THOUGHTS ON EMOTIONS: I’ve just finished re-reading the (long) first chapter of Martha Nussbaum’s book, “Upheavals of Thought.” In that chapter, and in the rest of the book, she argues for an account of what she calls “emotions as judgments of value.” Now, I take this thesis in its general form to be true, even to be obviously true: there’s a cognitive component to our emotions, which reflects our evaluations of objects in the world. The thing I’m curious about is whether the cognitive component actually exhausts what emotions are, which is what Nussbaum suggests. On its face, this thesis seems extreme – certainly emotions without feelings aren’t really emotions at all; we’ve left something important out if we say that. But Nussbaum gives a very complex account of the exact cognitive content of the emotions (how they involve the imagination as well as propositions about the objects, etc.) which, in its sophistication shows how her thesis might very likely be true. I have two worries, however.

First, we might run a thought experiment against Nussbaum, similar to the one David Chalmers runs against Daniel Dennett. We might imagine, that is, that there could be something like “emotional zombies” who had all the cognitive content related to a particular emotion, say love or grief or fear, but who nonetheless lacked affect, and so did not really feel emotions at all. (Chalmers' thought experiment is that there could be zombies who we like us in all physical and behavioral respects, but who nonetheless were not conscious).

Nussbaum, on my reading of her, would be committed to saying that the zombies in fact did have the emotions (as Dennett is committed to saying that the zombies are conscious). I find this to be implausible, because it seems to me that there could be emotional zombies who were cognitively identical to a person with an emotion, but would still not feel anything, making him unlike the person who did have the feeling (in addition to the relevant judgments).

Of course, this is only my intuition, and the zombie thought experiment only an intuition pump. Nonetheless, it might encourage us to give something like “affect” a place in our account of the emotions, by which I mean an independent place – independent of the cognitive content of any particular emotion. We might be moved to say, as Nussbaum records Michael Stocker as saying, that “affect” is a primitive concept, unanalyzable in terms of cognitive content, but which goes alongside it when we explain what emotions are. I am wary of such arguments, and Nussbaum is as well. But I think without positing “affect” as a primitive, we risk losing an adequate description of the phenomena. This doesn’t mean, necessarily, that we can say nothing about affect, only that it resists a reduction to the cognitive. To further explore affect, we would have to proceed, I think, by contrasting it with belief (which Nussbaum also does, though in a way unfavorable to treating emotion and belief as distinct).

Second, while Nussbaum’s account of the emotions might work well with some types of therapy, i.e., psychoanalytic or cognitive-behavioral, I wonder how it would handle psychopharmacology. The thing about traditional therapies is that they do seem to work with a patient’s beliefs. So, on a crudely Freudian picture, you go back and find out what past events your present day emotions are linked to, and you work through them. On the cognitive behavioral account, you work to identify your negative and pessimistic thoughts, and to combat them by looking at them more realistically. But what about taking a Prozac, and having this change your emotional outlook? Doesn’t the fact that one might change one’s mood by taking a pill every day say something about how emotions work, that they might not solely be a cognitive affair?

- posted by Chad @ Saturday, August 30, 2003