Sunday, April 13, 2008

Rationality and Reflective Endorsement

The following has the ring of truism: belief aims at truth, and preference/desire aims at the good. But Liz Harman argues against the latter claim on the grounds that loving a person as they actually are may lead us to prefer this actual state even to an alternative that would be better in every important respect. (Suppose you have a disabled child, now grown up. You love this person as they are, and so do not wish that their disability had been cured at birth -- even assuming this would have been better for them and everyone else too -- because then they would have grown up to be a radically different person from the one you actually know and love.)

My initial response is to simply bite the bullet and insist that love makes us irrational in these respects. It is not, strictly speaking, rationally warranted to be biased towards the actual, worse state. The better state is always preferable (it warrants preferring). It's just that the contrary bias is one we may be happy to have. It's fortunate that we prefer our loved ones as they actually are. Indeed, these biases are reflectively endorsable in a way that is familiar to sophisticated consequentialists: we do better to be biased in these ways rather than single-mindedly pursuing the good. But that doesn't mean the attitudes in question are rationally warranted (in the strictest sense). It just shows that sometimes we foreseeably do well to be irrational.

One conclusion you might draw from this is that we shouldn't be too concerned about having warranted attitudes. I think that's exactly right. My writing sample defended a form of 'rational holism', according to which we are instead advised to reason according to those norms which we can reflectively endorse. I thus agree with Julian Nida-Rumelin that it would be perfectly advisable to "refrain from point-wise optimization because you do not wish to live the life which would result." (He actually says it would be 'perfectly rational', which I earlier endorsed too, but for now I want to draw a sharper distinction between advisability and rational warrant.)

Further: if rationality and warrant and so forth are meant to be action-guiding -- concerned with reasons we can follow -- then perhaps we should simply identify these with 'advisability', and give up on the notion that preference aims at the good, as Liz suggests. That's compatible with a continued insistence that it's the impartial good which is the ultimate normative source. It's just that we end up with a neater and more coherent framework overall if we tie rationality more closely to reflective endorsement, rather than identifying it with the strict and alienating impartial ideal.

Anyway, I'm basically just thinking aloud here, and I'm not even too sure what my question is. (Perhaps: "What is the best way to think about rationality?") But if anyone else can shed some light on the issue, that would be immensely helpful...


  1. > You love this person as they are, and so do not wish that their disability had been cured at birth -- even assuming this would have been better for them and everyone else too

    Seems to me that might be more of a misunderstanding of the hypothetical than illogicality of love. Since in the hypothetical there exists the parent of a 'abled' child who is happy that their child is not disabled.

    Or it could be a matter for (largely western) culture related to a reluctance to admit a disabled persons life might be inferior to an abled persons in a material sense, so one twists how one views hypotheticals on that basis rather than due to love itself.

  2. Or just that "preference/desire" is a poor term for the dispositions we have toward those we love?

    Most of the time, our saying we prefer X possiblity to Y state of affairs now goes along with our evaluating the counterfactual "were X the case, I'd be more pleased" as true.

    But I think there's a disconnection there. We can prefer our loved ones just the way they are, but we might still be more pleased after a positive change.

    That disconnect suggest to me that we're just fundamentally using the wrong term for our dispositions toward the qualities of our loved ones.

  3. I don't know. I'm actually inclined to take a sort of hard line here -- one that might well make me unpopular. I'm not sure about this, but let me give it a try.

    So I have a disabled child, Sarah, and I love her for who she is -- for being Sarah. In no sense is my love for Sarah conditional on her being disabled. I'd love her no matter what.

    Why on earth would I not prefer that she had been born healthy? We're supposing that Sarah would have been better off. When something would make someone I care about better off, I want it to happen. When some past thing would have made someone I care about better off, I wish that it would have happened.

    Suppose I am indifferent to whether Sarah was dropped on her head as an infant (the cause of her disability) -- or maybe I even prefer the actual state of affairs where she was so dropped. I think this criticism is both natural and apt: if you really cared about Sarah, you'd be sorry that something bad happened to her.

    Now of course, it's probably not a good idea to go around saying things like "I wish Sarah wasn't disabled", even if they're true. Because they pragmatically implicate that Sarah is disappointing.

    I think that fact drives a lot of the relevant intuition. I suspect another (probably related) factor is that we are often reluctant to admit that disabled people would be better off, were they not disabled.

  4. Jonathan - I've some sympathy, and in fact I think that's the view an impartial benevolent spectator should take. But note that Sarah is now grown up, and her personality, values, etc. have been shaped by her disability (deafness, say, since there's a flourishing Deaf culture to enhance this aspect of the story). It's true that if Sarah's deafness had been cured at a young age, you would have loved her just as much. And we're stipulating as part of the scenario that in fact she would have been (slightly) better off. But she also would have become a radically different person, not someone you can presently recognize in the Sarah you actually love (though of course you would have grown to love this alternative Sarah too). All the particular quirks you're so attached to would be replaced by different quirks -- which, again, you would have become attached to, but that's different from being attached to them now. And it's our actual attachments that shape our preferences.

