Saturday, September 06, 2008

(Ir)rational Emotions

In his (independently interesting) post on 'Emotions and Moral Skepticism', Andrew Cullison writes:
It seems that if emotions are going to be rational, then the best explanation of this will have to be that they are rational in virtue of the rationality of the beliefs included in the emotion. Why is it irrational to get angry that there is an odd number of pebbles in the fishbowl? Because it’s irrational to believe that this is a bad states of affairs. Why is it rational to get angry at someone who steals money from you? Because you’re rational in believing that this is a bad state of affairs.

One quick point: the cognitive components of emotions are not properly 'beliefs' -- they might even conflict with our beliefs. But I agree with the broader point that cognitive misjudgment is one common source of irrationality in emotion. Might there also be others, though? Justin Oakley argues that emotions are complexes of cognition, affect, and desire; it seems to me that there may be rational failings in respect of all three.

Consider someone who is disproportionately angry about some minor wrongdoing. Their judgment of wrongdoing seems accurate enough;* the problem is instead an excess of affect. The excessively angry agent obsesses over this one problem, to the detriment of more pressing issues. For another example:
Good things generally warrant positive reactions. But envy consists precisely in feeling bad or resentful about the good things that happen to other people. Doesn't that just seem fundamentally misguided? Similarly for schadenfreude, or taking pleasure in others' suffering. It's simply perverse; not an apt response to the normative features of the situation.

Moreover, jealousy may involve a desire for the target to lose the benefit you so resent them for. But it is generally irrational to desire a Pareto loss -- for someone to be harmed and nobody to benefit. So jealousy and the like are irrational for this reason too.

* Now, it may be claimed that the desirous and affective components of emotions should also be incorporated into their cognitive component, in the form of normative claims. Maybe envy doesn't just consist in judging another to have benefited and feeling negative affect about this; perhaps we could say it also involves an implicit judgment that it is bad that the target benefited so. To the jealous desire we might add a judgment that it would be right to deprive them of this benefit. And so on. But if we build everything into the cognitive component to begin with, then it's trivial that it will suffice for all aspects of rational evaluation. So I think the substantive point to note here is that affect and desire are open to rational assessment too, even if they are ultimately incorporated into the cognitive component of emotion.

Incidentally, this proposed melding of cognitive and non-cognitive components of emotion relates to my earlier comments about 'rich' or interpreted phenomenology. Perhaps the broader lesson is that these enriched "components" are not really distinct and independent components at all, but instead draw our attention to a single (conceptually inextricable) state that may be variously described as a kind of judgment, a kind of desire, or a kind of phenomenal affect or 'feel'. Really what it is is just one thing, an affectively motivating intentional state, i.e., an emotion.

17 comments:

  1. I wonder about the sources of irrationality. Suppose you find someone who is fully informed about the object of his anger, X, and yet responds in way that seems disproprotionate. You exhaust the sources of his apparently mistaken response, and you find none. He knows everything that you thought he might have failed to know about X. He then says that fully informed emotional responses are brutal. They might be atypical, but they're not irrational.

    ReplyDelete
  2. Mike - we might need to fill in more details here. For example, does the guy acknowledge that the wrongdoing was actually relatively minor and unimportant in the grand scheme of things? If so, this seems to be in tension with having an enflamed emotional response, for that is effectively to treat it as important and meriting significant attention and cognitive resources. Arguably, it is not merely 'atypical', but indeed 'irrational', to treat unimportant matters as though they were important.

    ReplyDelete
  3. I see Mike's wonder and raise him a suspicion. The suspicion is that the subject of the irrationality judgement isn't the emotion as experienced, but the disposition to think/act that usually goes along with it. To *treat* unimportant matters as important is irrational, but that's not . You can quietly seethe with envy and jealousy, but then be aware of this and think/do the right thing anyway. Emotions don't have to feed into decisions and actions.
    Anyway I know that point's a bit murky, but if we are clear in these thought experiments to isolate emotion as being the only difference (i.e. spell out what happens next as being not dictated by them) then I'm not sure the intuition that emotions can be irrational will be as strong. At least, it isn't for me.

    ReplyDelete
  4. For example, does the guy acknowledge that the wrongdoing was actually relatively minor and unimportant in the grand scheme of things? If so, this seems to be in tension with having an enflamed emotional response, for that is effectively to treat it as important and meriting significant attention and cognitive resources. Arguably, it is not merely 'atypical', but indeed 'irrational', to treat unimportant matters as though they were important.

