Wednesday, September 17, 2008

Psychologists Mangle Philosophy

Haidt and Bjorklund ('Social Intuitionists Answer Six Questions about Moral Psychology') suggest a model ('SIM') according to which human moral judgments typically result from gut reactions, post-hoc rationalization, and social conformity, rather than good philosophical reasoning (though the latter is also possible; just rare). No great surprise there, I should think. But at the end of the paper they grossly overreach in suggesting alleged "philosophical implications" of this psychological claim.

"1. Moral truths are anthropocentric truths":
On our account, moral facts exist, but not as objective facts which would be true for any rational creature anywhere in the universe. Moral facts are facts only with respect to a community of human beings that have created them...

Most people aren't philosophers, therefore moral realism is false? I'm not seeing the connection here. The core of the paper was talking about moral judgments or beliefs. Now they're suddenly drawing conclusions about the objects of those beliefs -- the "moral facts", not just our thoughts about them. Isn't this a whopping non-sequitur? Is it yet another instance of the lazy conflation of belief and truth (as suggested by their subsequent reference to the "moral truths held by members of [a] society") or is there some more charitable interpretation that I'm missing?

"2. The naturalistic imperative: all ought statements must be grounded, eventually, in an is statement" They offer as an example of a good is-to-ought inference:
Sheila is the mother of a 4-year-old boy, so Sheila ought to keep her guns out of his reach.

But of course even non-naturalists can endorse this (enthymatic) inference. That's because there are further moral premises implicit in the background, e.g. that mothers ought to keep their children safe from harm. This is not an example of empirical grounding of a moral claim, but of the empirical application of a more general moral principle.

This raises a more general worry I have about naturalistically inclined philosophers. When they claim that ethics must be "empirically informed", they could mean one of two things. If they simply mean that empirical contingencies affect how moral principles apply to our actual situation, then that's utterly uncontroversial. (A consequentialist will be interested to learn what would actually bring about the best consequences, for example.) But if they mean something stronger, that empirical findings should influence our first principles -- not just which conditional claims turn out to apply, but what should be conditionally claimed in the first place -- I don't see the faintest reason to believe this.

"3. Monistic theories are likely to be wrong"
If there are many independent sources of moral value [Haidt classifies humans moral intuitions under 5 categories: harm, fairness, authority, purity, and ingroup/outgroup] then moral theories that value only one source and set to zero all others are likely to produce psychologically unrealistic systems that most people will reject.

Note, again, the sloppy conflation of descriptive and normative "value" (belief and truth). Haidt likes to accuse liberals of failing to "understand" anti-gay sentiment, for denouncing it as immoral 'homophobia' rather than "appreciating" that it's a moral judgment which derives from our evolved 'purity' module. We parochial liberals limit ourselves to the first two "sources of value", and "neglect" the latter three. But here's the thing: just because there's a mental module which inclines people to judge yucky things as immoral, doesn't mean they're right. (Duh.) Sloppy, sloppy, sloppy.

4. Haidt and Bjorklund's fourth "implication" is that a moderate relativism is true:
If relativism is taken as the claim that no one code can be proven superior to all others, then it is correct, for given the variation in human minds and cultures, there can be no one moral code that is right for all people, places, and times.

Says who? Again, nothing in the descriptive theory of SIM implies any such thing; these guys are just making stuff up. (Aside: we should also take care to distinguish situation-sensitivity vs. relativism proper.)

They go on to reject radical relativism, understood as they claim that "no one code can be judged superior to any other code". Apparently a code that "radically violate[s] the values and wants of a large proportion of the people involved" fails to qualify as "well formed". It's nice that they don't want to endorse slavery and the like, but this ad hoc principle reinforces my sense that they're just making stuff up as they go along.

"5. The methods of philosophical inquiry may be tainted":
If the SIM is right and moral reasoning is usually post hoc rationalization, then moral philosophers who think they are reasoning their way impartially to conclusions may often be incorrect. Even if philosophers are better than most people at reasoning, a moment's reflection by practicing philosophers should bring to mind many cases where another philosopher was clearly motivated to reach a conclusion and was just being clever in making up reasons to support hear already-made-up mind... The practice of moral philosophy may be improved by an explicit acknowledgment of the difficulties and biases involved in moral reasoning. As Greene has shown, flashes of emotion followed by post hoc reasoning about rights may be the unrecognized basis of deontological approaches to moral philosophy.

