Thursday, August 30, 2012

Desiring, Valuing, and Believing Good: Almost the Same Thing

Philosophers typically think of belief as one thing and desire as quite another. Beliefs, for example, represent the world as it is; desires picture the world as we want it to be. Desires are intrinsically motivating; beliefs are maps by which we steer once motivated. Action, it's often thought, requires the copulation of a belief and a desire -- for example the desire to get a cupcake and the belief that the store I'm passing has them at a reasonable price. In generating my action, these attitudes play entirely different roles, sometimes labeled "cognitive" and "conative".

And yet I am struck by how belief and desire can sometimes seem to blur into each other. To believe that it would be good for X to happen is probably not quite the same thing as to desire that X happen, but it takes an unusual psychological wedge to pull them apart. It might not be possible to have an entirely canonical case of the one alongside an entirely canonical lack of the other. And valuing seems a kind of intermediate case: To value one's privacy seems to be almost, on the one hand, to have a kind of belief or set of beliefs about one's privacy but also, on the other hand, to have a certain sort of desire for privacy. Where's the sharp line between cognitive and conative?

I've come to think there is, in fact, no sharp line between the attitude types. This falls out of a general theory of psychological attitudes on which I have been working. (A paper is in progress, but not yet in circulating shape.)

On my view, to believe that something is the case -- for example, to believe that gold is more valuable than molybdenum -- is just to live in a particular way. It is to act and react, and to be disposed to act and react, across a wide variety of hypothetical scenarios, in ways that ordinary people would tend to regard as characteristic of having that belief. So, for example, it is to be willing to say, ceteris paribus (that is, all else being equal or normal, or absent countervailing forces like the intent to deceive), that gold is more valuable than molybdenum. It is to be willing to trade away molybdenum for gold. It is to be disposed to judge others' molybdenum-for-gold trades as wisely done. It is to feel happier upon receiving gold than upon receiving molybdenum, and to judge it reasonable to price the one higher than the other. And so forth. If space aliens were to visit tomorrow and they exhibited this psychological pattern, we would rightly say they believe that gold is more valuable than molybdenum, even though we may know virtually nothing about their internal operations.

To value one thing over another is also, I think, to live in a particular way. In fact, valuing A over B involves almost the same set of characteristic patterns of behavior, subjective experience, and cognition as does believing that A is more valuable than B. The same way of living that makes it true to say of our space aliens that they believe that gold is more valuable than molybdenum also makes it true to say that they value gold over molybdenum.

The same goes for desiring one thing more than another. Almost the same set of dispositions, perhaps with a shift of focus or emphasis, are involved in desiring A over B as in valuing A over B and believing A more valuable than B.

Now perhaps in using the term "desire", we put more emphasis on something like visceral reward or impulse, while "valuing" seems more intellectual or cognitive and "believing valuable" seems more intellectual still; but this is a subtlety. It's a subtlety that can make a difference in an unusual sort of case, where the intellectual and the impulsive pull apart -- where one has an impulse, say, to eat a cupcake but also a sense that it would be bad to do so all things considered. Then we might say, "I want that cupcake, but I don't think it would be good to eat it". But this is not really a canonical case of wanting something. In a way, it seems just as accurate to say that what one really wants is not to eat the cupcake, and what one is fighting is not so much a desire as an impulse.

Shortly after moving into one of my residences I met a nineteen-year-old neighbor. Call him Ethan. In my first conversation with Ethan, it came out (i.) that he had a beautiful, expensive new pickup truck, and (ii.) that he unfortunately had to attend the local community college because he couldn't afford to attend a four-year school. Although I didn't ask Ethan directly whether he thought owning an awesome pickup truck was more important than attending a four-year university, let's suppose that's how he lived his life in general: Ethan's inward and outward actions and reactions -- perhaps not with absolute consistency -- fit the profile of someone who wanted an awesome pickup truck more than he wanted to go to a four-year school, who valued having an awesome pickup truck more than he valued going to a four-year school, and who believed it better or more important to have the one than the other. Approximately the same set of dispositional facts makes each of these psychological attributions true.

Desiring, valuing, believing good, believing valuable -- if to have these attitudes is just a matter of living a certain way, of being disposed to make certain choices, to have certain feelings, to regret some things and celebrate others, etc., then although these attitudes might differ somewhat in flavor and thus sometimes partly diverge, the difference between them is vastly overstated by philosophical talk of the cognitive-conative divide and of the very different psychological roles of belief and desire.


