A common picture of value determination with respect to pains, pleasures and other such mental states seems to be the following: the intrinsic value of such states is determined by their phenomenological character. Thus, some people hold that all pleasures are intrinsically good, and all pains intrinsically bad. (Call this the straightforward theory.) Of course, straightforward theorists can allow that some pleasures are extrinsically bad, and some pains extrinsically good. The enjoyment of another’s suffering can lead the malicious to harm others in the course of seeking out such pleasure; the pain of punishment, on the other hand, may prompt the wicked to reform. Straightforward theorists will insist, though, that the phenomenological character of the mental state is the fundamental, if not the only, determinant of its intrinsic value.
Certain sorts of ethical baggage tend to come along with such views. For instance, many people think that since pain is always intrinsically bad, there is always at least some moral reason to eliminate it. Similarly, some people think that there is always at least some moral reason to promote or increase any particular (potential or actual) pleasure.
Some people accept a slightly more complex axiological theory than this, allowing that at least some pleasures (pleasure in another’s suffering, for instance) might be intrinsically bad. A few are even willing to say that some pains (like the deserved punishments of the wicked) are intrinsically good, though my sense is that it is far more common to hold that these pains are bad in themselves, but extrinsically justified.
I think that the proponents of more complex theories are moving in the right direction, but that we must go still further. We must recognize that there is a greater variety of counter-examples to the straightforward theory than is commonly appreciated (I’ll describe a couple in a moment), and that the landscape of value is thus significantly more complex than has been realized. And more fundamentally, we must reject the idea that the phenomenological character of a mental state is the primary determinant of its intrinsic value.
The theory I’d like to propose to replace the straightforward theory is the following. The primary determinant of the intrinsic value of a mental state is the question of its appropriateness. That is, a mental state is justified, and intrinsically good, if it is appropriate; it is unjustified, and intrinsically bad, if it is inappropriate. Where a state is neither appropriate nor inappropriate, its phenomenological character may then determine its intrinsic value. Thus, a pain that is neither appropriate nor inappropriate, but simply the brute result of a causal mechanism in the world, can be considered bad because it is unpleasant to experience, but only after the prior question of appropriateness has been settled.
There are many varieties of (in)appropriateness. Let me mention just a couple of examples to help bring out the flavor of the theory. As mentioned above, the deserved sufferings of the wicked are sometimes allowed to be intrinsically good. The most straightforward explanation of such a view, I think, is that such suffering is appropriate. But we need not concentrate on punishment, or the wicked, to find such an example. Suppose Gordon’s friend has died, and Gordon is grieving. Grief is clearly a species of suffering, as anyone who has experienced it will know. But, while it is surely unpleasant, it is not clear that it is intrinsically bad. The world is not a worse place for its happening (though it is, of course, a worse place for the event which triggered the belief.) And the idea that there is (at least some) moral reason to eliminate this suffering seems implausible. (Imagine I had a pill that would wipe out Gordon’s grief altogether. Would I be morally obliged to offer it to him? Would there be any moral reason to do so? It seems not.) Grief, then, is a kind of suffering that, where appropriate, is both justified and intrinsically good precisely because it is appropriate. (Inappropriate grief, of course, is neither justified nor intrinsically good.)
One other example. As mentioned above, it is sometimes allowed that pleasure in another’s suffering is intrinsically bad; it is, after all, malicious. However, we can imagine situations in which this is not the case. Suppose that Anne is suffering from grief as a result of her father’s death. Jill, Anne’s mother, who had worried that Anne was so self-centered that she literally did not care whether her father lived or died, might be glad to see that Anne is in fact grieving. The pleasure that Jill takes in Anne’s response is not in any way malicious. In fact it is quite appropriate, given that Anne’s response is itself an appropriate response. So we should have no moral qualms about the existence of Jill’s pleasure, any more than we should have moral qualms about the existence of Anne’s pain. What matters here, in terms of intrinsic value, is not the phenomenological character of the mental states but rather their appropriateness as responses to events in the world. Moreover, I claim, this is at least implicitly true with respect to any mental state: where such a state’s phenomenological character is relevant, it is only as a secondary matter, and only because the state is neither appropriate nor inappropriate. The latter sort of consideration, where present, always trumps.
Finally, I want to mention one advantage of such a theory, which is that it might help contribute toward a unified theory of intrinsic value with respect to mental states and actions. Actions, obviously, do not have a phenomenological character the way mental states like pains and pleasures do; so if we stick to the straightforward theory with respect to the latter, we will need a different sort of account for the former. However, actions, like these mental states, may be appropriate or inappropriate. This seems to me somewhat promising – though this suggestion, I should say, is quite off the cuff. But I’d like to know what people think of it.