This raises interesting issues about the genetic fallacy, and the extent to which explanatory reasons can debunk normative reasons. In assessing "clever" arguments, I don't usually care what "motivated" the author. Whether deontology is the most rational and systematically defensible approach to ethics seems logically independent from the question of its psychological "basis" in human proponents. I guess a spectator might draw on genetic considerations as a kind of Bayesian meta-evidence. But from within the practice of philosophy itself, I'm not sure these considerations are really all that relevant.

So, there you have it. Five alleged "philosophical implications", of which the first four are blatant non-sequiturs, and the fifth is at least dubious. Whether SIM is true or not doesn't really affect any of these questions. (Would the truth or falsity of SIM really influence your beliefs here, or even shift your credence in the slightest?) We need to do a priori philosophy to determine (1) whether there's a uniquely rational moral system, (2) which empirical contingencies have moral relevance, (3) whether disgust - as well as harm - can ground moral facts, (4) whether some moral codes are better than some or all others, and (5) in what conditions 'genetic' considerations are philosophically relevant.

Those are the fundamental philosophical questions here, and (so far as I can see) empirical psychology in general - and SIM in particular - doesn't have any implications for any of them. Certainly Haidt and Bjorklund haven't shown any. Perhaps other readers might have more luck?

24 comments:

  1. Whether deontology is the most rational and systematically defensible approach to ethics seems logically independent from the question of its psychological "basis" in human proponents.

    The authors need not claim a logical, but merely an evidential, connection between psychology and morality. It is hard to see how such a claim could be contested. Appeals to authority clearly do have epistemic import in some cases, and so do appeals to origins. The cause of a belief can be evidentially relevant, just as the identity of the believer can.

    ReplyDelete
  2. Come on, you can't have it both ways. If you are going to claim that the fifth claim by Haidt and Bjorklund is " is at least dubious" you cannot also acknowledge the point I raised in my comment. In fact, to this date I don't quite understand your position on this particular issue. You like to say that people who rely on genetic considerations as guilty of committing the "genetic fallacy" and yet you recognize that such factors can be sometimes be used legitimately against think-tanks, religious apologists, and the like.

    ReplyDelete
  3. I am very sympathetic to the authors' point of view, so maybe I can help you find the missing argument. I think they are appealing to something like Occam's Razor. If moral intuitions are fully explained by evolution, sociology, and psychology, why would we need to posit some sort of eternal truth to morality?

    The only reason we tend to believe in morality is that we have moral intuitions. But if we have those intuitions for a perfectly good reason that doesn't involve moral realism, then we no longer have any need for moral realism.

    This explains their first point, and if you accept it, I think #3-#5 will follow. Your criticism of #2 seems valid, though.

    ReplyDelete
  4. I'm with Richard. Without reading the original, perhaps the authors (and Pablo?) are just collapsing 'morality' into the anthropology of in-situ morality. That's something they can study with the tools they have, so perhaps it's their empirical hammer making all problems look like nails.

    @ PoeticExplosion:

    Sounds like the Gil Harman example about cat-burning? I.e. our moral disgust reactions are psychological states with psychological causes, and we need not posit an objective moral property of 'wrong' to explain them.

    But who ever seriously thought that the only interesting role moral properties/reasons could play would be to explain moral intuitions? It seems like doing that rules out a hell of a lot from the very outset...

    ReplyDelete
  5. PoeticExplosion, I think Richard's main point is that the authors are confusing descriptive morals with prescriptive morals. Just because we have certain intuitions, that doesn't mean they're the morally right ones.

    Richard, number 5 seems to be just be a psychologistic stance towards moral reasoning, and it can be rejected in the same way Frege neatly rejected psychologism about mathematics and logic.

    ReplyDelete
  6. GreatLeapingCrab, I largely agree with Richard's first four objections, and am certainly not collapsing morality into anthropology.

    Lvb, I don't think number 5 expresses a psychologistic stance towards moral reasoning, but instead the rather simple point that we should distrust arguments which are rationalizations of pre-existing beliefs formed by emotional or affective processes.

    ReplyDelete
  7. Pablo, as Richard also said, the soundness of an argument does not depend on the psychological motivations behind the argument's construction. The argument stands or falls on its own. It seems to me that the opposite attitude would be a kind of psychologism.

    Besides, I don't think they're simply claiming that one should distrust arguments that are post-hoc rationalizations, but rather that a lot (maybe all) of moral arguments ARE of such a nature. See in particular the last sentence in (5): "As Greene has shown, flashes of emotion followed by post hoc reasoning about rights may be the unrecognized BASIS of deontological approaches to moral philosophy" [My emphasis.]

