Inspired by Uriah’s post below, I invent a new kind of zombie (a hedonic-zombie), draw a distinction between strong and weak internalism about the good life, and argue that Nagelian thought experiments involving hedonic-zombies support weak, not strong, internatlism. First, some definitions:
- Hedonic-zombies (or h-zombies) are exactly like us in all physical respects and mostly like us in terms of their conscious states; however, unlike us, they are incapable of experiencing pleasure or pain, or any of their cognates, such as joy and anguish, satisfaction and frustration, etc. Although, like us, h-zombies have desires and aspirations about how their lives might go, they are never displeased when things don’t go as they wanted them to and are never pleased when things do go as they wanted them to. They take no joy in anything, not in getting what they want, not in watching a beautiful sunset, not in discovering some new philosophical insight, not even in making love.
- Strong Internalism about the Good Life: Two lives differ in their welfare value if and only if they differ in their psychological features.
- Weak Internalism about the Good Life: Two lives differ in their welfare value if and only if they differ in their pleasantness, but how pleasant a life is needn’t depend solely on the psychological facts about that life.
Now it seems to me that the lives of all h-zombies will have the same welfare value: namely, zero welfare value. So although two h-zombies may desire to be loved and respected and whereas one, Hank, is both and the other, Herald, is neither, it seems correct to say that Herald is no worse off (or better off) than Hank is. But this intuition about the welfare value of the lives of h-zombies doesn’t support strong internalism; it only supports weak internalism. And, as Fred Feldman has ably shown, weak internalism is consistent with the view that Tim, who is pleased that he seems to be loved and respected and is loved and respected, has a more pleasant life than Tom, who is pleased that he seems to be loved and respected but isn’t loved and respected. That is, one can accept truth-adjusted intrinsic attitudinal hedonism (TAIAH), according to which the value of an episode of attitudinal pleasure depends not only on its intensity and duration, but also on the truth value of its object. Other things being equal, an episode of attitudinal pleasure where one is pleased that P and P is true is more pleasant and, consequently, of more value than an episode of attitudinal pleasure where one is pleased that P and P is false.
In conclusion, I think that although zombie arguments might be usefully employed against theories that reject weak internalism, such as, the desire-fulfillment theory, they cannot be used to refute weak internalism.
Doug,
I find it very odd to think that one episode of attitudinal pleasure could be “more pleasant” than another due to non-psychological facts. Suppose that A and B are psychological duplicates of one another, and each is pleased that her husband loves her. But A’s husband really does love her, while B is sadly deceived. I’m happy to say both that A’s life is better than B’s (in this respect), and that A’s pleasure is worth more; but if they are psychologically identical, it seems false to say that the situation of one is “more pleasant” than that of the other.
So I’d prefer to see the idea expressed by weak internalism expressed as something like this:
Weak Internalism About the Good Life (Troy’s Version): Two lives differ in their welfare value if and only if they differ in the total quality of the pleasures they contain, but the total quality of the pleasures contained in a life needn’t depend solely on the psychological facts about that life.
I’m not asserting WIAGL(TV), but I think that it is at least a viable candidate, since it is consistent with the fact (as I take it) that pleasantness is a wholly psychological property.
Doug,
I don’t think that’s Feldman’s view. The view is not that true pleasures are more pleasant than false pleasures. It’s that they are intrinsically better (other things equal). He might accept Troy’s version of weak internalism, if by ‘quality’ Troy means ‘value’ (and if something appropriate about pains were put in there).
Troy,
Thanks. You’re right: it does sound odd to say that one episode of attitudinal pleasure could be “more pleasant” than another due to non-psychological facts. So I take it back. In any case, the core idea behind weak internalism is that one can accept that the bearers of prudential/welfare value are pleasures and pains and so deny that h-zombies have the necessary bearers of prudential value, nevertheless one needn’t hold, as the strong internalist does, that the prudential value of those pleasures and pains supervene solely on their phenomenological qualities. Their prudential value can, for instance, depend on the truth value of the objects of those attitudinal pains and pleasures. So I would be happy to accept your friendly amendment to WIAGL or at least something close to it.
I’m sympathetic to the idea that h-zombies don’t have a welfare but I want to reject strong internalism.
Ben,
Yeah, you’re right about Feldman. My mistake.