    So hopefully that makes it a bit clearer how one might sincerely have the preferences Liz describes. I'm very confident that some people (incl. myself) really do have these sorts of biased preferences. So this raises the question: is it irrational?

    Paul - I'm not sure I understand your comment. Why doesn't the disconnect instead show that pleasure is not the relevant thing to assess here? (Another example: Suppose an angel offers to go back in time and radically change your life history so that you end up with completely different friends, interests, etc., and a slightly greater net happiness.)

    In any case, my post is primarily addressing the rationality of preferences/desires, not in "the dispositions we have toward those we love". So it is not the 'wrong term' for my purposes. What reasonable preferences we can have about (situations involving) loved ones is precisely the question I'm interested in.

  5. "I think this criticism is both natural and apt: if you really cared about Sarah, you'd be sorry that something bad happened to her."

    To which one makes the natural and apt response: sure, focusing our attention purely on the bad aspect, of course I'm sorry Sarah suffered that. But look at the big picture. This was a transformative experience, an essential part of making Sarah who she is today. And I'm very attached to who she is today. So all things considered, I'm really glad things turned out as they did. (I'd prefer that Sarah could be just as she is without having suffered the harms of deafness. But those harms are not so great that I'd be willing to give up who she now is in order to be rid of them -- and, like most deaf people, she feels the same way, so I hope you're not about to accuse her of insufficient self-care!)

  6. I'm inclined simply to reject the truism: belief doesn't aim at truth, but at practical utility, and preference aims not at the good but at something like your 'advisability'. I think there is a sense of 'desire' where desire aims at good (it's a very different sense from that according to which it can be identified with preference, which may be preference out of a selection of non-good things) but to say that it aims at the good is a long way from saying that it aims at the best. And really that's what Harman's argument requires: that aiming at the good is optimizing rather than (at least sometimes) satisficing.

  7. Richard, to clarify: that's precisely why I think the word "preference" is wrong. We ought to reserve the notion of preference for when we have a disposition to bring about X, AND we think we'd be happier were X the case.

    Abberant cases -- where we think we'd be happier were X the case, but we don't have a disposition to bring about X -- like the angel thing -- or where we have a disposition to bring about X but don't think we'd be happier afterward -- just don't make sense in that context. So it's wrong to say that those preferences are irrational because they don't aim at the good, or are rational because preferences don't need to aim at the good -- we're just not talking about preferences.

  8. Paul - you must be speaking a different idiolect to me. My term 'preference' certainly applies to these cases.

    But never mind the word. I'm talking about the kind of mental state which one expresses by saying things like, "I wish X were the case," or "I don't regret that Y happened," or "I'd rather the world be in state S1 rather than S2", etc. I take it these are expressing a judgment of sorts, a rational evaluation (not merely a behaviour disposition). It's these judgments (and their rational status) that I'm interested in.

  9. This whole notion just seems mysterious to me. Do you really think some kind of judgment is being made when someone loves her disabled child for what he is and fails to actively regret his condition/wish it were different? I'd be inclined to deny that there's anything so fancy going on. Or that there should be.

    On reflection, I think you're right, I shouldn't have succumbed to the instinct to reduce that kind of preference to facts about dispositions to change states of affairs and to hedonic states. Perhaps I've been hanging around too many revealed preference economists. But I still think there's a fundamental mismatch between our ordinary notion of preference and these cases.

    Maybe a better way to think about it is that certain kinds of relationships between people preclude evaluative judgments of that sort. It's just not appropriate to have a preference ordering over the states of those we love, or the appropriateness is constrained in some sense. (Appropriate to wish they weren't alcoholics, not appropriate to wish they weren't blind, etc.)

    That helps us make sense of the opposite case as well. If my lover is perfectly healthy, I'm something of a jerk if I prefer her that way -- if I think that I would prefer her less if she were disabled. That's just not the kind of evaluation we make about those we love.

    If that's the case, I think Harman goes off the rails: we simply don't have preferences over those states, the preferences we *do* have can still aim at the good.

  10. "If my lover is perfectly healthy, I'm something of a jerk if I prefer her that way -- if I think that I would prefer her less if she were disabled."

    I think you're confusing the concept. It's not that I would "prefer" my lover less if she were disabled (this clumsy locution seems to be talking about my counterfactual desires, rather than my actual preferences). I hope that I would continue to love and cherish her no matter what. Nonetheless, I do prefer that my lover remains healthy: when I bring to mind the alternative possibilities, I find that I have a preference for the actual state of affairs. I would rather she remain healthy than not (for her sake as well as mine). Note that we might not go around saying such things for the reasons Jonathan points to. But there's no question at all that we (at least, I) do have these preferences.

    (And in the earlier cases as well, it's not as though we refrain from all comparative judgments of states of affairs which involve our loved ones. No, we positively prefer that the actual state of affairs obtains rather than some counterfactual scenario where they turn out radically differently.)