Troy,
That’s an interesting theory, though I think that it may serve best as an explanation of some of people’s erroneous intuitions about value. I may say more about that later (perhaps on my blog). For now, I’d like to raise some examples where the Appropriateness Theory (AT) does not seem to correspond with our intuitions about intrinsic value.
1. Scottie watched a mindnumbingly boring video, but, inexplicably, found it hilarious and enjoyed the experience very much. This inappropriate mental state seems intrinsically valuable, as I would recommend the tedious video to someone else if I thought that they would have the same experience.
2. Troy McClure loves fish, with an erotic kind of love. Fish are an inappropriate target of love, and his love an inappropriate kind of feeling to have towards fish, but if that’s how he feels then his relationship with fish is valuable to him, and, I would say, intrinsically valuable. Something similar could be said for many other instances of nontraditional sexuality.
3. Hansel and Gretel were in danger, great danger, as the evil woman attempted to cook them alive. Hansel felt fear and panic, as was appropriate, while Gretel felt a completely inappropriate calm and lightheartedness. AT would hold that Gretel’s state of mind was intrinsically bad while Hansel’s was not, yet the opposite seems to be the case.
One other question – it’s not clear to me what you mean by opposing appropriateness considerations to causality with the phrase “pain that is neither appropriate nor inappropriate, but simply the brute result of a causal mechanism in the world…” Perhaps, a clearer definition of appropriateness could make some of these problematic examples go away.
There might be something to the idea, but there are some big problems to be avoided. Suppose I stub my toe in the presence of a million people, all of whom feel appropriate pity for me. Then you’ve got a million intrinsic goods, and one intrinsic bad, which evidently would make the whole situation really good overall. That can’t be right. (I think I remember a conference where Blackburn gave an example like this as a counterexample to Hurka’s view, which is a lot like yours. Hurka probably discusses this somewhere.) Maybe that situation is intrinsically better than one where the million people get *pleasure* out of my pain, because malicious pleasure is intrinsically bad. (Though I’m not sure about that.) But that’s consistent with the pity not being intrinsically good.
Troy,
I think you provide an interesting alternative to the straightforward axiological view here. Let me do two things: one is to ask a question and the other is to play the devil’s advocate.
First the question: you are presenting this as a theory of value. However, as Fred Feldman and L. W. Sumner urge, there are different sorts of value. The kind of value that has preoccupied utilitarians (and other consequentialists) has been what we may call “welfare value”. So my question to you is this: do you wish for this theory to be taken as an account of welfare (i.e. of what is intrinsically good for me, from my perspective; as that which will make my life go best for me), or do you wish for it to be seen as a much more general theory of intrinsic value (perhaps you reject the idea that there are, in fact, several different kinds of intrinsic value)?
Now to play the devil’s advocate. Let’s suppose that you are understanding your theory to provide a theory of welfare value. The supporter of the straightforward view seems to have an easy response to some of your examples. Take Gordon’s grief. Your claim is that Gordon’s grieving is a good thing because it is appropriate. The straightforward theorist will claim that the grief is not intrinsically valuable for Gordon in any sense because it feels so God awful. It may be extrinsically valuable for him since it may allow Gordon to grow as a person and come to have better experiences in the future (or something along those lines).
I say that I am playing the devil’s advocate here because I do not particularly like this response from the straighforward theorist. I am working on issues similar to your topic, and I have run into this response enough that I have dubbed it the “But it’s a question of welfare” response. Any thoughts on this ubiquitous response?
Troy,
Doug’s earlier post regarding moral and non-moral reasons suggested to me that your distinction between appropriate/inappropriate mental states, rather than relating to value, perhaps confuses the right and the good. That is, the ‘appropriateness’ of such states is a factor in determining whether it is morally right for a person to undergo such a state, but not in the value of the states. Several of the examples discussed so far seem to me examples of appropriate mental states because it is morally admirable (it’s hard to know what to make of the notion that mental states could be morally required, so I’ll stick with the less stringent ‘admirable’) for the persons in this situations to undergo certain pleasures or pains, but not because the pleasures or pains are intrinsically valuable states. In any event, divorcing the right from the good in this way is of course a controversial axiological move, but I could imagine Kantians in particular pressing this line of criticism,
Blar is right to urge me to try to give a “clearer definition of appropriateness,” since there are so many things that might be meant by the latter word. Let me say this: mental states can be divided into two classes. I’ll call them responsive and non-responsive. A responsive state is a response to conditions of the world as perceived by that person. They necessarily involve intentionality and (I think) some form of perception. Non-responsive states, on the other hand, might be caused by events that happen in the world but are not responses to such events in this sense. Thus, if I am sad because I have been fired, happy because I am in love, or suffering because a friend has died, then I am having a responsive state. If I am sad because of a chemical imbalance in my brain, happy because I have taken a pill, or suffering because I have a terrible headache, then I am having a non-responsive state.