    Hi Richard,

    Maybe rather than ask whether the object of his anger is all that important, we might focus on whether his response was apt. I realize that we would not see it as the typical sort of response. No doubt, it is statistically strange. But we are asking about his rationality. So if we say, well look this is a small event in the larger perspective. He might well say, "yeah, ok, but the question is not whether it is small in the larger view. The question is whether my response to it is rational". He knows all about the event, he is not confused cognitively. He event ecognizes that others don't respond the way he does. But so? Given that he is fully informed, all that is left is come causal connection between the event and his emotional response. It's unusual, but again, so? Is it irrational? I can't see the basis for the charge of irrationality. Maybe this analogy. You accompany someone to (part of ) Wagner's Ring Cycle. You are both fully informed about this opera and have some musical sophistication. The other person responds with profound emotion while you can't wait for this interminable bluster to end. Is one of you irrational? None of your beliefs are mistaken, you are both well-positioned to appreciate the event. But your reponses differ importantly. In the end, we have the brute fact about emotional responses that seems to elude further analysis. I don't think one of you must be irrational.

    ReplyDelete
  5. woops, typo. Related concern (apologies if you've addressed this before):
    You seem to want to say that the affect component of emotions can be irrational for being in tension with judgement, belief or desire (actual or ideal).
    But affect skates close to other non-voluntary mental events, like perceptions. Perception can be inaccurate or deceptive (i.e. in tension with ideal judgement), and it is irrational to believe what perception tells us when we have good reason to believe it to be so (or if it's wildly anomalous). But the perceptions themselves are surely not irrational, they just represent a 'rational danger' for the imperfect reasoner.
    Cognitive mental states can be rational or irrational, but non-cognitive ones in general aren't apt for the description. If emotions are complexes of cognitive and non-cognitive mental states, then why not just leave it at that? Perhaps there's a clear difference between emotional affect and perception, but I don't really see it just now.

    ReplyDelete
  6. GLC (#1) - let me see if I can ease your suspicions at all. One thing to note is that even if we can restrain our emotions, and so retain rational control over our behaviour, we can't control our thoughts or mind at large. Emotions aren't just feelings; they 'tint' our thoughts and perceptions, direct our attention to some things rather than others, etc. Even if we manage to behave aptly in the end, it doesn't follow that we're perfectly rational; it may be that we had to overcome various irrational tendencies along the way, which threatened (but fortunately did not succeed) to throw a spanner in the works.

    Further, as noted in the main post, it seems to me that even feelings themselves can be irrational. If you feel good when something bad happens, that's irrational -- it's not an apt response to the normative features of the situation. Or so I'd claim. (Do you disagree?)

    Mike - "Maybe rather than ask whether the object of his anger is all that important, we might focus on whether his response was apt."

    But my point was precisely that it is not apt to respond to unimportant things as though they were important. That is "the basis for the charge of irrationality." The rational agent has a measured, well-calibrated, or 'apt' response to the world: he attends more closely, and cares more deeply, about the things that truly matter. Insofar as one fails to do this -- e.g. by fixating on trivialities -- one is failing as an agent, i.e. being irrational.

    [Note that aesthetic qualities are famed for permitting a variety of responses, so I don't think we can infer much from the analogy. Quite apart from the feelings, I'm also permitted to judge either 'That was great' or 'That was tedious'. I don't think we can necessarily assume the same is true of everything else -- I assume you're not a moral relativist, for example.]

    GLC (#2) - Irrationality is not limited to cognitive states. (Cf. weakness of will.) Emotional affect is radically different from perception, for it is part of our response to the world and the reasons we find there. Perception, on the other hand, provides us our input (and clearly isn't responsive to reason the way our emotions are).

    ReplyDelete
  7. But my point was precisely that it is not apt to respond to unimportant things as though they were important.

    I agree with that. Here's what I want to say. I think I can be brief. I'm trying to call into question the claim that an unimportant matter X--one that is typically or statistically responded to in a mildly emotional way--is one that cannot rationally be responded to in an very emotional way. I want to say that someone who acknowledges that X is unimportant might nonetheless respond in a very emotional way to X. And I think there is no obvious way in which such a person is irrational. His response to the charge of irrationality is not that he knows something about X that you don't. He does not claim that he knows that X is very important, while you seem to think it is not. His claim is rather, this is how I in fact respond to matters of this sort, even after knowing all about things of this sort. It's again just a brute fact about this person that he responds in this way to such events. He might say that it is only a contingent fact that most others do not respond as he does. Fully informed emotional responses just happen. Why are you assuming that such responses must converge? What's in dispute is whether they must converge.

    ReplyDelete
  8. Mike, "unimportant" does not mean "typically or statistically responded to in a mildly emotional way". I can easily imagine a society where everyone is excessively angry, and goes into a wild rage at the drop of a hat. This is a society where everyone has miscalibrated emotions, treating unimportant things as though they were really important. Typicality simply isn't anything to do with it.

