Tuesday, August 26, 2008

Inverting Emotional Qualia

Conceivably, some aspects of our phenomenology (e.g. colour) could flip around whilst everything else in our minds remains the same. Could this happen with emotions? Might I have an emotionally inverted counterpart, just like me except that fear to him feels like anger to me, and vice versa?

I suspect not. For one thing, I'm inclined to identify emotions by their phenomenal feel, so the above scenario should really be described as his feelings of anger playing the functional role of fear, etc. More importantly, though, this scenario requires that our emotions be wholly independent of the rest of our mental economy (thoughts, judgments), which is arguably incoherent. Emotions are cognitive: part of the what you experience in your anger is a judgment (e.g.) that you have been wronged.

One might respond by trying to break down these components of the emotion, the judgment and some leftover phenomenal feel. But I don't think they are so extricable (compare: what is left over in scarlet once you take away the redness?). An essential part of the overall feel comes from the judgment, so you can't subtract that without significantly altering the phenomenology.

Relatedly, I've been puzzling over a passage from Justin Oakley's Morality and the Emotions (p.19, lightly edited):
Benevolence and gratitude may both [exhaustively?] involve feelings of affection and warmth... But if one emotion can feel like another, different emotion, then the distinctions between emotions cannot be drawn in terms of feelings.

Oakley must be using 'feel' in a very narrow sense, because it seems clear to me that the different emotions have a very distinct phenomenology. But if we restrict our attention to pre-cognitive or uninterpreted phenomenal aspects, perhaps he is right. This is the sense in which duck/rabbit looks the same to you whether you see it as a duck or as a rabbit. But in a broader (and more natural) sense, they obviously look quite different, for they are readily distinguishable experiences. The lesson from duck/rabbit is that interpretation makes a difference to phenomenology (in the broadest sense). And I think the same sort of thing is going on in case of emotions and their cognitive components.


  1. If feelings of various types are emergent phenomena based upon some underlaying 'stuff' then how could an emotion flip without some serious reworking of the brain? It seems akin to Nagel's question of what it is to be a bat.

    I guess in theory it could happen. Although I think we can say that only due to the huge amount of ignorance we have regarding the workings of the brain and their relation to the phenomena we experience. For all we know it may be impossible physically.

  2. I meant to be arguing that it's conceptually and metaphysically impossible -- a stronger claim than mere physical impossibility.

  3. Even if you can't separate the phenomenal feel of the judgement from that of the rest of the emotion without changing what the emotion is, it seems to me that it's still possible (conceptually) to separate the phenomenal and functional aspects of the judgement such that you could feel as if you were wronged but behave as if you weren't.

    BTW, I think this is the first time I've commented here (though I have been reading for about a year now) so it's a good time to say that I think you've got a great blog here.

  4. Hi Tim - that seems right, at least to a degree. I guess it depends how broadly we delimit the 'functional'. I would want to include not only behavioural outputs, but also internal patterns of thought. For example, we may think that part of what it is to have an emotion (rather than a mere sensation or fleeting thought) is for it to 'tint' our perceptions and patterns of thought -- when we are angry we will be more disposed to see flaws in others, etc.

    (P.S. Thanks!)

  5. A few thoughts.

    First I think it might be helpful to be a bit more explicit about the type of possibility at issue here. In particular people might have very different answers if this is phrased in terms of logical possibility (which I understand to mean something like not false by virtue of the meaning of the terms) versus physical possibility. For the rest of this comment I'll speak to logical possibility.

    To answer a slightly different question I think it's undeniably logically possible for me to be in the functional role regarding externally observable behaviors that now accompanies anger but be in the phenomenal state that accompanies the emotion we describe as fear. Why? Simply because there is no way to logically infer phenomenal content from physical behavior.

    To make the argument more precise I would claim that if I define L1 to be a language with only phenomenal descriptions (there is an experience that feels like such and such) and L2 to be a language that includes only purely physical descriptions (there are atoms of carbon at locations x, y, z) then there are no nontrivial logically necessary relations between the first and second languages. My claim above follows pretty easily from this.

    I would then argue that this is the only interesting question at hand. One can debate about the proper definition of emotions and whether the english word emotion refers only to things that are definable in L1 above but that's kinda a boring question of happenstance.

    Perhaps instead you meant to ask about whether our emotions could be switched without changing other kinds of mental content like judgements. While I agree that this is impossible once again it's mostly an uninteresting question of whether what we call a judgement can be defined in L1 or not.

  6. "Perhaps instead you meant to ask about whether our emotions could be switched without changing other kinds of mental content like judgements."

    Yup, that's what I had in mind. Does this really come down to the question whether judgments are purely phenomenal? This isn't obvious to me. (Prima facie, I'm inclined to think that they aren't purely phenomenal, though their meaning arguably depends upon a phenomenal aspect, at least. Is this incompatible with the position taken in my main post?)

    I guess I'm also skeptical of the claim that this question about the nature of judgments is merely terminological (or otherwise 'uninteresting'). For example, we may think that we have a firm grasp of what 'judgment' means, in terms of the theoretical role it plays in our mental economy. It could then be a substantive question whether phenomenal properties are necessary and/or sufficient to fill this role.


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.)