To say that a responsive state is appropriate is to say that it is merited by the circumstances it is a response to. Thus, the death of a friend merits grief, which is why a grief response is appropriate. Taking a pill may cause happiness, but it does not merit happiness. This is why I say that with respect to non-responsive pleasures and pains, it may be their phenomenological character, rather than their appropriateness, that determines whether they are intrinsically good or bad. We do not ask of non-responsive states whether they are merited or not. Thus, a mental pain whose duration and intensity was equivalent to, say, a standard episode of grief, but which was caused by a chemical imbalance in the brain, would be intrinsically bad on account of its phenomenological character. Evaluating a responsive state, on the other hand—say, an actual episode of grief—is more complex. The appropriateness theory claims that such a state is intrinsically good if it is appropriate, i.e. merited by the circumstances.
This sort of account helps me put aside at least some alleged counterexamples. For instance, Blar mentions cases of “nontraditional sexuality.” Now, nontraditional sexuality is surely considered inappropriate by many people; in fact some people might take ‘nontraditional’ and ‘inappropriate’ as near-synonyms. But my own view of sexuality is that it is not, except in rare circumstances, a matter of (in)appropriateness; at least as applied to private mental states, the ‘whatever floats your boat’ theory seems to me fairly plausible. (There are, of course, certain moral reasons not to act on certain such states in certain ways.) So if McClure derives sexual gratification from the contemplation of various situations involving fish, or even from various actions involving fish, the pleasure may be unusual but it is not itself inappropriate. (Again, the actions may be wrong if they harm the fish, or someone else.) I don’t know what to say about the value of “his relationship with fish” – for the moment I’ll continue to confine myself, to mental states, as this question is hard enough.
The boring video case is mysterious, because apparently Scottie didn’t find it boring at all. (Or is it the fact that it was so boring that made it so hilarious? But in that case, why would we say that his amusement was inappropriate?) Obviously the experience of watching the tape was in some way amusing; moreover, we expect others to have a similar experience. Perhaps we are supposed to think that the video causes pleasure the way a drug does: as a direct physical process of some bizarre sort. But in that case Scottie’s pleasure is a case of non-responsive pleasure, and so the question of appropriateness does not arise. I would need to have the case described in further detail in order to understand the case for viewing Scottie’s pleasure to be, on the one hand, a case of responsive pleasure, and on the other, an unmerited and thus inappropriate pleasure. (Notice that if we alter the case to make the pleasure clearly inappropriate, by imagining, for instance, that the tape depicts the torture of an innocent person, the pleasure or amusement that a viewing might cause in some no longer seems clearly intrinsically valuable.)
I’m also not clear on the Hansel and Gretel case: why is it supposed to be clear that Gretel’s calm and lightheartedness are intrinsically good? Perhaps because we would rather have our last moments on earth be of this nature, rather than dominated by fear and panic? But there is clearly something wrong (yet not morally wrong) with Gretel’s response. Suppose the oven door were opened at the last moment; Hansel would flee, and thus escape, while happy Gretel would be unprompted to move. This, of course, may only show that Gretel’s inappropriate response is (potentially) extrinsically bad. But it seems to me that there is something deeply wrong with Gretel, and, even supposing that what is in question is value to the agent (that is, the question of welfare, or what makes one’s life go well), it is not clear that completely unjustified and inappropriate calm and lightheartedness do add anything to the value of a person’s life. Suppose that, throughout her life, Gretel was constantly experiencing pleasurable feelings of this sort – feelings that were entirely disconnected from what was actually happening in her life. She felt just the same way when her parents died, when she got divorced, on September 11 2001 . . . Gretel was a very happy person; yet it would seem to me that her life was, in fact, seriously deficient, and (though I realize this will be controversial) not a good human life at all.