By the way, my colleague Peter de Marneffe has an interesting article entitled “An Objection to Attitudinal Hedonism” in Philosophical Studies 115 (2003): 197-200. He has the opposite intuition about h-zombies (or, in his case, a Spock who is more like an h-zombie than the Spock of Star Trek). He argues that because Spock has rational aims (e.g., becoming a starship officer), his life goes better for him if those aims are achieved whether he enjoys their achievement or not. What do others think?
Could someone say a bit more to motivate the claim that a ‘true’ pleasure is intrinsically more valuable than a ‘false’ pleasure? I sort of see the intuition, but I’m still finding it a bit odd. Using Troy’s example in his comment above, I can see why one would hold that A’s life is better than that of B, insofar as A’s husband does in fact love her (perhaps A is also better off in having true beliefs, wher B has false ones). But I’m puzzled by the additional claim that “A’s pleasure is worth more” (or is of a higher quality).
As people noted above, it’s not that A’s pleasure is more pleasant. What else goes into determining the quality of a pleasure, then? [Would taking pleasure in the suffering of others be a lower quality pleasure? Would pleasures involving use of our rational faculties be higher quality pleasures? I guess I also find these sorts of claims odd.]
Why not simply hold that if we see B’s life as worse, it is insofar as her desires are not actually satisfied, and (perhaps) that she has false beliefs? Why think that the quality of her pleasure itself would be affected?
One understanding of pleasure and pain are states that are intrinsically liked or disliked for what they feel like. Another understanding is that pleasure is a particular flavor of sensation that has a distinctive phenomenology. I am thinking you are assuming the latter understanding. However, so understood, I think shielding such flavors of sensation does not have the intuitive bite (against, for example, desire accounts) that you think it does. For now the thought is just this: suppose what I want goes one way and any particular flavor of sensation goes another way–which way determines the extent of my benefit. And this is a question that has overwhelmingly been answered in favor of the former.
Jason,
I think that the motivation for the sorts of claims that Feldman makes is to take what’s best from hedonism (perhaps, the idea that the good life is the pleasant life) while leaving behind the standard counter-intuitive implications that simpler versions of hedonism have such as the implication that Tom’s life is no worse than Tim’s. You also ask,
“Would taking pleasure in the suffering of others be a lower quality pleasure? Would pleasures involving use of our rational faculties be higher quality pleasures?”
Yes and Yes. As Feldman argues, there are versions of hedonism that have these (intuitive) implications.
David,
I don’t think that I’m presupposing that understanding of pleasure at all. Indeed, one of the reasons for opting for attitudinal hedonism is skepticism about whether all pleasures have any distinctive phenomenological feel/flavor that they share in common. In any case, could you say a bit more about why you don’t think that my h-zombie argument has the intuitive bite that I think it does against the desire-fulfillment theorist. Do you not share my intuition that h-zombies are incapable of living a life of any welfare value whatsoever or do you deny my contention that the desire-fulfillment theorist is committed to the view that h-zombies are capable of living a life with either positive or negative welfare value?
H-zombies sound like what Neil Sinhabubu calls Neutrals.
Hi Doug,
I’m not sure I think that h-zombies are conceptually possible. I don’t really have a theory of desire worked out in my head, but I guess I think that the concept of desire is analytically connected to the concept of pleasure in some way; maybe, desires are states that, in conjunction with certain beliefs, cause (or tend to cause) pleausure?
I’m sure that’s hopeless naive. But I would like to think more about the relationship between desire and pleasure before I draw any conclusions about welfare from considerations about h-zombies.
That said, I do accept that unless something has conscious states, it can’t have a welfare. I also accept the claim that a person’s welfare is n’t solely determined by his conscious states in some strong way.
Hi Doug,
Thanks for your response! I guess the move to talking of the quality of pleasures still strikes me as a bit ad hoc. Uncharitably, the proposal seems to be something like this “It sure seems like Tom’s life is worse than Tim’s, given that Tom’s wife really doesn’t love him, etc. But how to capture this with a form of hedonism? Their lives are equally pleasant, after all. I know – we’ll tack on the idea of the ‘quality’ of pleasures, and say that pleasures arising out of false beliefs are thereby of reduced ‘quality’ (or value, more broadly). Problem solved.”
