I. INTRODUCTION
Most philosophers working on the issue agree that the correct theory of human well-being will contain a subjectivist element. Of course, just how important this element is, is a contested issue. Some think it is all that there is to say about the issue, others take it to be just one item (of perhaps minor importance) among others on an objective list of goods, while yet others hold intermediate positions. The commitment (or admission) that a theory of well-being needs to have a subjectivist element is commonly expressed by saying that (part of) our well-being is anchored in our attitudes, more specifically in our pro-attitudes.Footnote 1 Consider, then
The Generic Subjectivist Thesis
ST*: If subject S has a pro-attitude towards event E and E obtains, S is better-off in virtue of that fact, other things being equal, than if E did not obtain or S did not have the pro-attitude.
I take ST* to be fairly uncontroversial;Footnote 2 but that is mostly because it is wildly underspecified. Almost everyone would agree that some version of ST* is true, but that is pretty much where agreement ends.Footnote 3 What exactly is to count as a relevant pro-attitude in this context, and are certain ways of forming such pro-attitudes (e.g. irrational or ill-informed ones) to be eliminated? Does ST* hold for all possible events or are there restrictions ruling out, for example, very evil events like murders or genocides? Do we count only intrinsic pro-attitudes? What is the proper range of subjects? Moreover, is the right version of ST* the complete correct theory of well-being or does it only describe part of the truth (maybe just one item on a pluralist list of goodsFootnote 4)?Footnote 5
In this article, I will focus on what I will call, following Stephen Darwall, the Scope Problem. This is a particularly vexing issue for pro-attitude-based theories of well-being. The problem is that the scope of events we can (and do) have pro-attitudes towards appears to be broader than the scope of events that are well-being-relevant in the sense implied by ST*. Compare my desire to be recognized as an authority in my field with my wish that people in the year 3000 live decent lives. Both of these may or may not be satisfied. But only in the former case does it seem plausible to say that this makes a difference to how well my life is going for me. Solving the scope problem requires that we find a principled way of drawing the distinction between well-being-relevant and well-being-irrelevant pro-attitudes. This is no simple task and there is currently no widely accepted solution in the literature. Indeed, some authors, most explicitly Richard Arneson, have gone so far as to declare the problem to be unsolvable, and take it to expose a fatal flaw in pro-attitude-based theories.Footnote 6 Others have recommended that desire theorists should simply bite the bullet on the counterintuitive cases.Footnote 7 But, in the face of robust intuitions accepted by large majorities, this seems like a desperate move.
I think that the scope problem can be solved, but that doing so requires taking a somewhat circuitous route. Instead of addressing the problem head on by devising a criterion sorting pro-attitudes in the two categories of well-being-relevant and well-being-irrelevant, we should begin by looking for a scale on which we can rank pro-attitudes from most relevant to the subject's well-being to least relevant. There is, of course, nothing new about the idea of such a scale. Every pro-attitude-based account of well-being needs an account of why it is that one satisfied pro-attitude contributes more to well-being than another. My contention is that, if we have the right scales for this task, we get the solution to the scope problem for free: that a pro-attitude is entirely well-being-irrelevant is simply the limiting case on the scale of well-being-relevance.
In section II, I will introduce the scope problem in a bit more detail. I survey some proposed solutions and their shortcomings in section III. In section IV, I argue that the way forward is to solve the scope problem by developing a scale ranking every pro-attitude in terms of its well-being-relevance. In section V, I outline a scalar model according to which the well-being-relevance of pro-attitudes is measured either by their hedonic tone, or by the subject's conative commitment. In section VI, I show how my solution to the scope problem avoids the related worry that pro-attitude-based theories of well-being make self-sacrifice a conceptual impossibility. Section VII expands the discussion of my model by considering some objections to, and implications of, its individual elements.
II. THE SCOPE PROBLEM
The most widely discussed versions of pro-attitude-based theories of well-being are desire theories. This is also the context in which the scope problem first attracted attention. In his influential defence of an informed desire account, James Griffin states the problem thus:
The trouble is that one's desires spread themselves so widely over the world that their objects extend far outside the bound of what, with any plausibility, one could take as touching one's own well-being.Footnote 8
The most well-known illustration of the problem is Derek Parfit's ill stranger scenario.
