulnerable to harm from those in the real world.) We’re far from alone in thinking that the experience machine doesn’t do the work against hedonism that it’s supposed to do. Just to take one of many examples, L.W. Sumner (perhaps the leading philosophical authority on the concept of well-being, and certainly no hedonist) said this in his treatise on well-being. There’s no need to relitigate the issue here, but my point is just that it can no longer credibly be claimed that the mere words “experience machine” constitute a major blow to the hedonic theory.

      Delete
    4. 5. The veil of experience

      Your example is very interesting, but it doesn’t undermine our point. We’re not saying that hedonism is the only theory that counts only experiences. We’re saying that it’s more plausible than other theories, and one reason is that — all else being equal — counting only experiences is more intuitive than not doing so. All else is not always equal, and you’re right that someone could have a theory of welfare that counts only experiences but that isn’t a hedonist theory or a plausible theory. In any event, our view (like Bentham’s) is that the positivity of experience is what determines welfare.

      6. “Here's another good argument against hedonism: Imagine a young person is molested. At the time they are too young to understand what's going on, so they aren't upset. A few years later they get older, realize what happened, and are horrified. According to your theory, the thing that harmed them was realizing they were molested, not the molestation itself. If they had been killed in a car accident before they got older then the molestation wouldn't have harmed them at all.”

      Our approach to this is clear from everything else we’ve said: there’s a difference between harm and wrong. As with the experience machine and any other stylized, incendiary example (all of which are entirely legitimate, as long as one can sort out which intuitions are coming from welfare concerns and which aren’t), this example may undermine some people’s intuitions about hedonism. But if you pick another theory, I’ll be able to give examples that are even bigger problems for that theory than these examples are for hedonism.

      Delete

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