By In NDPR Discussion Forum Comments (20)

NDPR Discussion Forum on David Sobel’s From Valuing to Value: A Defense of Subjectivism

Welcome to our discussion thread on David Sobel’s From Valuing to Value: A Defense of Subjectivism, recently reviewed by Ben Bramble for NDPR. We have invited Sobel and Bramble to provide any comments they’d like on either the book or the review, and we hope other readers of PEA Soup will chime in with thoughts on either the book or the review as well.

From the OUP blurb on FVV: “Subjective accounts of well-being and reasons for action have a remarkable pedigree. The idea that normativity flows from what an agent cares about-that something is valuable because it is valued-has appealed to a wide range of great thinkers. But at the same time this idea has seemed to many of the best minds in ethics to be outrageous or worse, not least because it seems to threaten the status of morality. Mutual incomprehension looms over the discussion. From Valuing to Value, written by an influential former critic of subjectivism, owns up to the problematic features to which critics have pointed while arguing that such criticisms can be blunted and the overall view rendered defensible. In this collection of his essays David Sobel does not shrink from acknowledging the real tension between subjective views of reasons and morality, yet argues that such a tension does not undermine subjectivism. In this volume the fundamental commitments of subjectivism are clarified and revealed to be rather plausible and well-motivated, while the most influential criticisms of subjectivism are straightforwardly addressed and found wanting.”

From Bramble’s review: “In making all of these arguments, Sobel develops an extremely innovative and nuanced subjectivism. In my view, Sobel’s is probably the most sophisticated defense of subjectivism given to date. That said, I have a number of concerns. First, I am not entirely persuaded by Sobel’s response to the Amoralism Objection. When I think of an agent who knows all there is to know about a given innocent, but remains unmoved by her condition or well-being, I am inclined to think that we have here, not some brute to be condemned, but someone who is making a mistake, someone who has either missed something (an evaluative fact) or is responding in an ill-fitting way to their full and accurate appreciation of the evaluative facts. Such an agent fails to care about something that seems in some sense worth caring about. Sobel’s response, and indeed his subjectivism more generally, seems to lack the resources to properly account for this intuition.”

Love this idea? Nominate it for the Annual PEA Soup Awards!

20 Responses to NDPR Discussion Forum on David Sobel’s From Valuing to Value: A Defense of Subjectivism

  1. David Sobel says:

    I am extraordinarily grateful to Ben Bramble for his wonderful review of my book. Bramble is remarkably careful to get my view right and focuses on just what seems to me worth highlighting about what I say. Furthermore, he is surely overly generous in his praise, saying that I am simultaneously both perhaps the best advocate for, and the best recent critic of, subjectivism. But his excessive kindness should not distract us from his concise and accurate summary, as well as his probing and valuable challenges.

    As tempting as it is to linger over his praise (and trust me, in my alone time I do just that), Bramble’s forceful and thought-provoking challenges require at least attempting a response. Despite this being required, however, some of Bramble’s challenges are broad enough that I won’t attempt to respond to them here. Here I will try to make a start on just three of his concerns. I should say that I find Bramble’s worries very interesting and I am confident they require more thought than I have so far managed to give them. What I say here is just an initial reaction that hopefully can advance our discussion.

    Bramble’s first concern has to do with how it would impact a subjectivist view if one concluded that nothing matters. He writes,

    “Suppose I were to conclude that nothing matters, no way the world could go that would be in any way better than or preferable to any other way it could go. It seems that such a belief would entirely extinguish my motivations to live. Or, if I somehow had some desires remaining (or believed that I had some), I would not think that I had any reason to fulfill them. I would think that I was a fool for still having any. And how would I regard others who did not realize that nothing mattered, who were going about their lives with a sense of meaning or purpose? I would think them deluded, blind to the pointlessness of their activity. Subjectivism (or at least, the sort Sobel defends, which appeals to one’s rationally contingent nontruth assessable favorings) seems unable to account for this. It seems committed to saying that the reasons of all these beings might be entirely unaffected by the fact that nothing mattered or could go better than anything else. If Sobel’s subjectivism were true, it should not matter to me whether anything matters. This seems wrong.”

    (It would not surprise me if, indeed I have some reason to think, Bramble and I here are to some extent replicating the Parfit/Street exchange in On What Matters, vol. 3 (OUP, 2017) and Does Anything Really Matter? (OUP, 2017) that I have not yet found time to study properly.)

    The subjectivist I favor denies that nothing matters. They think nothing matters independently of what people care about, but given that people care about stuff, lots of stuff matters. According to the subjectivist, valuing creates genuine value (hence the title). Given that, if we are forced to add the premise that nothing matters, we will be forced to abandon subjectivism. Since Bramble wants to test what happens if we combine subjectivism with the supposition that nothing matters, I suspect the test Bramble has in mind is to add to premise that nothing matters in the way that the objectivist thinks things matter. He thinks when we add that premise we should conclude that nothing matters at all. The complaint about subjectivism is that, under that stipulation, they do not get that result.

    Here I think once again the example of matters of mere taste is useful. I myself think, and I bet many others do as well, that there are no stance independent reasons to favor diet coke over diet Pepsi or Lagavulin over Talisker. Does that mean that my favorings in this domain lack normative upshot or that it would be quite psychologically weird to continue to think my favorings grounded reasons to go one way rather than another. I find no temptation to reach such conclusions in these contexts. Speaking for myself, I don’t think that diet Coke stance-independently merits my favoring it over diet Pepsi. I don’t think my reactions are warranted here by stance-independent value. I assume that view is quite common. Yet we care, often quite a bit, and I think without confusion, about getting what we like over what we do not even in contexts where there are no stance independent reasons to go one way rather than another.