  11. This strikes me as vintage Nozick. To wit, the whole experience machine thought experiment is designed to show that there is something important, normatively, in being in touch with the real world. That it is good to be in touch with the true.

    If that's the case, this makes sense of loving, for example, your disabled child instead of his superior, if non-existent, twin.

    Alas though, I have no assistance to offer as to the best way to think of rationality. I do agree, for what it's worth, that we shouldn't worry too much about rationality in regards to our preferences.

  12. Scott, I'm not sure about that. Nozick's thought experiment merely highlights that we have desires for external objects, i.e. we care about more than just our mental states. It doesn't straightforwardly follow that the actual external state of the world is preferable to some other possible external state of the world (which would be just as real and true were it to be realized).

    Indeed, we may often prefer to change how things are in certain respects, but that doesn't mean we're "out of touch" with reality. So I don't think the standard Nozickian insights are really what's doing the work here. I think it's more to do with the peculiar role of attachments in the good life. (As mentioned in my response to Jonathan, I think that unrelated persons should probably prefer that the superior twin had existed instead, since we don't have any personal attachments to bias our evaluations.)

    Just to clarify, on your last point: I do think that it's important to reflect on our preferences, and subject them to normative assessment (and, if need be, revision). It's just that a narrow conception of rational 'warrant' may not be the appropriate norm to assess them against. But there's some broader sense of rational 'advisability' which I think is very important.

  13. Richard, please strike from your mind my mention of Nozick. You're right--completely irrelevant. Sometimes one will take any flimsy insight as an excuse to comment.

    This all has a flavor of time travel, you know? Would I prefer the current state of affairs, where my child has grown up 30 years with a birth defect, or would I rather go back in time and have that defect repaired? He would have had a happier life and so would I, and yet, counterintuitively, I wouldn't make the change, and many others agree.

    So are we irrational? It's possible, but it's far from clear. In a sense, changing the past (assuming any of this is possible) would change who I am. I'd be losing myself to a twin with different memories and experiences. Failing to go back and tinker with things, by these lights, is almost self-defense.

    Now, let's look at another similar problem (it may be within the same class Harman was discussing, I don't know). Would I change the present to bring about a future state of affairs that is more different from the present than an alternate future state of affairs? Concretely, say my child has grown up, 30 years with defect and all, would I heal him with a new procedure that would cure him of the defect? Or would I prefer to keep the situation closer to the status quo?

    Now here, if the answer is keep things closer to the status quo, I agree the impulse is irrational. But I also think far fewer people would have that counterintuitive impulse, so it's less troubling. I would go ahead and heal the child.

    Here's another puzzle, related, that I first read Steven Landsburg mention: I know, if I have ten more children, I will love those ten more children more than any other possession I have. And yet, given their huge value, I probably won't have ten more children. Most people probably have the same possible feelings, and the same behavior.

    Thank you for responding to my comment. It's enjoyable talking with you.

  14. Yeah, I like your 'self-defense' point (cf. my old post on regret). It naturally extends to also preserving the identities of those we love. That is, it's not entirely clear that the proposal to go back in time and cure baby Sarah's deafness would really be better for our Sarah at all. Maybe it would effectively mean that she is wiped out of existence and replaced by a different, better-off, person. So maybe this is a 'worse' outcome according to our partial standards of goodness, and so no exception to the claim that preference aims at the good (in some loose sense). However, I also tend to think that any sort of partiality or self-favouring bias is irrational in the strictest sense, unless you really think that you or your loved ones matter more, all things considered, than everyone else. In other words: preference aims at the impartial or all-things-considered good. But that's another (more controversial) issue.

    Moving along... The 'extra children' case is another nice example (which Liz Harman also discusses). What these cases show is that the 'Reflection Principle' (i.e. have the same attitude now that you expect to reasonably have in future) for desires is false. Supposing you will go ahead and have extra kids, we can expect that in future you will be very glad you did. But - Liz argues - that doesn't show that it's now reasonable for you to want to have those extra kids (especially if it means you cannot provide for your existing children so well, etc.). This is because you don't yet have the attachments that could legitimize such a bias, so there's no excuse not to presently prefer the greater good (which, we may assume, is not to have 10 more kids).

    I think similar remarks would apply to the question of now healing the 30 year old. Assuming it would make him better off (rather than causing a sense of alienation, etc.), one rationally ought to prefer this change. But, again, if you left the disability in place we can expect that in a few decades later still, neither of you would regret this failure. So it just goes to show that having no regrets isn't sufficient to show that a choice was right at the time.

    (This is relevant to real world debates about abortion, curing deaf babies, etc. Love and identity-based biases may cause us to later be glad that, e.g., the teenage mother didn't abort her child, or that a child grew up deaf. Importantly: this doesn't show that it's not a bad outcome. So these preferences we have about existing people shouldn't stop us from endorsing abortion, curing disabilities, etc.)

  15. What paper of Liz Harman's are you referring to?

  16. '‘I’ll Be Glad I Did It’ Reasoning and the Significance of Future Desires'


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.)