Carl M. said...

It is my understanding that Indian Nyaya epistemology makes this same point: that belief alone can be motivating.

Eric Schwitzgebel said...

Carl, if you have a citation, I'd be interested to see it!

Tad said...

Seems very plausible, but I wonder how much mileage you're getting from a rather heterodox understanding of the propositional attitudes. Of course if PAs are just ways of living, then there won't be sharp boundaries between believing valuable, valuing, and desiring. But if PAs are, like orthodoxy seems to suppose, discrete kinds of mental states that make distinctive contributions to the etiology of behavior, then there are pretty clear distinctions between them, e.g., as you say, their distinctive contributions to practical reasoning, especially the different directions of fit that characterize belief and desire. So, it seems to me, the plausibility of your point here derives almost entirely from the plausibility of your understanding of the PAs, the arguments for which I eagerly await! (I read your paper in the Monist (I think) about this, but I can't wait for a book-length treatment).

Scott Bakker said...

To tie this back to another favourite saw of yours, what role does introspection play in this?

Someone like Carruthers, for instance, would argue that your access to your PAs is as interpretative as your attribution of PAs to others, that you are, in effect, using your 'mindreading system' on yourself.

If this is the case, then it seems plausible to presume that the 'borderline cases' you call attention to are simply instances that reveal the limits of the kind of heuristics we use to attribute PA's to others, that the actual machinery is far more complex (which is almost certainly the case, I think), forcing us to rely on a low-resolution, but robust and useful, information, not only when explaining, predicting and manipulating others, but ourselves as well.

Mark Alfano said...

Interesting framework, Eric. I'm excited to see the paper when it becomes available.

What do you think of the oft-asserted idea that it's somehow easier to have contradictory or at least inconsistent desires than it is to have contradictory or at least inconsistent beliefs. People like Connie Rosati argue that while there may be rational pressure on someone to update their beliefs if they realize that they're inconsistent, they're under no such pressure if they realize that their desires are inconsistent.

If this line of thought is on the right track, it would be a way of pulling apart beliefs and desires. My own view, though, is that there's actually quite a lot of rational pressure to revise desires in light of inconsistency, and that cases that people are tempted to describe as inconsistent desires are actually not. Instead, they're cases where desire is not closed under believed implication. "Get your government hands off my Medicare!" is a classic example. Someone who says this wants the government out of Medicare, does or at least should believe that the government runs Medicare, and yet doesn't want Medicare to cease to exist.

Jeremy Goodman said...

Once one comes to grips with 'creeping minimalism', one sees that there's nothing more to expressivism than the view that to believe that something is good is to be a fan of it, or something like that. And such metasemantic views are just false, since we can believe things are good through Burgean deference about "good" without being at all a fan of the things in question. A better view is that being a fan of something is a *way* to think that it's good. This avoids the deference problem, while capturing the striking metasemantic datum that we are willing to ascribe moral beliefs to agents solely on the basis of their patterns of evaluative attitudes. [End of rant]

Nick said...

The cupcake case. One could easily reframe things so that one is not desiring to P and ~P. It could be that one desires the taste of the cupcake (T), but does not desire the nutritional consequences (N). So they desire T and ~N. This seems a satisfactory account of behavior and it also explains the "inner" conflict. After all, a cupcake, necessarily involves T and N. You cannot have one without the other. I look forward to reading about how your theory is superior to this type of theory.

Also, it seems that believing/desiring/valuing gold more than an alternative entails that when someone asks one if they believe/desire/value gold more than an alternative, one will respond "yes" or "no."

If we go with your interpretation, then one would answer yes because they are aware of their predisposition to treat gold as more valuable. I struggle to imagine myself doing this. I suppose I could imagine myself asking, "Would I rather have gold or an alternative?" This kind of fits your interpretation, but I cannot imagine myself doing this in all cases. It will be tough to get your reader to believe (whatever that means) that when someone is thinking about this type of question, they are really considering their own predispositions to certain behaviors. Personally, I think that interpretations implies that most people have greater self-awareness than they otherwise demonstrate.

I look forward to reading your paper to see how these kinds of cases are framed. Thanks for sharing!