    "It's again just a brute fact about this person that he responds in this way to such events. He might say that it is only a contingent fact that most others do not respond as he does. Fully informed emotional responses just happen."

    I don't see how any of that is relevant. It may be a brute fact about someone that they have an irrational fear of mice. It's certainly a contingent fact that I have no such irrational fear. So what? It doesn't change the facts about what responses are rational and irrational.

    You seem to be assuming that once someone is fully informed, there's no further room for rational criticism. See my response to GLC #2 above. Even fully informed people can suffer from rational defects such as weakness of will, or fear of mice.

    ReplyDelete
  9. Re: emotions not just feelings
    Agree.
    Minor response: as complexes there are many different parts to them, some of which can be tagged rational or irrational. But the whole doesn't straightforwardly get the tagging of the parts.
    Less-minor response: the 'feeling' part (the affect?) and the 'attention-grabbing' part still don't seem apt for a 'rational' or 'irrational' tagging to me. But perhaps we just disagree.

    Re: irrational feelings
    Yes I think I do disagree that feeling good about something bad happening is in itself irrational. It's unvirtuous certainly, but I'm not a rationalist/virtue theorist about morality so don't see a problem there. I also think envy can be a perfectly appropriate (not 'rational', obviously) response to someone else getting something I myself would like, even if they deserve it. Not very nice of me perhaps, but there's a natural fit with my wanting it in the first place. What's irrational is revising your beliefs about them on the basis of the envy alone.
    If it's a rational requirement to have desires & emotions in a one-to-one line-up with judgements of importance or desert then that's either waaay too strict a notion of rationality or a recipe for moral lunacy.

    Re: "part of our response to the world and the reasons we find there". First conjunct, obviously true. The second I flat out disagree with. Perhaps that's the basis of the disagreement, but if so it looks to be either terminological or programmatic.
    Anyway I find the 'me vs not-me' contrast with perception unconvincing. It's not my area and perhaps I'm confused or semantically deviant, but I thought perception was not just input it involved attention-grabbing and affect. Pain, for example. At least: emotion is not unique in being "part of our response to the world". (Unless all ouchy or yummy sensation is emotion?)

    For purposes of clarity: suppose a radical feminist man is momentarily distracted and aroused (against his principles) in response to some cleavage hoving into his field of view. What by your lights is driving that response to the world? Perception or emotion? And is it irrational?
    My intuition: it's awkward but the tension is not a rational one. And the affect and behaviour-guiding components of perception and emotion are not that easy to separate.

    When drawing the line between me and 'the input' for normative modelling, you might choose to disown the role of V1 but own that of the amygdala (or some other brain structure more closely tied to affect) but personally I don't see why you'd want to do that unless compelled to by a specific philosophical project. If the goal here is to rattle the Humean's cage then I'm not sure it's working.


    PS I also don't see how the Raz stuff supports your view here.

    Anyway. back to work...

    ReplyDelete
  10. I don't see how any of that is relevant. It may be a brute fact about someone that they have an irrational fear of mice. It's certainly a contingent fact that I have no such irrational fear. So what? It doesn't change the facts about what responses are rational and irrational.


    Well, that won't do. Maybe I can help. Maybe you'll agree that you need some basis for the irrationality claim. Take the fear of mice example. Maybe that is an irrational fear, maybe it isn't. If the person is afraid of mice because he thinks they take humans as prey, then it is irrational: it is based on a false belief. Maybe he is afraid because he thinks showing fear to mice is the best way to befriend them, and he really likes mice. But maybe it is none of these. Maybe he is just psychologically unusual. Maybe his beliefs about mice are thorough and accurate. Further, the person needn't be suffering from any pathology. He might be as psychologically healthy as you are. But, despite having accurate and thorough beliefs about mice, and suffering from no pathology, he might remain afraid of mice. Now to the point I've been trying to make. You'll let me know if any of this is at all relevant. If the person is (1) fully informed about mice and responds with fear to mice (2) suffers from no pathology, and (3) you want to say that his emotional response is irrational then (3) you need some basis for the claim. What would it be? I claim that there is no basis that is not question begging. Maybe you'll claim that the example I offer is not possible. The person would have to have some false belief or some pathology. I'd deny it. I know perfectly healthy people that have such fears. Being psychologically unusual does not make you psychologically unhealthy.
    Looking for some basis, you will find yourself in the same position as someone trying to convince a fully informed, psychologically unusual person that it is irrational to eat mud. If the person desires to eat mud and he knows all of the implications of mud-eating, what is the basis for the claim that he is irrational? There is no non-question begging basis, I claim. All you can really say is that the person is statistically unusual.