    If that is right, then the generality of Bramble’s worry seems overstated. It thus seems to me overstated to say that because nothing matters in a stance-independent way that therefore nothing matters. It seems to me overstated to suggest that it is either psychologically weird or normatively untempting to continue to act as if stuff matters after accepting that nothing matters in a stance-independent way.

    Matters of mere taste are cases where, I maintain, we can see that stuff matters simply because it matters to us, even if it does not matter regardless of our concerns. Bramble thinks, I believe, that if nothing matters in a stance independent way, then nothing matters at all. I think the case of matters of mere taste can show us how to block that worry. Here, I maintain, we have a context in which there is no stance independent value at stake but still, clearly enough to my mind, there remains stuff that matters.

    Second, and likely, given Bramble’s views, the concern that matters most to him is this one:

    “Sobel also worries that it is hard to see how such a sensation of pleasure could be good for somebody who happened to be indifferent to it. He writes: “Most likely pleasure seemed a uniquely plausible recommendation partially because the vast majority of actual people like it. But of course, in other possible worlds, most people do not like that sensation. What could then be said on behalf of [it]?” (226) But should we expect to have reliable intuitions about creatures so unlike ourselves? Perhaps we feel that beings must have some pro-attitude toward what is good for them only because everyone around here happens to like or want pleasure. Or perhaps it is no coincidence at all that all beings with whom we are acquainted like or want their own pleasure. Perhaps we all like or want our own pleasure because pleasure is the most obviously valuable thing. Why is it so hard to imagine beings who do not want their own pleasure? Perhaps it is because it is hard to imagine beings who are so dense as to not realize the value for them of feeling good.”

    I take it to be beyond question that we have reason to go one way rather than another in matters of mere taste. Thus those who would resist that such reasons are grounded by an agent’s stance or favorings must find an alterative source of grounding for such reasons. Because the Benthamite Hedonist surely offers the most plausible alternative, it provides perhaps the most important challenges to my claim that our reasons of mere taste are the clearest case in which our favoring attitudes ground our reasons.

    The Benthamite hedonistic view is that there is a flavor of sensation or set of flavors that grounds such reasons. These flavors of sensation would be fundamentally stance independent states—not instances of this kind because anyone likes them. (A rival view of pleasure would say that a sensation is a pleasure because I intrinsically like it for its intrinsic phenomenological features.) The fight between the stance dependent theorist and the Benthamite is likely the key fight in deciding whether we must allow that our favorings sometimes ground reasons, at least in matters of mere taste.
    My central claim is that it is not tempting to normatively recommend a flavor of sensation, simply on grounds of what it feels like, when the agent informedly dislikes that sensation. There is no categorical imperative, it seems to me, to choose any flavor of sensation regardless of whether one likes it or not. Part of Bramble’s response is to maintain that it might in fact be rare that people fail to like this sensation, not because liking it makes it pleasure, but because stance independent pleasure is so manifestly good that we recognize its goodness and respond rationally by being for it. I will explain shortly why I doubt this picture, but even if we accept it we still must deal with the case of the person who does not like the flavor of sensation that is claimed to be pleasure and I still find the suggestion that this person ought to choose such a sensation not for the sake of morality or for the sake of other people but due simply to what it will be like for them to experience that sensation, when they really in no way like it and in fact quite dislike it, hard to credit.

    But additionally I find reason to doubt the explanation Bramble offers for why, on his view, most people like or favor the flavor of sensation of pleasure. He says it is so obviously valuable that most people value it. But most philosophers who have thought about the question have denied the existence of any such sensation that underlies the valuable experiences they are tempted to call pleasurable. The epistemic status of this flavor of sensation is thus awkward—it is so obviously noticeably valuable that children (and perhaps animals?) find it so and respond rationally by valuing it, but many excellent philosophers such as Sidgwick intently focus on the search for such a sensation and yet deny its existence.

    It is true that we are often inarticulate about our sensations. I myself have a very hard time verbalizing what I am experiencing when I am drinking a wine. But I, for one, have a much less hard time noticing similarities in phenomenology. I can re-identify the same wine and distinguish it from other wines and I’ll bet you can too. If this were generally very hard to do then it would be surprising to me that people bother to keep track of what foods they like and what they dislike. This seems to me a much easier task than putting into words what I am experiencing. But in the case of pleasure, I put it to you, it is not true that we think “oh, there is that stance independent flavor of sensation again that I previously felt when I was taking a hot bath and when I was falling in love with my spouse and that I now felt again while winning this tense tennis match.” That is just what most philosophers who have thought about this question deny exists. Indeed, given the difficulty in identifying the sensation that the Benthamite says has normative status, it is a bit unclear to me how, in principle, to carry out an empirical investigation into whether people in fact tend to go for such a sensation, let alone imagine that such an empirical finding has already been run and vindicates Bramble’s hypothesis.

    Third, Bramble wonders how tempted I am towards a view like Chris Heathwood’s that does not require that we combine into one head what it would be like to live very different lives in order to determine how valuable the lives are. I had mentioned the availability of such a view in the postscript to my “Full Information Accounts of Well-Being” paper and Bramble wonders if I had changed my mind that such “isolatable” accounts of value are untempting. I want to say that I find views like Heathwood’s quite interesting and acknowledge that they avoid some of the pressures that Velleman, Rosati, Loeb, and I had noted on full-information views that seem to require stuffing so much into one head. Nonetheless I do not find such Heathwoodian views very tempting, partly for the sort of reasons Bramble outlines. But this initial reaction to such views is just a sort of hunch and I very much feel pressure to try to articulate my concerns about them and see if they can survive scrutiny. But to answer Bramble’s question: I don’t currently find such views all that tempting. I place more hope on the pair-wise comparison model. But in any case, I share Bramble’s worry that we must find some alternative to stuffing all the info about all possible lives into a single head—and that, I quite agree, is work the subjectivist owes us and has yet to provide.