Eric Schwitzgebel said...

Thanks for the comments, folks!

@ Tad: I agree. I regard it as a positive feature of my general approach to the attitudes that it delivers this result, while more orthodox approaches do not.

Eric Schwitzgebel said...

@ Scott: I agree completely with the point in your last paragraph. One advantage of my approach over others, like Carruthers', is that it makes no bones about this, whereas other approaches typically have to characterize the underlying mechanisms as much simpler than they in fact are.

I think Carruthers has an important piece of the story about self-knowledge of attitudes, but I think he overplays his empirical hand when he tries to make it the *whole* story. I discuss this, as well as my own unfortunately convoluted position on self-knowledge of attitudes, in my 2011 paper "Knowing Your Own Beliefs":

Eric Schwitzgebel said...

@ Mark: I think I'm with you on this one, and might go you one farther. The difference is exaggerated, with inconsistency in beliefs more tolerable than is often portrayed and inconsistency in desires less tolerable. That's not to say there's *no* difference in consistency pressure.

Sometimes when we say our desires are inconsistent they're really not. It might seem inconsistent, for example, for me, at noon, to want to work six more hours and to want to be home by 5; but we can render them consistent by maneuvers such as saying that what I really want is that is that there be an extra hour in the day today, or that it be 11:00 rather than noon, or.... (It would help here if we allow desires for impossible worlds.)

Eric Schwitzgebel said...

@ Jeremy: Good thing I'm not an expressivist then!

Eric Schwitzgebel said...

@ Nick: I agree with most of what you say in your first paragraph. I'm not saying that one simultaneously desires P & desires that -P. In fact, that would be straightforwardly impossible on my view. So I'm unsure what you think the source of the conflict is here.

I also agree with your second paragraph, except I think "entails" is too strong. I prefer: "contains as part of the constitutive stereotype (enough of which must, ceteris paribus, be satisfied)". On your last point: I don't think we normally think of ourselves as self-ascribing attitudes based on our knowledge of our dispositions; but I don't think my view entails that we do think of ourselves that way.

This short post opens a web of these types of issues, and obviously I can't address them all in depth here. If you're curious to further explore my position on them, you might check out my "A Phenomenal, Dispositional Account of Belief" (2002) and "Knowing Your Own Beliefs" (2011), both available on my website.

Doug Portmore said...

Hi Eric,

Consider your (A) and my (B):

(A) "To believe that it would be good for X to happen is probably not quite the same thing as to desire that X happen, but it takes an unusual psychological wedge to pull them apart."

(B) To believe that it would be better for X to happen than it would be for Y to happen is probably not quite the same thing as to prefer X's happening to Y's happening, but it takes an unusual psychological wedge to pull them apart.

I'm thinking that if (A) were true, then we should expect (B) to be true as well -- at least, if we think the relationship between good and better is like the relationship between, say, tall and taller, which seems right to me.

But now (B) seems false. To illustrate, let X be the state of affairs where two 8-year-old strangers are saved from drowning, and let Y be the state of affairs where my 8-year-old daughter is save from drowning. I believe that it would be better (other things being equal) for X to happen than for Y to happen, but I don't think that "it takes an unusual psychological wedge" to pull them apart. I prefer Y to X not because of any unusual psychological wedge, but because of the usual psychological preference that people have for the survival of those to whom they have close ties over the production of the greater good.

Arnold Trehub said...

Eric: "Beliefs, for example, represent the world as it is; desires picture the world as we want it to be."

I don't think that beliefs represent the world asit is, Beliefs represent the world as we think it is -- what we sincerely take to be true about our world as it is consciously experienced.

Tony Dardis said...

I like this kind of view. But, sort of following Tad here: part of the point of the cognitive/connative distinction is separability. Ways of living need to be sensitive to fact and opportunity: if I desire that p, my actions with regard to p work out better if I'm responsive to facts relevant to realizing that p. (Eric Saidel has a nice paper in Robert Lurz' Animal Minds anthology on this theme: separability is his mark of the presence of mental representations.) Similarly, it's often useful to accumulate cognitive states even if you don't know what you're going to do with them. And they can be picked off more or less one-by-one by appropriate evidence.

Looking forward to seeing how your're thinking about this!

Anonymous said...