    ReplyDelete
  11. GLC - I don't think there's a strict 1:1 correspondence -- most situations allow for a variety of permissible responses. Also, I'd say bodily arousal and pain are sensations rather than emotions. I agree that the former are not responsive to reason (and so not open to rational assessment). I'll have to think more about how to clarify the distinction I have in mind.

    Mike - I'm afraid we're going around in circles. I've already explained that "statistical" atypicality is nothing to do with it. And I've explained that the basis for my claims is the idea that the irrationally emoting agent has emotions that are inapt, poorly calibrated, or out of sync with the normative features of the situation.

    For example: to fear something is to respond to it as though it were dangerous. If something, e.g. a mouse, is not in fact dangerous, then it is patently inapt to respond as though it were. We can call this 'pathological', or we can say that 'perfectly healthy people' are capable of being irrational, but the terminology doesn't change the plain facts about what sorts of responses are apt or called for by what situations.

    Similarly, excessive emotion involves treating unimportant things as though they were important. I believe your earlier comment actually agreed with me that such responses are not "apt". It follows that they're irrational.

    If you don't think emotions can be irrational in the way I've described, you only have two options that I can see:

    (1) Retract your earlier agreement and instead deny that it's inapt and irrational to respond to things as though some false normative claim were true.

    Or (2) Deny that emotions involve responding to things 'as if' certain normative claims held. For example, you would have to deny that fear involves responding to something as though it were dangerous, or that enflamed emotion treats its object as important or significant.

    I guess you mean to take the second route, but either strikes me as an awfully large bullet to bite.

    ReplyDelete
  12. I share some of GLC's puzzlement here. I agree that emotions are subject to some sort of normativite evaluation, but why are the norms in question rational? I disapprove of people who get angry at small slights, but I don't think that such people are somehow logically inconsistent, nor do I think they're pragmatically self-defeating. I'd say my disapproval of inappropriate anger has more to do with the moral and social norms I accept than with judgments of consistency. Richard, I take it you want a concept of rationality that permits people to be rational without being inconsistent.

    Does it matter that the aptness of emotions is culturally relative? Most Americans find insect-eating disgusting, while southeast Asians would be happy to eat a bowl of tasty fried grasshoppers. To pick up on GLC's example, feeling aroused by cleavage is inappropriate at a radical feminist commune, and appropriate at a Renaissance Faire. Do the norms of rationality vary from culture to culture, or does their correct application vary from culture to culture, or what?

    ReplyDelete
  13. Oh, crap, read too hastily. Cut the radical feminist/Renaissance Faire example.

    ReplyDelete
  14. "I take it you want a concept of rationality that permits people to be rational without being inconsistent."

    Well, I think that for other reasons: there are any number of logically consistent positions that it would be ridiculously irrational to believe. Solipsism, for example.

    For my above claims though, I think it suffices to note that the implicit 'claims of emotion' are in some sense contradicting obvious normative facts (which the reasonable agent presumably believes). See my response to Mike directly above. There's a kind of inconsistency involved in (i) believing that mice are harmless whilst (ii) reacting as though they are a source of great danger.

    ReplyDelete
  15. (I assume that quote was meant to say something like, 'I take it you want a concept of rationality that may qualify one as irrational without being inconsistent.'?)

    ReplyDelete
  16. Yes, thank you for supplying the missing negation.

    Just to be totally clear, is the irrational thing having emotions that would be appropriate if some false normative claim were true, or having emotions that would be appropriate if some normative claim that you believe is false were true? You said the first thing in response to Mike, and the second in response to me.

    A worry about the first thing: surely having false beliefs is different from having irrational beliefs?

    Worries about the second thing: what if your explicit normative judgment is wrong, like in Bennett's Huck Finn example? Or what about areas where explicit reasoning is generally a less reliable guide to truth than intuition? For example, you might explicitly believe on the basis of good evidence that person X is OK, but find yourself unable to shake the nagging feeling that X is dodgy. These seem like cases where your judgment is appropriate, even though it conflicts with your normative beliefs.

    (I'm full of fundamental disagreements at this point, but they don't belong on this thread, and they'd probably make for a boring conversation, so I'm sticking with your moral realist assumptions.)

    ReplyDelete
  17. Yeah, I've been a little sloppy there. Really I mean something in between belief and truth: our emotions shouldn't conflict with the normative claims that are "warranted", or that we ought to believe.

    There's also some irrationality present in an agent whose emotions and actual beliefs are inconsistent -- but the problem could be the belief, as with Huck Finn. In cases where intuitions are reliable, they should be taken into account as evidence. A good judge of character should (in some cases) believe X is dodgy even though she cannot explicate the grounds for her judgment.

    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.)