  2. Sergio Tenenbaum says:

    Really looking forward to this discussion! Just a quick comment on your response to Ben’s discussion of pleasure. Let us assume the stance-independent theorist hedonist (SITH) goes for a view of pleasure more like Kagan’s view of pleasure. According to Kagan, pleasure is like loudness: loudness is not a separate sensation that occurs independently of pitch and other qualities. Rather it is a quality that can occur as an aspect of widely different sensations. If we accept this view, the evidence that philosophers deny the existence of sensation of pleasure does not seem to me to speak against the view; they might have been simply confused about what the sensation of pleasure was supposed to be. So you could think of the Sidgwick of sound saying: “I cannot find a sensation of loudness no matter how much I look for it; the sensation of a large orchestra playing and the sensation of a child screaming are completely different. To say that an auditory experience is loud is simply to say that we can easily discern its content”. Here it seems plausible to say that Sound Sidgwick is being misled not by a failure to reidentify an aspect of his phenomenology, but by his philosophical views about categories of sensations.

    It also seems to me that the SITH can explain away the intuition that I seem to have no reason to pursue pleasures in the few cases in which we don’t like pleasant sensations if the SITH accepts Ben’s view that pleasure is obviously valuable. It might be hard to accept the stipulation that that I don’t see the value in the pleasure, and so we might tend to slip into conceiving the case as one in which the experience is not pleasant contrary to the stipulation. Our intuitions would be more reliable in cases in which it’s clear that my not liking is the consequence of distorting influences that are somehow blocking the otherwise easy appreciation of the goodness of pleasure. But in such cases I don’t think we have any intuition that we do not have reason to pursue what we do not like ( full information subjectivists implicitly agree with this point). Due to my strong religious views on the evil of soft pillows, I do not to like the pleasant feeling of the pillow on my head; I’d rather lay my head on a rock. But here I don’t find particularly intuitive that you would have no reason to choose the soft pillow.

    By the way, I am not loyal to the SITH, but I thought I would join them just for a day.

  3. Ben Bramble says:

    Thanks, Dave, for your kind and stimulating response.

    Suppose you, Dave Sobel, come to think that nothing matters. On your view, this would be because there just happened to be some change in your conative states—you’d happened to cease caring about things, or wanting anything to be true. Now, suppose that, as a result of this change of heart, you are lying in bed, utterly withdrawn and apathetic. You look outside through your window at other people walking down the street, some with a spring in their step, going about their lives with a sense of purpose or passion. What are these people seeking? You know the answer: Friendship. Family. The flourishing of others, more generally. Achievements of various kinds. Pleasures of various kinds. And so on. Now, what is your attitude toward all of these things? These things, you think, are all pointless or worthless. However, given what these individuals are like—given that they, unlike you, do care about these things—you are committed (given your subjectivism about reasons) to thinking that they have plenty of reason to go after them. These people, it seems you must hold, are not in any way foolish or in the grip of some false picture of reality. On the contrary, they are perfectly justified in their effortful activities. This, I would suggest, is an odd place to end up, psychologically. (It is to endorse the following: “Friendship, family, pleasure, it’s all pointless, worthless, a total waste of time. But these people have powerful reasons to go after these things. There is nothing wrong with their thinking about matters. They are not mistaken about anything.”) Or would you rather want to say that these people are wrong in thinking that these things have value (since here you’d be expressing or reporting your own lack of passion for anything), but insist that they have reason to pursue them nonetheless? This seems an equally odd place to end up. (Here, I confess, I lose a grip on what you mean when you say they have plenty of reason to go after these things.)

    The oddness in thinking both that nothing matters, and that those people who are going about their lives with a sense of purpose or passion are not mistaken about anything, can be accounted for, I think, by the fact that in going about our lives with a sense of purpose or passion (or indeed going about them at all), we are thinking that some things matter in a stance-independent way. Our effortful activity is contingent on our thinking that things matter in this way. Our evaluative thinking is prior or comes first. For most people, when they lose their belief that things matter in a stance-independent way, their motivations to live are radically sapped. I, for example, believe it matters how I and others fare, and that in thinking this, I am not merely expressing or reporting my feelings or preferences about things. If I came to think that I was just expressing or reporting my feelings or preferences here, I would lose much or all of my motivation to live.

    This connects with the point I made in the passage that Shoemaker excerpts, but that doesn’t make it into your response. Namely: “When I think of an agent who knows all there is to know about a given innocent, but remains unmoved by her condition or well-being, I am inclined to think that we have here, not some brute to be condemned, but someone who is making a mistake, someone who has either missed something (an evaluative fact) or is responding in an ill-fitting way to their full and accurate appreciation of the evaluative facts. Such an agent fails to care about something that seems in some sense worth caring about.”

    I agree with you that much rests on the case of matters of mere taste. You say that even if we stopped thinking that things matter in a stance-independent way, we would still want to go after some flavours rather than others (and we would still seem to have reason to do so). You write: “Matters of mere taste are cases where, I maintain, we can see that stuff matters simply because it matters to us.” As you know, I agree that our reasons in such cases are in some sense grounded in our “responses”. But this is not because they are grounded in our contingent desires. It is rather because they are grounded in our pleasures (and pains)—i.e., particular felt states—which have stance-independent value.