Ben’s case struck me as very troubling at first; after some reflection, though, I’m not so sure. Is it really so clear that all this compassion (that word has nicer connotations than ‘pity’) can’t outweigh the badness of one little old stubbed toe, and make the overall situation intrinsically good? I’m not sure: intuitions are hazy here. Suppose we change the case: imagine that, rather than merely a stubbed toe, a person is actually undergoing utter agony. Now it seems more plausible to say that no number of compassionate responses could make up for the sufferer’s suffering. The problem, though, is that this might be explained simply by considerations involving the difficulty in summing up various goods and bads. Imagine the following case: a sufferer is hooked up to a mechanism such that by causing him to suffer utter agony, a small but not insignificant amount of pleasure is caused in each of x persons. (I am not imagining that this is malicious pleasure: the x persons are not enjoying his agony. Suppose they are not aware of how the mechanism works. The pleasure is piped directly into their brains.) It seems plausible to many people to say that no matter how large x is, the overall situation could not be good. You just can’t add up pleasures that way (so that lots of small ones counterbalance a great pain). But if so, then we might also think that you can’t add up other small goods that way (so that lots of small compassionate responses, for instance, counterbalance a great pain.)
Scott and Michael ask, respectively, whether I might not be confusing (1) welfare value with some other form of (intrinsic?) value or (2) the right and the good. Take the second point first: I think what consideration of grief, for instance, helps show is that we cannot separate the right and the good as easily as some have thought. Grief is the right response to certain sorts of circumstances. It is also, I think, a good response, in the sense suggested above regarding the Gretel example: a person who was unable to grieve the passing of her friends would be living a deficient and hence bad life, even if this lack had no extrinsic bad effects (denying opportunities for personal growth, etc.) (Perhaps rather than ‘bad’ I should be more cautious and say ‘less than fully good.’) Such a person is much like the subject of Nozick’s Experience Machine: she is happy, but her happiness is worthless (or at least, less than fully valuable) because it is not caused in the right way. (Merited grief, then, is the converse of experience machine happiness: it is unpleasant but, because it is caused in the right way, it is valuable.) Of course, the fact that merited grief is intrinsically good (if it is) in this sense does not imply that more such grief makes one’s life better; the point, rather, is that failing to experience such grief when it is merited makes one’s life less good. But it is, I am suggesting, a question of good, and not just a question of what it is right or admirable to feel, or what we have reason to feel.
As for Scott’s question – I’ve been talking about this intrinsic value as if it were welfare value: a person’s life is less good, I have claimed, if she fails to feel what is appropriate. I could, I suppose, be somewhat more cautious, allowing that pleasure always makes one’s life better and pain always makes it worse, but claiming that, in terms of ‘objective’ intrinsic value (or something like that) the fact that someone’s life is made better by the experiencing of an inappropriate pleasure can make the overall situation objectively worse. (Something like this is, I think, the way people like Noah Lemos want to go with respect to malicious pleasure.) Perhaps ultimately, this is the way to go; there does seem, after all, to be a valid distinction between what is good for a person, and what is objectively good (from the standpoint of the universe, so to speak), such that a person’s having a good life might well be bad in the latter sense. Still, I’d like for someone to convince me that inappropriate states – malicious pleasures, for instance – really do have positive intrinsic welfare value, i.e. that they make a person’s life better. I remain fairly skeptical about this.
Troy,
That was a thoughtful response. However, I think that the Hansel & Gretel example, in particular, still points in a direction contrary to Appropriateness Theory. My intuition is that Hansel’s frightened experience is intrinsically bad, while Gretel’s calm experience is close to neutral (not necessarily positive), while AT seems to have Gretel’s state of mind as intrinsically bad and Hansel’s as neutral or good.
Looking at the Hansel side of the story, I’d say that part of the evil of the old woman is the fear and despair that she made Hansel feel. We can isolate this effect in another example. Suppose that Ricky insults Stu (perhaps by saying something untoward about Stu’s deceased mother), and Stu appropriately feels offended. Since Stu’s negative mental state is appropriate, AT would consider it to be good (or at least not bad). I would like to say that Ricky’s action was wrong because of the way that it offended Stu, but that statement seems strange to make if Stu’s offended mental state is not intrinsically bad.
Going back to Gretel, we could imagine 2 different explanations of Gretel’s calm & lightheartedness. One is that she is a monk who has learned that a relaxed equanimity is the best response to worldly suffering. The other is that she is a giddy masochist & sadist who enjoys experiences in which she & others are suffering. Is her mental state appropriate in either case? Perhaps it is responsive in neither, since it departs drastically from the normal reaction of fear. I would be more inclined to say, though, that it is responsive in both, since it is based on the way that she interprets the facts of the situation according to her worldview. I do not see a plausible way to claim that she is responsive in only 1 of the 2 cases.