    I'll concede it's possible I read too much into this. It's not a very important point anyway.

    ReplyDelete
  8. Pablo - my claim is that qua philosophers we should consider the first-order arguments on their merits, but that external assessors might take genetic considerations into account (just as they might take authority and other philosophically irrelevant 'epistemic clues' into account). It's also worth noting that the epistemic weight of such meta-evidential clues is inversely proportional to our grasp of the first-order evidence. For the former only matter insofar as they give us clues about the latter.

    (I'd also add that - contra H&B - SIM in particular doesn't seem to tell us anything important here.)

    PoeticExplosion - "If moral intuitions are fully explained by evolution, sociology, and psychology, why would we need to posit some sort of eternal truth to morality?"

    I respond to precisely this argument in my post 'Explaining Beliefs'. (In short, the mistake is to think that moral realism adds more entities to our ontology, when really it merely requires that some moral beliefs be more rational than others. Occam's razor clearly has nothing to say about the latter question.)

    It's also worth noting that this old argument doesn't require SIM in particular, but in fact could be pulled out no matter what the true psychological theory happens to be. As such, it would be odd for H&B to suggest it as an 'implication' in their theory, when (like, say, '2+2=4') it would equally well be an implication of any competing theory. So I don't think that's what they had in mind.

    ReplyDelete
  9. Pablo, as Richard also said, the soundness of an argument does not depend on the psychological motivations behind the argument's construction. The argument stands or falls on its own. It seems to me that the opposite attitude would be a kind of psychologism.

    No. Motivations can be evidentially relevant without being logically relevant. If Fermat says that a certain formula is a theorem, that gives me good reason to think it is, even if the truth of this assertion is logically independent from Fermat’s speech act of asserting.

    Besides, I don't think they're simply claiming that one should distrust arguments that are post-hoc rationalizations, but rather that a lot (maybe all) of moral arguments ARE of such a nature.

    I agree. I focused on the first claim because it was being disputed by some people in this thread. As far as I can tell, no one disputed the second claim, which stands and falls with the relevant empirical evidence.

    my claim is that qua philosophers we should consider the first-order arguments on their merits, but that external assessors might take genetic considerations into account (just as they might take authority and other philosophically irrelevant 'epistemic clues' into account).

    I don't see much justification for this principle. As the authors write, "Even if philosophers are better than most people at reasoning, a moment's reflection by practicing philosophers should bring to mind many cases where another philosopher was clearly motivated to reach a conclusion and was just being clever in making up reasons to support hear already-made-up mind." Why should we ignore these facts?

    Is it because "the epistemic weight of such meta-evidential clues is inversely proportional to our grasp of the first-order evidence"? But clearly you cannot assume that philosopher's grasp of first-order evidence is perfect, so there's always room for invoking meta-evidential clues in expert philosophical discussion.

    Perhaps our persistent disagreement over this issue can be traced down to our different views about the status of philosophy as a discipline. Unlike you, I don't think the findings of philosophy warrant much confidence, and am as a consequence much more willing to rely on external clues to adjudicate philosophical disputes.

    (Of course, I also think that you have a much higher opinion of philosophy because it's psychologically comforting to think one has devoted one's professional life to a worthwhile activity. I'm sure dentists, too, think of dentistry as highly important.)

    I'd also add that - contra H&B - SIM in particular doesn't seem to tell us anything important here.

    Why not? If it is indeed true that facts about origins have epistemic import, then a view on which "human moral judgments typically result from gut reactions, post-hoc rationalization, and social conformity, rather than good philosophical reasoning" does tell you something important.

    ReplyDelete
  10. Pablo, well I guess if you're offered an ethical argument by a philosopher and you're not interested in engaging with the actual argument, then you can use your knowledge about the philosopher and his motivations to estimate to what extent you should believe in the argument's conclusion.

    But if you actually take the time to examine the argument itself, what will be the use of this second-order evidence? Either the argument is sound or it's not. The only alternative is that you believe the psychological basis for the argument somehow fundamentally affects its soundness (in which case you accept psychologism).

    I agree with Richard on this one: as a philosopher, you should be interested in examining the actual logical status of arguments, not external evidence. And your Fermat example brings up a good analogy: _I_ might choose to accept Fermat's assertion that a given formula is a theorem on the basis of his authority, but if I were a mathematician, his authority would mean nothing to me, and I'd better investigate to see if there's an actual _proof_ of the purported theorem.

    ReplyDelete
  11. Pablo - "Why should we ignore these facts?"