    You are reluctant to grant that there exist these particular felt states that have stance-independent value. This is for (at least) two reasons: (1) There is no ‘common feeling’ underlying all pleasures (no ‘feeling of pleasure itself’). Pleasures are not united by how they feel, so must be united by our attitudes toward them. (2) Even if all pleasures did feel alike in some way, this feeling would not plausibly be good for someone who happened to only dislike it.

    In defense of (1), you cite your ability to keep track of similar wines. By contrast, you write, “in the case of pleasure…it is not true that we think “oh, there is that stance independent flavor of sensation again that I previously felt when I was taking a hot bath and when I was falling in love with my spouse and that I now felt again while winning this tense tennis match.”” You also write: “most philosophers who have thought about the question have denied the existence of any such sensation that underlies the valuable experiences they are tempted to call pleasurable. The epistemic status of this flavor of sensation is thus awkward—it is so obviously noticeably valuable that children (and perhaps animals?) find it so and respond rationally by valuing it, but many excellent philosophers such as Sidgwick intently focus on the search for such a sensation and yet deny its existence.”

    In past work, I have tried to explain why we can have trouble mentally focusing on the feeling of pleasure, by saying that the feeling is unusual in that it permeates our experiences, rather than being ‘tacked on’ to our experiences in some crude way, and also by mentioning our fallibility when it comes to introspecting our own phenomenology. Another reason why we do not think “oh, here is that feeling of pleasure itself again” during the diverse pleasurable episodes you mention is that most of us feel pleasures of different kinds all throughout each day (pleasures are simply too varied and common for such thinking to be in any way interesting or useful). By contrast, wine pleasure is (for most of us, at least!) comparatively unusual.

    I should add that this might be one case where thinking too much about a subject matter can actually lead one astray, or to lose sight of things. When you ask kids if there is anything that feels the same about different pleasures they feel (say, of icecream, playing with their friends, watching tv, swimming in the ocean, etc.), they nod and say “yes, silly, the good/pleasant feeling!” or something like that. In trying so hard to introspect the feeling of pleasure itself, some philosophers, I suspect, are succeeding only in losing their own natural grasp of it.

    What of somebody who informedly dislikes the feeling of pleasure itself (and doesn’t like it one bit)? I’m not sure there could be such a creature, since on my view we start desiring things in the first place only by coming into contact with the most obviously valuable thing—namely, the feeling of pleasure itself. (For more on this, see my papers “Evaluative Beliefs First” and “Evolutionary Debunking Arguments and Our Shared Hatred of Pain”, both available on my website (

    But suppose there were such a creature. Should we hope, for their sake, that they get to experience the feeling that gives them pleasure (makes them feel good), or rather the quite horrible feeling that they (given their quirky nature) merely happen to want? Should we hope that they feel good, or that they do not? I think we should hope that they feel good.

  4. David Sobel says:


    You ask me to consider the situation in which I come to think nothing matters. If I came to think that nothing matters for anyone, whether they have concerns or not, I would have then come to reject subjectivism. (I’ll use “matters to” for the psychological claim and “matters for” as the normative claim.) So instead what we should imagine is that nothing matters to me but I remain a subjectivist. It can be difficult to imagine a case where literally nothing matters to me, as I have to imagine not minding in any way that someone saw off my arm, etc. but let us try to suppose this is the case. The key here for me would not be the cognitive claim that I think that nothing matters but rather the fact that nothing conatively matters to me. I am also to imagine things matter to other people. They have pursuits that matter to them quite a lot and I know this. You think it awkward for the subjectivist to have to say, in this scenario, that nothing matters for me but lots of stuff matters for other people. I think that is exactly the most obvious thing to say in such a situation. I see no awkwardness at all in my thinking that if others value hanging with family and like eating chocolate and I do not that there is a key difference between us in who has reason to do such things.

    It feels to me part of the philosophical lore about morality that it is special in that it, unlike most other stuff, matters regardless of one’s concerns. I do think that is a sensible thing to say about morality and I think if subjectivism were true there would have to be some revision of common sense about morality. But I try to argue in “Subjectivism and Reasons to be Moral,” these revisions can be independently motivated, rendered explicable, and are less profound than it would initially appear.

  5. David Sobel says:


    You are not sure there could be an instance in which someone did not like the stance independent sensation you call pleasure. I find that a remarkable psychological claim. But when you suppose that there could be such a situation and address the question of what makes normative sense for the agent in that scenario, you say they should feel “good”. I find that phrasing unhelpful. If good just means a valuable state, then you and the subjectivist agree that they should feel good, they just disagree about what is good in this situation. I think you are saying that there is a stance independent sensation that the agent dislikes as much as you like yet, simply in virtue of what that sensation is like for that agent, we should hope for their sake that they feel that greatly disliked sensation. I think that is seriously implausible.

  6. David Sobel says:


    I need to check my interpretation of you on at least one matter first. I am thinking of your pillow example. I think I agree with you that standard Railtonian “wanting to want” full information views might have the upshot that first order likings do not necessarily have normative upshot. But I don’t want to buy into any completely subjectivist framework at this stage, let alone a specific version of such. At this stage what I am trying to say is that some favoring attitude or other plays at least a partially grounding role in our reasons of mere taste. So if liking for its own sake the feeling of resting one’s head on the soft pillow is one’s grounds for thinking one has some reason to do so (despite the religious resistance to such) that still feels a case to me where the view I am currently trying to support wins. There are issues, I grant, about how to keep such claims and incorporate them into a fully general subjectivism. But the claim that interests me here is that one cannot do without the favoring attitudes here, not vindicating a specific version of subjectivism.