I’m still not sure why we should think that the badness of false beliefs (or of desires that are not actually satisfied) should then ‘infect’ the associated pleasures (or pains). Perhaps this is the wrong way of thinking about how the quality of a pleasure is determined. But then what is occurring? (if not ‘infection’) – why would taking pleasure in the suffering of others, or pleasures that are a result of false beliefs about desire satisfaction, come to have reduced value or quality? Maybe my broader question is – what goes into establishing the quality or value of a pleasure (beyond its pleasantness)?
Doug,
Well it seems to me you need to deny the alternative story according to which pleasure is had when an agent gets a feeling that she likes (where that is to be taken as in the desire-based sense). You need to rule this out because on this conception of pleasure, don’t you, in order to get your thought experiement going. But then I don’t know what understanding of pleasure you have other than the flavor of sensation or family of flavors of sensations view.
Jason,
Yeah, I can see why it seems ad hoc to you, but, these days, I have less of a grip on why such putatively ad hoc moves are supposed to be bad. As I see it, we’re just trying to gain reflective equilibrium on our intuitions about welfare, and, unfortunately, we have two seemingly conflicting intuitions: one the intuition that lies at the core of hedonism and two our intuition that Tom’s life is worse than Tim’s. But Feldman argues that there is a way to accommodate both intuitions by adopting intrinsic attitudinal hedonism. This seems like a great move, ad hoc or not. If we’re just looking for reflective equilibrium here, what’s wrong with making such a move to avoid the counter-intuitive implications that plague more orthodox hedonism?
David,
Why do I need to rule out this alternative story (the desire-based story) to get the thought experiment going?
It seems to me that all I need to get the thought experiment going is the idea that there could be beings who have many of the propositional attitudes that we have (e.g., “fear that,” “believe that,” “desire that,” etc.) but just not the following one: “pleased that.” It seems to me that I could even accept a desire-based analysis of sensory pleasure (remember that on my view what makes a life go better is attitudinal pleasure): x is a sensory pleasure just in case x is an experience that the subject desires that she possesses or an experience that the subject desires that she will continue to have. H-zombies will have such sensory “pleasures,” but they won’t take any pleasure in having them. For instance, an h-zombie may want her feeling of exhilaration to persist, but she won’t be pleased that it persists. She won’t enjoy the experience. This seems very sad for her. Indeed, it seems right to conclude that such a life would be a life of no welfare value whatsoever.
Well tell me about attitudinal pleasure. It sounds like it can be ruled out by ruling out certain kinds of phenomenology–that is ruling out certain kinds of flavors of sensation. You, I was thinking, need to rule out the desire-based understanding of pleasure (now attitudinal pleasure) because otherwise your allowing that the agent has the relevant desires would just conceed that she has the relevant kind of pleasure. You want an agent with all the desires you like, but none of X. Then we are to feel the force of lacking X. But clearly if one had a desire based account of X, the thought experiment would not work. Thus I think you need a non-desire based account of X and I am not sure what it would be if it were not a flavor(s) of sensation account of pleasure.
David,
I’ll try, although my recollections of Feldman are a bit sketchy. In any case, to quote Feldman: “A person takes attitudinal pleasure in some state of affairs if he enjoys it, is pleased about it, is glad that it is happening, is delighted by it.” Now it may be that these attitudes all share some phenomenological quality in common in the way that the attitudes “hope that,” “wish that,” and “desire that” share some phenomenological quality in common. But to accept this is not to assert that all the sensations/experiences that we tend to enjoy (e.g., the sensations of soaking in a hot bath, having an orgasm, eating chocolate ice cream, etc.) share some phenomenological quality in common. Furthermore, I can reject the view that all the sensations that we tend to enjoy share some phenomenological quality in common without saying that what makes the having of one sensation but not another of prudential value is that I desire the one but not the other, for there is another propositional attitude, distinct from the attitude of “desiring that,” which I can appeal to. Indeed, I can claim that what makes the having of one sensation but not another of prudential value is that I’m enjoying the one but not the other, and I can enjoy a sensation that I desire not to experience and I can fail to enjoy a sensation that I desire to experience. So if I’m getting what you’re saying (and I’m not at all sure), you want to say that I have to understand pleasures either as experiences that share some phenomenological quality in common or as experiences that we like or desire. And my response is to say that I’m not understanding pleasure as either. Attitudinal pleasures aren’t sensations/experiences; they’re propositional attitudes.