    As I see it, the role of an academic discipline (be it philosophy or mathematics) is to advance the state of [first-order] inquiry in that discipline. This is based on the conjecture that long-term progress is more likely to be made by assessing arguments on their merits than by psychologizing their proponents, etc.

    That is, I believe there is a tension between:
    (1) forming the most accurate beliefs at present; and
    (2) bringing about even more accurate beliefs in future.

    I think the role of academics (qua members of their academic community) is to advance the latter goal, not necessarily to maximize their own epistemic ends.

    Though, actually, I'm dubious that rampant psychologizing (as implemented by actual humans) would even help the first goal. Evidence suggests that the most egregious bias of all is the tendency to ascribe bias to others, and so excessively discount their views. A perfect bayesian might do better to carefully weigh the meta-evidential clues for the little they're worth, but humans are far more likely to mess this up.

    ReplyDelete
  12. You know that I am utterly unsympathetic to reducing morality to experimental psychology. But in the spirit of charity, a recreation...

    (1') aka a revision of (5), The methods of philosophical moral inquiry are tainted because "moral judgments are the result of gut reactions, post-hoc rationalization, and social conformity."

    (2') There are moral truths. The best candidates for moral truths are agreements among people about what is and what is not moral. Hence, moral truths are a matter of coordination among our human moral judgments.

    (3') aka (1)-(4), Hence, moral truths are anthropocentric, hence monistic theories of value are unlikely to be correct (because of the pluralistic reduction base of moral truths), hence moderate relativism is correct (because of variation of moral views), hence the natural imperative (moral truths reduce to moral judgments and their statistical distributions).

    What do you think?

    ReplyDelete
  13. Jack, personally I am convinced that moral values are grounded in, and obtain all their authority from, human psychology (but that does not mean that I think the soundness (perhaps I should say validity) of specific moral _arguments_ depends on psychology). So I actually agree with a lot of what you (and the authors Richard is criticizing) wrote.

    However, if you are a moral realist - that is, if you believe there exists a unique objectively superior moral system whose values can be uncovered through rational deliberation - then presumably you don't care about the empirical fact that different groups of people have different moral systems. That just means all of those systems (except maybe one) are flawed, and does not imply that one should accept moral relativism.

    ReplyDelete
  14. Jack - your (2') ["The best candidates for moral truths are agreements among people about what is and what is not moral."] seems unmotivated, especially in light of the empirical fact that typical moral judgments are formed through such dubious processes. This empirical fact should (if anything) lead us to discount actual judgments, not valorize them. The real question is: what would people conclude if they were to reason better? Philosophical inquiry may be imperfect, but it is our best shot at answering this question. (And regardless of our practical ability to achieve it, the ideal standard is there as a matter of metaphysical fact.)

    ReplyDelete
  15. Richard—A bit of a dialectical impasse: the ”judgments are formed through such dubious processes” only if they are not constitutive of the moral truths.

    Let me try again for (2’).

    (a) Our moral intuitions, and to a lesser extent our moral judgments more generaly, are tracking something. After all, there is surely some reason we have tend to have the intuition that P rather than the intuition that not-P.
    (b) Call what our moral intutions tracking the ”moral truths.”
    (c) The best candidate for what our moral intuitions are tracking is some amount of human coordination, be it of genetic or social cause.
    (d) Hence, the moral truths are some amount of human coordination.

    The psychologist might need another premise, a dreadfully dubious one at that,

    (e) Moral intuitions are all on a par. Hence, we couldn’t being doing better morally. If we acted en masse, we could make a more rational morality true (by systematically changing our moral intuitions). But (these guys think) the moral intuitions are constitutive of the moral truths.

    I don’t know. I think I am with you in calling this a mangling.

    ReplyDelete
  16. That is, I believe there is a tension between:
    (1) forming the most accurate beliefs at present; and
    (2) bringing about even more accurate beliefs in future.


    This is an interesting argument. But what makes you think that psychologizing proponents is less likely to yield fruits in the long run? To the extent that it yields any fruits at all, this approach is, as far as I can see, no more likely to yield them now than later.

    ReplyDelete
  17. I'll tell you a kind of second-order evidence that I _do_ use when dealing with philosophers, and which I personally find very helpful at times -- namely, independent intelligence gauges. There are certain philosophical positions which seem clearly wrong-headed, and yet they can be hard to refute completely. I'm talking about, for instance, truth relativism, phenomenalism, rortyan pragmatism, and other abominations. People who truly hold views like these are always impossible to argue with; they're usually experts at twisting around everything you say to fit their own perverse 'world view', and it can be quite frustrating.