  7. David Sobel says:


    I have thought more about Crisp’s determinate/determinable thought here than Kagan’s volume of sound version. Crisp thinks, I think, that pleasure is not a specific determinate feeling but rather a thing that different determinates have in common. I don’t know what that could mean other than that there is some phenomenological, independent from liking, commonality between the different experiences. I maintain that introspection tells against such a claim and I don’t know what would tell for it if not introspection.

  8. Sergio Tenenbaum says:

    David, let us start with the question is “What does Sidgwick fail to find through introspection”. Now, it can’t be that he finds that he cannot tell which experiences are pleasant and which are not; certainly Sidgwick thought he could do as much by introspection. Now if you ask SIdgwick “Which separate sensation is such that if you add or subtract to the experience of tasting ice-cream, it will make the different between having pleasure in eating ice-cream and not having pleasure in eating ice-cream”. Here I grant (at least for the sake of the argument) that Sidgwick could, by instrospection, determine that there is no such sensation. But the same would be true if he were asked a similar question about loudness. But the claim by Kagan that pleasure is an aspect of certain experiences (and that cannot be experienced in isolation from other aspects). I don’t know why such a claim about the nature or structure of an introspectible quality must itself be introspectible (I don’t even think that all intrinsic properties of an introspectible experience must themselves be introspectible). It seems to me that this must be false; the very fact that there are competing philosophical views on the nature of pleasure would tell at least against the view at least that it is easily introspectible. But, more importantly, you also ask what else could tell for or against this claim. I take it that it must be its capacity to explain other things that we know, or are intuitive, about pleasure. For instance, one could say in favour of Kagan’s view that it explains, well, why Sidgwick fails to find a separate experience of pleasure, but also (unlike stance-dependent views) but it also captures an intuitive direction of explanation between liking and pleasure (I like eating ice-cream because it is pleasant rather than the other way around–I find the parallel claim about pain even more intuitive), and even allows for the possibility that we fail to like pleasant experiences (I confess not to remember very well Kagan’s arguments for his view, so this is very much my own speculation).

    [I think you could make similar points about Crisp’s view, but you would need to have a better memory of Crisp’s view than I have…)

    I feel compelled to say at the end of every such comment that I am not a fan of the SITH view; I just don’t think it can be refuted by this kind of introspection.

  9. Ben Bramble says:

    Hi Dave,

    Concerning your first response, I think what I’m missing is your view about the good simpliciter. I am clear on your views regarding well-being and reasons, but you don’t say nearly as much about the good simpliciter. What’s your view about it? (Once I’m clear on this, I can respond.)

    To the second matter. On the view I’m currently tempted by (a far out theory, I know!…but give it a chance), all action is aimed at the (perceived) good—i.e., behind every action (indeed, every motivation) is an evaluative belief. This is a necessary truth, and so a design constraint on desiring beings. So, to get creatures desiring things in the first place (an extremely useful adaptation in the race to survive and reproduce), evolution had to get them having evaluative thoughts. To do this, it had to introduce good things for them to encounter. Now, the most obviously good thing is pleasure, and the most obviously bad thing is pain. This is why we evolved to feel pleasure and pain. The ability to have these feelings initially came about as a mutation (a mere fluke), but then took hold because of its motivational efficacy. We feel pleasure now only *because* it’s a feeling we want. *This* is why it is so hard to imagine beings who dislike the feeling of pleasure. What, we rightly wonder, would have motivated such beings to get to where they presently are in their lives, in the absence of a preference for pleasure over pain? Of course, there might be beings who later on in their lives ceased to believe that pleasure is good (ascetics, philosophers of various kinds, etc.). These beings, indeed, might no longer like or want the feeling of pleasure itself. But, for them, I have no reluctance to say they are better off feeling pleasure than some other feeling that they falsely think to be good for them, or having no feelings at all.

    But there is another way to make the point without bringing in this complex machinery. Imagine a pain or unpleasant feeling (I think it works better with pain than pleasure) that one is feeling *without realising it*. Intuitively, it is bad for one. Here, its badness for one clearly doesn’t require that one be disliking how one is feeling, for one doesn’t even realise that one is feeling this way. Now, amend the case so that one *is* aware of the feeling, but doesn’t dislike it. Why should this alteration makes a difference? Why should it remove the disvalue for one of the pain? The feeling is still painful/unpleasant. It still *hurts*. That seems enough for it to be bad for one.

  10. David Sobel says:


    Certainly Sidgwick (and let’s keep in mind he was not unique in his introspective powers) allowed that some things were pleasant and some not. But one possible way of marking that distinction, and that seemed to tempt him, was to look to sensations that are liked or otherwise favored for what they intrinsically feel like. So allowing that some things are pleasant is no advance for the SITH view. Such a favoring view of pleasure would explain more easily Sidgwick’s finding. I agree that we would naturally say we dislike pain due to what it is like. But I think the favoring attitude view can naturally explain this. What is it that I dislike about that sensation? The answer in cases where we want to call the sensation pain is what it feels like, not something else, such as it suggesting I have a disease. The dimension of experience of loudness is a dimension we are certainly very much aware of and have independent ways (decibels) of establishing. What would unite a dimension of experience we are not aware of and don’t have an independent way of establishing? Further, Kagan’s argument is, at best, a sort of possibility proof, not a reason to think this dimension actually exists. I’m interested in why you, given what you say above, would reject the SITH view.