Doug:
I like this way of organizing the logical territory. It suggests that the strong internalist must see through a two-step argument. The first step is this:
1. H-zombies are metaphysically possible;
2. If weak internalism is false, then h-zombies are metaphysically impossible; therefore,
3. Weak internalism is true.
The second step requires an argument that would take you from weak to strong internalism. I think what the strong internalist needs here is some sort of internalism about pleasure, or more specifically the following thesis:
(*) How pleasant a life supervenes on how the non-relational properties of the person whose life it is.
Now the argument would go:
1. Weak internalism is true;
2. (*); therefore,
3. Strong internalism is true.
The question – or a question – is whether (*) is plausible. I think it is, but it seems some of us think it isn’t.
I am afraid I still don’t understand what attitudinal pleasures are.
But I can manage a more articulate further question than that. You say that “I can enjoy a sensation that I desire not to experience”. I can think of lots of ways this could be true. I could enjoy that I am getting a certain unpleasant experience because it signals that I win a million dollars or something. But can I intrinsically and informedly prefer to feel X rather than Y at time t on the grounds of what they each feel like yet still enjoy Y more than X?
Small point. Do you really want ‘Strong Internalism’ to be a biconditional? As written, it implies that any two lives that have the same welfare value are psychologically identical. That would seem to make equality in welfare between two people very rare.
David,
Suppose that you had asked instead: “Is there some metaphysically possible being that can intrinsically and informedly prefer to feel X rather than Y at time t on the grounds of what they each feels like yet not enjoy X more than Y?” I would think that the answer to this question is: “yes.” It seems to me that h-zombies are metaphysically possible and I see no reason why one could’t desire the sensation of redness rather than blueness at time t because the former is a “warmer” color sensation and yet this h-zombie would not enjoy the one sensation more than the other, for she wouldn’t enjoy either. I don’t know whether humans ever have such preferences, but I don’t see why a being couldn’t have such a preference.
Here’s another thought. Suppose we accept the view that pleasures are just experiences that one intrinsically and informedly desires for the sake of how they feel. In that case, can’t we imagine a being that desires to be loved and respected but is, nevertheless, incapable of experiencing pleasure because she is incapable of liking/desiring their own subjective experiences. These h-zombies would have desires but would be incapable of desiring/liking their own subjective experiences. Such h-zombies would have a welfare on the desire-fulfillment theory, and yet this seems counter-intuitive: even when they believe that their desires are fulfilled and consequently feel loved and respected they take no pleasure in these feelings.
Campbell: My bad: it should read “only if,” not “if and only if.”
Here’s a further off-the-cuff thought on the metaphysical possibility of zombies with desires but no qualia, beings that are, therefore, incapable experiencing pleasure and pain. Are people under general anesthetic temporarily zombies of this type? They have desires, right? We wouldn’t want to say that one has a desire that P only if one has an occurrent mental state that consists in yearning that P. But while under general anesthetic, people are incapable of experiencing pleasure/joy or pain/anguish. Isn’t it metaphysically possible to have a being that desires to be loved and respected in whatever sense that I desire these things while under general anesthetic, but who wasn’t ever and never will be conscious? They would be qualia-free in just the way that people under general-anesthetic are qualia-free. So maybe I can now understand Uriah’s claim that zombies can have desires. Or maybe I can’t. May one must be conscious at some point in order to be said to have a desire. I’m not sure. Thoughts?
David,
I realize that I didn’t really answer your question. Let me just retract my earlier statement: “I can enjoy a sensation that I desire not to experience.” What’s relevant to the issue at hand is whether a being can intrinsically and informedly desire an experience that she doesn’t enjoy. I think that she can. Do you see any reason to think that she can’t?
Doug,
You wrote:
Here’s another thought. Suppose we accept the view that pleasures are just experiences that one intrinsically and informedly desires for the sake of how they feel. In that case, can’t we imagine a being that desires to be loved and respected but is, nevertheless, incapable of experiencing pleasure because she is incapable of liking/desiring their own subjective experiences. These h-zombies would have desires but would be incapable of desiring/liking their own subjective experiences. Such h-zombies would have a welfare on the desire-fulfillment theory, and yet this seems counter-intuitive: even when they believe that their desires are fulfilled and consequently feel loved and respected they take no pleasure in these feelings.