    But then, in my experience these people are always pretty dense when you have discussions with them about other (more clear-cut) things. So even though I can't refute them with first-order arguments (at least not on their terms -- or terms we can both agree on), I still know they're wrong, because they're stupid.

    So I do agree second-order evidence can be useful sometimes.

    ReplyDelete
  18. Pablo - "what makes you think that psychologizing proponents is less likely to yield fruits in the long run?"

    I was thinking something like the following: our total evidence can be treated as the output of a two-place function from (i) first-order evidence and (ii) meta-evidential clues. Assuming the meta-evidential clues can be added at any time, and aren't really susceptible to improvement over time, our best means of increasing our future total evidence is by improving our grasp of the first-order evidence for now. We might say that first-order inquiry is a compounding investment, whereas meta-evidence yields a constant payoff. So we maximize our epistemic profits by investing our attention in the area where growth is possible, namely first-order inquiry.

    On further reflection, however, I guess there isn't any good reason to think that meta-evidential transformation won't have compounding effects. For example, if it causes us to focus more of our attention on those first-order theories that have more potential (say, consequentialism rather than deontology), then this could yield greater fruits in the long run.

    So I retract my claim that such meta-evidence is irrelevant to philosophy in principle. The problem is merely practical: as noted in my earlier comment, it turns out that attempting to incorporate psychological meta-evidence is more likely to lead us astray (through exaggeration and selective use).

    ReplyDelete
  19. Hey Richard, been a while!

    Interesting post. I've been reading a bit of Haidt and find his experimental findings very interesting (although even within psychology they are somewhat contentious). I don't think Haidt plays the "philosopher game" very well, and I can see the validity of your criticisms.

    Still, I'm not completely convinced that you disagree with him as much as you suggest. For instance, you've pretty much always said you were a utilitarian/realist. But further up the thread you mentioned what seem to be to be some very minimalist criteria for realism. I'd definately be a realist by your definition (I think that some sets of 'moral beliefs' are more coherent than others). But I would have thought that I was a non-cognitivist or even nihilist. After all, if Haidt is right, what is the *source* of morality? I'm sure you'd agree that we are moral because of evolutionary and social constraints. Philophers and society as a whole may make progress in ironing out inconsistencies between means and ends, but how do we get to know any moral principle as true?

    We can announce it by fiat - I remember you having said many times that moral questions *just are* questions about pleasure and pain (of course with interesting qualifications - rule utilitarianism, contextualism, and all that stuff), and left it at that. But isn't that just playing into Haidt's hands? You might think it's simply an analytic truth, but that's awfully convenient considering that (at least according to Haidt's empirical evidence) most people across the world use a different definition than you do. For them, it seems that sanctity, obedience and loyalty are equally important as deontological or consequentialist principles. I imagine that you'd respond: "you just fell prey to the naturalistic fallacy!" But look at the two alternatives more closely. Person #1 just states that morality is defined as A. Person number 2 cites evidence that for the majority of people, morality is defined as {A, B, C, D, E}. That's a relevant point to make. Of course, you can stand your ground and just insist that they are *wrong*! You're welcome to do that - after all, you're just ironing out inconsistencies in belief sets, but the consequence (If Haidt is right) is that you'll be talking about something different to the majority of everyday people.

    I wrote too much to go back and proof-read it now, as I've got some work to do! I may have left some thoughts unfinished, but would love to see your response anyway. As always, fun reading your blog,

    Patrick Kerr

    ReplyDelete
  20. Hi Pat! I think my earlier self was mistaken to adopt that 'definitional' approach, for the reasons you point out and more which I discuss in 'Is Normativity Just Semantics?'

    On the "source" of morality, we need to be careful not to conflate two very different questions:

    (i) What causally gives rise to our moral judgments and behaviour? This is a purely descriptive -- psychological rather than philosophical -- question. You'll need to say more if you think anything of philosophical interest follows from it.

    (ii) What are the epistemic grounds which justify some moral views over others? This is a normative philosophical question, and I think the answer is complicated but probably has something to do with rational coherence and its push towards greater universalizability, etc.

    In short: my view now is not that person 2 is making some linguistic error about the meaning of moral language, but that they are making a rational error: they think disgust reactions have intrinsic moral significance, and further reflection would disabuse them of this notion. That's the realist's claim, anyway, and it seems the only way to assess it is to engage in rational reflection and see whether there is some deep incoherence to the other views. The mere empirical fact that people hold these (possible incoherent) moral positions is neither here nor there.