  11. David Sobel says:


    I have views about what is good for or morally good but I am not sure I have views about what is good simpliciter. Or in any case I need help interpreting that question to know if I do.

    I think broadly what we have here is a case where intuitions run so strongly in opposite directions (perhaps in a theory-assisted way) that it can be difficult to see how to get put genuine pressure on the view we don’t find intuitive or distinguish the neutral to this debate data from what seems obvious to partisans.

    I find many terms in this debate, including pain, pleasure, hurts, feels good, is good, etc. to be contested between the Bethamite and the Favoring Attitude View. So when you say it still hurts, I find that unhelpful. I think we are disputing the best understanding of what it is for a sensation to hurt or be good, etc. So I wonder if we can find ways of describing our views and what we like about them that the other side can readily accept? It may not be all that easy.

    Just to see how the other side would see your case of the unnoticed pain, let me say this. I would not want to deny the possibility of unnoticed pains. But I don’t see that as warming us up for the Benthamite view. I just think it shows us that we might not notice that we prefer one option over another in a certain respect. One could allow that without being tempted to say that in understanding the sensations that benefit or harm it does not matter what one favors or disfavors.

  12. Ben Bramble says:


    (1) By talk of the good simpliciter, I mean talk of what makes things (or, if you like, ‘the universe’) go better rather worse, impersonally considered. One such thing might be well-being (i.e., things going well for particular beings). Another might be things being just, fair, equal, beautiful, or whatever. This notion really doesn’t figure in your theory?

    (2) I thought your claim was that *even if* a felt-quality theory of pleasure/pain is true, it is implausible that the feeling that is pleasure is good for beings who don’t like it (and that the feeling of pain is bad for beings who do not dislike it). But if we’re supposing a felt-quality theory of pleasure/pain is true, then there is surely nothing amiss in saying that the feeling that is pain still *hurts* (even for those who do not dislike it). That’s just what it is to suppose that the theory is true. (Perhaps you find the felt-quality theory of pleasure/pain so implausible that you have trouble even pretending that it is true?)

    (3) Concerning unnoticed pain, (as you know) I think there are plausibly cases of pain where we aren’t aware *on any level* that they are going on (so they cannot be the object of some de re pro-attitude). But that’s a tangent…

  13. David Sobel says:


    I have a view of what the Consequentialist ought to say is worthy of moral promotion—roughly what the agent would put into the aggregation machine on her behalf assuming that she knows how the machine works and that other morally considerable things will be fairly taken into account. But that probably is irrelevant for our purposes here.

    I guess much hinges on what it would be, in your sense, to assume a felt quality theory of pleasure/pain is true. Obviously it would not be to assume that the benefit is grounded by the “felt quality”—that won’t be common ground. It might be to think that the felt quality view best captures what pleasure and pain mean in English. I do think I could, in principle and depending on the evidence, accept that (while denying that pleasure/pain in that sense is directly normatively relevant). But I still don’t think that the most useful terminology for philosophers to use when discussing the Benthamite vs Favoring view. In part because I think it so unclear which is the best understanding of the English and we are really fundamentally interested in the normatively relevant uses of pleasure and pain.

  14. Hello Dave, Ben, and Sergio.  Really interesting discussion so far.  I was going to join the fray, if I might.  I was interested in raising something pertaining to the Sergio thread and the Kagan view of pleasure; but first, my curiosity and self-centeredness are getting the best of me, and I’m wondering if we might be able to unpack and discuss the ideas in this paragraph from Ben’s review:

    “Likewise, why is Sobel now more tempted than he originally was by an isolatable version of subjectivism (beyond its ability to overcome the objections in question)? Such isolatable theories seem to me quite inadequate. How can my preferences concerning just how my actual life is going be sufficiently well-informed to count as authoritative if I do not know a heck of a lot about other possible lives I might have led? Incidentally, given Sobel’s interest both here and elsewhere (namely, in his contemplated restriction to now-for-now desires mentioned earlier) in Chris Heathwood’s views on these topics, it would be interesting to hear more from him on exactly where and why he does not want to go along with Heathwood.”

    I confess to sharing this interest.  But to get things started, I thought I might say something about Ben’s suggested worry.  

    Roughly, and leaving out certain features of the view that I think are not relevant here, the form of subjectivism about well-being that I favor determines the relative welfare value of two possible lives that a person could lead in the following way. The value of each life is a function of all of the occasions in that life in which the person got something they wanted and of all of the occasions in that life in which the person failed to get something they wanted. The former are desire satisfactions, the latter desire frustrations. The former have positive welfare value and the latter negative welfare value, each proportionate to the intensity of the relevant desire. To get the welfare value of a whole life, you just add up all of these values. So, to compare the relative value of two possible lives that a person could lead, you determine the value of each life in isolation (hence Sobel’s name “isolatable” theory), and whichever has the value is the better life.

    Sobel, by contrast, favors a rather different subjectivist approach, on which (roughly) the relative welfare value of two possible lives that a person could lead is determined by which of those two whole lives the person would prefer. (Incidentally, Parfit uses the terms “summative theory” for the approach I favor and “global theory” for the approach that Sobel (and, incidentally, Parfit himself) favors.)

    Bramble asks, of the “isolationist”/summative approach, “How can my preferences concerning just how my actual life is going be sufficiently well-informed to count as authoritative if I do not know a heck of a lot about other possible lives I might have led?” My basic answer is that I just don’t think that being well-informed is important on isolationism/summativism (though I agree that it is important on globalism). It may help to have a concrete case at hand:

    Suppose that in my actual life, the only music that I know about and that is available to me, is Muzak. It think it’s ok and I sometimes have mild desires to hear it, which I satisfy. This gives me modest benefits.