Instinctively i think about this as a question of the definition of "me".

so two parts of 'me' (one impulsive and one non impulsive) might 'want' differnet things, or want them conditionally or want them now but not later.

One might have a more intelectual aspect to it and the other more 'base', and one of those may tend to win in general.

I suggest it would indeed be possible for me to want a cupcake and not want it at the same time, in the same way, simply via running two streams of conciousness at the same time - it is just that for most normal people that would be very hard to maintain unless there was a reasonable difference in type of conciousness.


Unknown said...


Unknown said...

Thanks. I hope to read those papers this weekend.

Unknown said...

Not a philosopher but... Is the punch line that we start w desire then retroactively rationalize and/or enhance the associated belief and/or valuation to justify our conduct and opinions to others?

Eric Schwitzgebel said...

Scott: I definitely think that happens, and that's one reason for the blurring, but I don't think that's the only reason. Desire and belief good might not be so different even to start with.

Eric Schwitzgebel said...

@ Doug: Interesting case. "Better" is multifaceted. If one reads it as "objectively better from an impartial point of view" what you say seems right. But I think there's a less objective dimension of "better" -- probably the more central dimension? -- on which your point carries less force.

Eric Schwitzgebel said...

@ Arnold: Yes, I was speaking elliptically.

Eric Schwitzgebel said...

@ Tony: One problem with being liberal about accumulating lots of conceivably separable nearby mental state is the "number of planets" issue. I believe that the number of planets is less than 9. And I believe that it is less than 10. And etc. We presumably don't want separate storage of an infinite number of beliefs about the number of planets.

Eric Schwitzgebel said...

GNZ: I think I see your point, but my general preference is to go for vagueness and approximate fit and splintered dispositions in such cases rather than distinctly separate selves or fully possessed conflicting attitudes. Part of my reason for favoring the latter is that there are so many *different* ways of splintering that insistence on fault lines that can be cleanly labeled is doomed.

Scott Bakker said...

That was a fantastic paper, Eric. Thanks for the tip. When you say you think Carruthers 'goes too far' I'm guessing that you're referring to the way he engages in the 'name that mechanism game.'

We are in almost total agreement. (The only argument I have against dispositionalisms like yours is their theoretical utility: they are certainly safe from a commitment point of view, but they don't hazard the risks that science requires. But I'm sure you've heard that before.)

I'm wondering though, how far you are willing to go in your criticism of the cognitive/conative divide. Could it be that folk-psychology is as much a culprit as philosophical culture?

One of the things about taking the evolutionary perspective to these questions is the way everything becomes an adaptive, problem-solving artifact - and what distinguishes the cognitive from the conative are things like accessibility and priority. What we're talking about are likely nested 'enabling systems,' some (the 'cognitivish') generally pigging-backing on others (the 'conativish').

I was wondering whether I could tempt you to pull aside the dispositional veil and share your hunches regarding the architecture of the mechanisms responsible.

And more generally, don't you worry that dispositionalism is - or will shortly become - a 'black box' philosophical approach, a prudential 'stand-by' position, no longer necessary now that cognitive neuroscience has developed so many tools?

Eric Schwitzgebel said...

Scott, I do think folk psychology is partly guilty here, too, through its reification of "what we want" and "what we believe", though it's less deeply committed to those reifications than are some orthodox theories of belief and desire. Also, it's not that I think there's *no* difference between believing good and desiring -- just that the difference tends to be overstated as though there were these two vastly different sorts of things, beliefs and desires.

On the mechanisms and neuroscience: I'm inclined to think that things are highly complicated and we haven't progressed very far in understanding the mechanisms underlying the attitudes; it's an almost-unfathomable dynamic-systems swirl. I don't rule out the possibility of trading up from folk psychology to something more scientifically rigorous in the future. But I think it's a mistaken to suppose that we have gone very far down that road yet.

Arnold Trehub said...

Hi Eric,

You wrote: "I'm inclined to think that things are highly complicated and we haven't progressed very far in understanding the mechanisms underlying the attitudes; it's an almost-unfathomable dynamic-systems swirl."

I agree that there is much yet to do in understanding brain mechanisms underlying attitudes, but a start has been made on the problem. You might take a look at "Set Point and Motive: The Formation and Resolution of Goals", in *The Cognitive Brain*, here:

Eric Schwitzgebel said...

Yes, hence "almost"!