    ReplyDelete
  21. I get what you're saying, and I agree that they are of course two separate questions (in principle, additional premises might link them somehow, but I'm undecided about whether there are such premises right now).

    Let's take the made-up-example of Fred the Layperson. He believes that morality is all about fairness (deontology), happiness (consequentialism), and obedience towards God (authority, one of Haidt's 5). You and I both would be uneasy about the latter. But what rational options are open to us in arguing against Fred?

    We could, as you say, point out that his commitment to obedience could lead him towards either unfairness or suffering. This would show that our moral intuitions are not fully coherent (Incidentally, Dan Dennet argues something along these lines in Darwin's Dangerous Idea). There are a couple of replies open to Fred. First, he can rationalize the inconsistencies away (this isn't really a rational option so we'll write it off). Secondly, he can just agree with you - "Sure, my multiple moral commitments sometimes come into conflict, but nobody said being a good person was easy!"

    I guess there are a few responses to that. You might point out that there must be some kind of rational principle to deal with conflicts such as this. If there is, I'd have to concede that THAT overriding principle was morality! Well, I guess I'll admit that people's heads don't explode when faced with moral dilemmas. However, that just shows that people have a psychological mechanism to solve the problem. Is there also a rational principle that prescribes how to solve it?

    I'm not sure there is. Sure, consistency is great, so Fred might decide that he only cares about obedience when it doesn't conflict with, or perhps only insofar as it functions as a means towards, say, fairness. But is that the only rational solution? Maybe Fred will conclude that some actions are wrong with respect to fairness, and right with respect to obedience, but that both are fundamental facets of morality. By what rational principle would we deny him this option?

    If I were to declare by fiat that morality was about some one thing, I'd probably go for utilitarianism. But I don't think I really have any tricks up my sleeve to rebuff someone who insists that morality is about something else. Well, sometimes - they might be incoherent, but they don't necessarily have to be - if James said that morality was solely about growing as many grapefruit as possible and nothing else, I don't see how that could be incoherent. The only responses I can think of are to say either "Nobody else thinks it's about grapefruit" or "You go ahead and think about grapefruit, but I'm going to devote my time to reasoning about pleasure and pain". Neither response seems to show that James is wrong or irrational. What more can be said about the situation? I guess an important fact is that many people will find the "fruits" of your philosophical labour more useful (unless they are grapefruit farmers). However, that's just the same kind of fact as Haidt brought up in the first place.

    ReplyDelete
  22. Richard,

    It looks like you're defining "X realism" to mean "I can be wrong about X."

    In a narrow way, this could be true of morality and aesthetics. If morality for an individual is maximizing some property (e.g., happiness), then that individual could be wrong in his assessment of which actions will best satisfy his goal. For example, he may misidentify disgust as something that leads to less overall happiness.

    But this is realism in the same sense that mathematics is real. If I accept certain mathematical axioms, certain theorems really follow. (I can make mistakes in my calculations.) However, mathematics does not tell us whether the axioms themselves are really true (e.g.,I can't say that Hilbert was wrong on the grounds that he assumed axioms of infinite dimensional spaces).

    Likewise, it seems that Haidt et al are saying (and I would agree) that our moral axioms are not provably true, but are merely accidents of history. I can describe why I hold to certain axioms (like valuing happiness), but I can never prove the axioms correct. This doesn't mean I cannot think rationally about morality. I can think rationally about morality in the same way I can think rationally about creating art or devising novel mathematical systems.

    I think Haidt suggests that maybe the person who thinks disgust is a fundamental attribute is not confused. Maybe that person really would prefer a strategy that reduced his disgust at the cost of some happiness. IOW, maybe he does have different moral axioms.

    And I see no objective way to validate one moral axiom above another. Moral philosophy is a matter of modeling our moral feelings, yet those models can't predict anything apart from our feelings. And when a model comes into conflict with our feelings, I don't see where moral philosophy gets off claiming the map is the territory. Again, this is not to say that reason isn't useful in moral decision-making.

    ReplyDelete
  23. It's one thing for Haidt to just "say" what his personal philosophical views are. It's quite another to claim, as his article does, that these are implications of his descriptive theory.

    I don't mean to get into the realism vs. anti-realism debate here. (Plenty of other threads for that.) My point is simply that the psychological data Haidt points to are not evidence one way or another. They're completely independent of the philosophical question whether some moral views are rationally superior to others.

    ReplyDelete

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.)