I think I disagree with that. First, a desire theoriet is free to pick and choose the kinds of desires that they think are relevant to well-being, so I suppose a desire theorist could get the result that this person lacks the capacity to be benefited or harmed if that was thought to be the right thing to think. Such a desire theorist would focus only on intrinsic desires for various kinds of phenomenology.
But I don’t yet think it is the right thing to think because I think other things matter to well-being besides just how good things feel as you are experiencing them. I suspect you might agree with that, but claim that the capacity to like phenomenology is necessary to the value of the other things. I don’t know what I think about that yet.
I agree that the relevant question is whether there are metaphysically possible creatures that can X but not Y, not whether humans in fact X without Y-ing. I think to better understand what I would want to say about the possibility of intrinsically desiring in the right way something one does not enjoy, I need to feel that I understand better the notion of enjoyment. I feel that I understand the desire-based understanding of enjoyment you allowed for the sake of argument above. At least if we had that understanding in place, I would feel further along in addressing the question.
For the record, I think Scanlon is a bit frustrating in just this neck of the woods. He wants to say that desires never give reasons, but that we have reason to eat a kind of ice cream we “enjoy” rather than another flavor. And then I think he is quite cagey about how to understand “enjoyment”. His latest, as I understand it, is that desires track such enjoyment without making it the case that one enjoys something.
David,
Okay, you’re right that the desire-fulfillment theorist could tag on some sort of experience requirement and thus avoid this sort of objection. My argument only concerns versions of the desire-fulfillment theory that claim that Tom’s life is worse than Tim’s because, whereas Tim’s desire to be loved and respected is fulfilled, Tom’s desire to be loved and respected is unfulfilled. And I agree that “other things matter to well-being besides just how good things feel as you are experiencing them.” This is just to say that I’m sympathetic to weak internalism as opposed to strong internalism.
I currently lean toward the view that the capacity to be pleased that P (or “enjoy the fact that P” or “like the fact that P”) is a necessary condition for having a welfare, although the relevant proposition, P, doesn’t have to pertain to some qualia. I can be pleased that I’m experiencing a tingly sensation, but I can also be pleased that I have a tenure-track job. So I think that I would deny that “the capacity to like phenomenology is necessary to the value of the other things.” The object of the liking (or the enjoyment) doesn’t have to be the having of some phenomenological experience.
I think something along the view you attribute to Scanlon might be right: “desires track such enjoyment without making it the case that one enjoys something.” That is, it certainly seems true of human beings that if we enjoy a certain kind of experience and know that we do, we will (at least, normally) intrinsically desire it. But I question whether every experience that is intrinsically desired must necessarily be enjoyed. Thus intrinsically desiring that some experience persists doesn’t seem sufficient for enjoying it. If this is right, then there can be h-zombies who intrinsically desire certain experiences but who don’t enjoy having them. And these h-zombies may also desire to be loved and respected when they’re not, but I have hard time seeing how this or anything else could be bad for them if they’re incapable of enjoyment.
Doug,
Given that Scanlon’s latest (that I know of) might appeal to you, I thought I would provide the reference. Social Theory and Practice, April 2002, p. 338-40. I have a paper that, among other things, argues against this view of Scanlon that is recently out at: “Pain for Objectivists: The Case of Matters of Mere Taste,” Ethical Theory and Moral Practice, Vol. 8, No. 4 (August 2005), 437-57.
David,
Thanks for the references. As a matter of fact, I’d been reading your paper just a couple weeks ago but had gotten only a little less than half way through it when I got sidetracked with a bunch of personal stuff and now I have a huge stack of papers to grade. I do like what I’ve read so far. And although I’m attracted to the sort of strong objectivism that Scanlon and Parfit endorse and that you argue against, I must admit that your argument regarding matters of mere taste is quite compelling. I’m trying to come up with some objection to it, because I have a feeling that if give up ground here, you’ll push me all the way toward subjectivism. Anyway, I’ll come back to it soon when I have the time to give it the attention it deserves. And who knows; maybe it will give me something to blog about.
Cool!