In general, a wide variety of responses can involve intentionality and perception, even though many seem inappropriate to us. What makes malicious pleasures, even, non-responsive? The “merited by the circumstances” test seems to accord better with our intuitions, but that is because it already includes an unjustified moral judgment.
Perhaps a more basic problem with responsiveness is that the prospects of separating ‘true’ feelings from merely physiological ones do not look promising. If you get fired, your brain chemistry does change, and that change in brain chemistry is at least correlated with the sadness that you feel. Further, the quality, depth, and duration of that sadness is likely to depend on your pre-existing brain physiology. This brain physiology, in turn, is a consequence of both previous experiences and other things, including genetics. The closer you look, the harder it gets to make a crisp distinction.
Troy – that’s a strategy you might use. But in your pleasure-mechanism case, an alternative explanation is available: those getting the pleasure, even though it’s not malicious, arguably don’t deserve it, as a result of the mechanism by which they get it. So the situation might fail to be good for that reason. The same can’t be said in the compassion case.
I guess in the toe-stubbing case my intuitions just aren’t hazy at all. The general idea that seems hard to deny is that you can’t get a good world just by putting in some pain and some appropriate responses to it. Somebody somewhere needs to be enjoying themselves in order to get a good world. That’s my main reason for not wanting to say that appropriateness has intrinsic value.
Troy,
I wonder whether you might say a bit more about appropriateness, and the role it plays in determining the value of responsive mental states.
In particular, when you speak of ‘merited’ responses, is merit based on the agent’s perception of a scenario, or the actual facts? That is, if Alicia feels tremendous pity for a another person who stubs her toe, would this pity be merited if Alicia believes (falsely) that the other person is feeling intense pain?
[This goes back to Blar’s Gretel. What if Gretel simply didn’t understand the danger she was in? Her lack of understanding of her situation seems inappropriate – an epistemic problem, at least. But given that she does not see the situation as dangerous, her lightheartedness seems quite appropriate. So, there might be a further question. Would you want to say that a responsive state could be made inappropriate if it is based on an understanding of a situation that is false or epistemically inappropriate / unjustified – even if the responsive state would be appropriate relative to the flawed understanding of the case?]
I should perhaps add – things get rather tricky if we hold that responsive states that seem appropriate given false or unjustified beliefs are in fact inappropriate. For example, if I falsely believe that someone is in great pain, and I feel pity yet this is deemed inappropriate (given my false belief), we might ask: what then would be the appropriate response? Wouldn’t we be disturbed by a person who did not feel pity when she believes that others are in great pain (even if falsely)?
Perhaps we could distinguish between judgements conerning the intrinsic value of these responsive states, and the value of the character of those who have the states. [So in the case of a person who feels sorrow based on the false belief that a friend is suffering: the sorrow is inappropriate, and thus intrinsically bad. But the person’s moral character at least is shown to be good – insofar as she’d have these reactions to the suffering of others.]
Troy,
Would you want to generalize your view to cover other types of responsive states?
The generalized view would go something like this: For all responsive states, the intrinsic value of that responsive state depends on whether or not it constitutes an appropriate response to its given intentional object.
Now you do want to say that Smith’s being pleased that P, where P is something like “Jones is in pain,” is not an intrinsically valuable mental state. But do you also want to say that the intrinsic value of believing, desiring, fearing, etc. depends on whether or not they’re appropriateness in their given circumstances? For instance, does the value of Smith’s desiring that P depend on whether or not P is desirable? And does the value of Smith’s fearing that P depend on whether or not P is something that warrants fear (i.e., something that constitutes a real danger)?
Intuitively, it seems that fear is an intrinsically bad state whether or not it is appropriate in the given circumstances. It seems that it is intrinsically bad for Smith to fear that a terrorist attack is imminent even if that fear is appropriate. Of course, that fear might be instrumentally valuable; it might, for instance, move Smith to do something to make himself less vulnerable to such an attack, but the fear itself isn’t valuable. It would be better if he could rid himself of the fear while still doing whatever he can to best prepare for the possibility of such an attack.
Now you could respond by simply saying that fear and pleasure are not analogous; you can say that whereas the intrinsic value of fear depends solely on its phenomenological character, the intrinsic value of pleasure doesn’t. But what then is the crucial difference between the two? Why does the intrinsic value of the one depend on its appropriateness (with respect to its intentional object) but not the other? Or do you, instead, want to say that fear is intrinsically good (or, at least, not intrinsically bad) when it’s appropriate? But, then, why does it seem like a good thing if Smith could just take a pill and rid himself of his fear, assuming that he’s already done all that he can about the potential danger?