    There is another life I could lead, which I know nothing about. If I were to lead that life, I would be exposed to Mozart, and it would really knock my socks off. I would become obsessed with Mozart and, luckily for me, would regularly be able to satisfying my strong desires to hear more and more Mozart. (For good measure, let’s just say that all else is equal between these two lives; they differ only with respect to music.)

    My actual preferences are ill-informed regarding the Mozart-obsessed possible life. But the isolationist/summative theory can still deliver the intuitively correct result that the Mozart-obsessed life would be a better life for me. For the Mozart-obsessed life has a greater number of desire satisfactions than does the actual life, and the Mozart desires are also more intense than the less numerous Muzak desires.

    Ben, does this speak to the worry that you were raising, or did I not really get it?

    David, can you say why you are inclined to reject isolationism/summativism?

  15. David Sobel says:


    This is stuff I really need to think more about. Thanks for joining in. That is a helpful example and in the neighborhood of cases that concern me about isolatable views. I grant that as you describe the case, it gets the intuitive answer. But it is not clear to me that we have a good way of measuring intensity of desire that is bound to get the plausible answer here. Why not think the person is crazy excited about the best music they have ever heard, muzak, but that when they hear Mozart they see that what they were previously listening to not nearly as awesome. It feels to me, at least in this case, like what is clear is that the agent has an informed preference here and that seems to me to make us look for a way of getting to the conclusion that the muzak preference is less intense. But it is not obvious to me how we can be assured to get this plausible answer with an independent notion of intensity of desire. If the person had a more intense desire for muzak (while uninformed about Mozart) than they had for Mozart (while uninformed about muzak) but they had a strong pref for Mozart over muzak when informed about both, would that not tempt you to think Mozart seems superior for them?

  16. David Sobel says:


    Let me try to say what I was trying to say above in a different way. Suppose we know that in the absence of info about what it would be like to be Socrates dissatisfied, a person desires to be (what we would say was) a fool satisfied and desire that more strongly than they desire to be “Socrates dissatisfied” although they do not know what the latter is like. This same person, if they first knew what it was like to be Socrates dissatisfied would desire to be Socrates dissatisfied, in the absence of knowing what it would be like to be a fool satisfied, and desire it more than being a “fool satisfied”. But if they knew what both options were like, they would have a stable pref for Soc dissatisfied. I assume that none of what I have so far said licenses us to infer that the desires in the uninformed states are such that the pref to be Socrates dissatisfied is more intense than the pref to be the fool satisfied. I take it to be compatible with all of the above that the desire to be Socrates (uninformed by what it is like to be the fool satisfied) is weaker than the desire to be the fool satisfied (uninformed about what it is like to be Socrates dissatisfied). Is that right? If so, then I am tempted to think that it does not matter what the relative intensities of the desires are in such a case because the informed comparison seems to carry authority regardless of those comparative intensities. Even if we think there are real worries about putting a person in a position to appreciate all options simultaneously, it seems to me when we can put options simultaneously before a person, her informed prefs carry authority.

  17. Sergio Tenenbaum says:

    David, you say “What would unite a dimension of experience we are not aware of and don’t have an independent way of establishing? “. I take it that the view does not say that this is a dimension of experiences you are not aware of. I was trying to draw a distinction between being introspectibly aware of a dimension of experience and being introspectibly aware *that* pleasure is a dimension of experience. The latter does not follow from the former, and I wouldn’t take Sidgwick’s or anyone else’s introspective evidence as probative in this matter (or for any other philosophical theory about pleasure).

    My main objection to Kagan-type view is that one of their central motivations is to provide a ranking of intensity of pleasure (much as you could do with loudness). I am very skeptical that there is a SITH ranking of pleasures across all types of pleasures (even a rough and imprecise one).

    By the way, I’m not sure I understand your claim that “Kagan’s argument is, at best, a sort of possibility proof, not a reason to think this dimension actually exists.” Again, I’m not sure I’m being faithful to Kagan’s actual argument, but it would be an argument on the basis of the explanatory power of the view (of course, you don’t think that Kagan’s view explains the phenomena better than the non-SITH views, but I thought you meant something else by this claim).

  18. Thanks for laying that out, Dave. It seems to me that the cases under consideration are ones that essentially involve future-directed desires (which I acknowledge are very natural desires to consider in this context). However, I actually don’t think that desires that occur *before* some outcome would occur are relevant to the welfare value of that outcome. What are relevant, in my view, are just the desires that occur *within* that outcome.

    Now, I do think that past desires are often good indicators of the value of some outcome that is future relative to them. And for these desires, I agree that the better informed they are, the better indicators they are. This is because informed desires tend to be fairly stable over time, and to persist into the very outcome that they are about. If you prefer at t1 your getting chocolate at t2 over your getting vanilla at t2, and this preference is informed about what chocolate and vanilla are like, this is good evidence that getting chocolate at t2 would be better for you than getting vanilla.

    But, in my view, this preference at t1 isn’t what makes getting chocolate at t2 better for you. What makes that the better outcome is how strongly you are desiring the chocolate as you are eating it compared with how strongly you would have been desiring the vanilla as you were eating that. I think that our antecedent informed preferences are good indicators of the relative strengths of these two later desires, but it is the two later desires that determine which outcome would be better for you.

    Returning to your cases, then, I agree that it would probably be better for this person to be Socrates rather than a Fool. This is because their informed preferences suggest that they would more strongly want to be Socrates if they were be Socrates than they would want to be a Fool if they were a Fool. But it is this latter fact (if it is a fact) that determines which of these two outcomes would be better for the person.

  19. Ben Bramble says:

    Hi Chris, thanks for joining in!

    I think I agree with David on this. If (i) I, as an X and knowing no other, am extremely pleased about being X, and (ii) I, as a Y and knowing no other, am only slightly pleased about being Y, but (iii) I, knowing both X and Y, would much prefer being Y to X, then it seems to me most plausible (assuming some form of subjectivism is true!) that Y would be better for me. At least, this is my intuition, for what it’s worth.

    Might you want to say that (i), (ii), and (iii) *couldn’t* all be true?

    (I wasn’t sure why you thought that Sobel’s response required that the desires in question be forward-directed.)

    Here’s a related question for you: To what extent do you think the plausibility of your response (or your case for an isolatable view) depends on its being only one’s desires for one’s current experiences that determine one’s well-being? (I have a follow-up question for you, but would like to know your answer to this beforehand, if that’s ok.)

  20. Hi Ben,

    That’s an interesting case (it takes the issue of future-directed desires out of it). I also think it’s a pretty weird case. To make it clearer, let’s make it less abstract. Let’s say we have a friend from a distant land who knows nothing of chocolate or vanilla ice cream. And these things are true of him:

    (i) If we were to give him chocolate, thus acquainting him with it (he would remain unacquainted with vanilla), he would LOVE the chocolate and devour it.

    (ii) If we were to give him vanilla, thus acquainting him with it (he would remain unacquainted with chocolate), he would like the vanilla pretty good, but not love it.

    (iii) If we were to give him a sample of each, he would much prefer vanilla to chocolate.

    I certainly don’t want to say that such a case is metaphysically impossible, but I hope you agree that it is rather baffling.

    Before we decide what to say about the case, I think we should divide it into two different ways it could go (for we might want to say different things about these two different versions of it):

    WAY ONE: If we were to give him a sample of each, and then give him chocolate, he would love the chocolate. And if we were to give him a sample of each, and then give him vanilla, would he merely like the vanilla. Despite this — despite the fact that he would love eating the chocolate but only sort of like the vanilla — he would still prefer to get vanilla over chocolate.

    This is very odd! If this were the true description of the case, I would be inclined to say that when it is said that he would love eating the chocolate but only like the vanilla, we are talking about his pro-attitudes towards the way they *taste*. But when it is said that he would prefer to get vanilla over chocolate, this can’t be a report of his attitudes towards the way they taste (otherwise, I fear that the case would be genuinely incoherent). Maybe, then, this preference is about some other feature of eating chocolate vs. eating vanilla, such as the ethics of the situation. Maybe, although he prefers the taste of chocolate, he believes that tasting tastes of this kind is immoral, and this explains why he would rather get the vanilla.

    What to say about the welfare facts in such a case? The simplest answer would be to balance all of these conflicting desires: gustatory desires of different strengths for chocolate and vanilla, moral desires of different strengths for these options as well. It could very well turn out on my view that getting vanilla is overall better for him (as according to your intuition in the original abstract version of the case).

    (I actually think that there is a further complication; I think that some moral desires should be excluded from well-being (not *because* they are moral but for a different reason). But let’s not open that can of worms.)

    Here is the other main way the case could go:

    WAY TWO: If we were to give him a sample of each, and then give him chocolate, he would only merely like the chocolate. And if we were to give him a sample of each, and then give him vanilla, would he love the vanilla. Accordingly, he would prefer to get vanilla over chocolate.

    This second way that the case could go is also a little odd, but for a different reason. It is a little odd, I think, that having been exposed to vanilla would cause him to lose his love for the way chocolate tasted in itself. But I’m sure this sort of thing does happen. If I give my one son something he usually loves, but then he sees that his brother got an even better thing, the first son may be unable to enjoy what he got. This version of our case may be sort of like this.

    Here is what I want to say about the welfare facts of this case. If our friend will never actually get acquainted with both flavors — if he will get acquainted only with the flavor that he gets a full serving of — then we should give him chocolate. After all, he’d love that more than he would love vanilla, if we gave him vanilla. I don’t think it matters at all that IF he were to have sampled both, THEN he would have loved getting vanilla and only liked getting chocolate. This is a place where I think idealized or full-info theories go wrong.

    Is it your view, by the way, that this counterfactual fact matters (at least assuming that subjectivism is true)? If so, then that may be a fundamental disagreement between us.

    Finishing up, if, on the other hand, we will in fact acquaint our friend with both flavors, then, after so acquainting him, we should give him vanilla, and the explanation as to why is the same as in the above: he’d love that more than he would love the chocolate, if we gave him the chocolate (after acquainting him with both).

    I don’t see anything troubling in all of this for the sort of view that I favor. I wonder if you still do.


    This post is already very long (sorry), but I also want to answer the other part of your post, Ben. It is NOT my view that only one’s desires for one’s current experiences determines one’s well-being? The view Parfit calls “Preference Hedonism” has this feature (as I interpret Parfit), as do some views of others (e.g., D.W. Haslett). But on my view, desires that we have for states of the world are also relevant to well-being. Still, my view is nevertheless a mental-state theory; this is because, on it, our desires don’t actually have to be satisfied for us to benefit — we just have to think that the object of the desire is true. These “subjective desire satisfactions” are pure mental states.

    But this raises a version of your question: To what extent does the plausibility of my response (or my case for an isolatable view) depend on this feature of my view, its being a mental-state desire theory? That’s a good question! I think my hunch is that it doesn’t so depend, but that is just a guess. I will have to think more about it.