Suppose that I meet a stranger who has what is believed to be a fatal disease. My sympathy is aroused, and I strongly want this stranger to be cured. We never meet again. Later, unknown to me, the stranger is cured. On the Unrestricted Desire-Fulfilment Theory, this event is good for me, and makes my life go better. This is not plausible.Footnote 9
There are many other cases in the literature and they all seem to show the same thing: not all of our (informed and rational) desires are such that their satisfaction would make us better-off.Footnote 10 Generalizing this problem from desires to the broader class of pro-attitudes we may say that only some of our pro-attitudes are well-being-relevant (i.e. such that their satisfaction is good for us). Others are well-being-irrelevant. Any pro-attitude-based (part of a) theory of well-being needs to be able to clearly mark that distinction. In other words, we need to account for the apparent fact that the satisfaction of the desire in the ill stranger scenario does not enhance Parfit's well-being, but, say, my taking pleasure in the prospering of my children does enhance mine (as most people agree). There are a number of proposals for how this demand may be met.
III. SUGGESTED SOLUTIONS
While an exhaustive discussion of every suggested solution to the scope problem would be tedious, let me motivate the shift to starting with a scalar model with a brief overview of the literature. Existing proposals can be helpfully distinguished into three classes in accordance to whether well-being-relevant and well-being-irrelevant pro-attitudes are supposed to differ in terms of the grounds on which they are held, their objects or their psychological kind.
Wayne Sumner, among others, claims that well-being-irrelevant pro-attitudes are those that are held on disinterested (as opposed to self-interested) grounds. He suggests that the reason his satisfied desire for apartheid in South-Africa to end has not made him better-off is that he did not desire apartheid to end on self-interested grounds; he wanted it to end because it was unjust.Footnote 11 But this line of reasoning rests on an ambiguity concerning what it means to have a self-interested desire. When a subject has a disinterested desire, such as Sumner's desire for justice in South Africa, they do not desire something for themselves. Any theory that implied the opposite would be an implausible form of psychological egoism. But whether or not a subject desires something for themselves is a distinct question from whether or not the satisfaction of a desire does in the end benefit them.Footnote 12 This is the second way in which we can understand the phrase ‘to have a self-interested desire’. Once we recognize this distinction it is easy to come up with examples where the two senses of ‘self-interested desire’ come apart. Many parents desire or take pleasure in the prospering of their children. And while some of these pro-attitudes may have egoistic grounds (parents trying to live vicariously through their children), surely not all of them do. But it would be a mistake to say that the prospering of someone's children is a benefit to them, only if they favour it for their own (i.e. the parent's) sake rather than for the sake of the children themselves.
Many authors have thought that we need to focus our attention on the objects of our pro-attitudes. We may, for example, think that only pro-attitudes about the subject's life are well-being-relevant.Footnote 13 Unfortunately, as Parfit admits when making this suggestion, ‘it may be unclear what this excludes’.Footnote 14 Mark Overvold has suggested a way of making this thought precise. His idea is that a pro-attitude is well-being-relevant only if its object necessarily implies the subject's existence at the time the pro-attitude is satisfied.Footnote 15 In other words, a pro-attitude is about my life (and thus well-being-relevant) only if it is impossible for the pro-attitude to become satisfied when I am dead. Thus, Parfit's stranger is ruled out, because it is possible that he recovers after my death.
But Overvold's restriction is both too permissive and too restrictive. It excludes too much by ruling out, for example, an author's desire that her books will be appreciated. This desire could be satisfied either during or after the author's lifetime. Most people would say that, at least if the desire is satisfied during the author's lifetime, it is well-being-relevant. But Overvold's proposal rules this out, simply because the desire could also be satisfied after the author is dead.Footnote 16 The proposal is also too permissive. As Krister Bykvist has pointed out, it fails to rule out pro-attitudes ‘whose object is of the form I exemplify feature F . . . even where F is a mere Cambridge-property’.Footnote 17 Overvold's proposal would allow, for example, that a desire to live in a universe in which the number of stars is even is well-being-relevant. But it is plainly implausible that such a desire should pass the criterion of being about the subject's life.
Other attempts to draw the distinction in terms of objects fall prey to similar worries.Footnote 18 But there is also a more general objection to object-based views. Here is Dale Dorsey:
Imagine that I desire to spend my life traveling to the poorest villages of the world to help vaccinate young children against common maladies. Is this a prudential or nonprudential desire? Surely it depends on the person in question: for some this might be a life-long goal upon which the success of life depends. For others this might be regarded as an instance of self-sacrifice one desires to undertake for beneficent or moral reasons. Hence the desire's object cannot seem to distinguish between prudentially relevant valuing and its contrary.Footnote 19
Dorsey's point is that it is possible for the very same event to be the object of a pro-attitude that is well-being-relevant for the subject, and of a pro-attitude that is not. If that is right, all views that spell out well-being-relevance in terms of a pro-attitude's object are mistaken.Footnote 20
The class of pro-attitudes is very broad and includes such attitudes as wanting, desiring, taking pleasure in, being happy about, craving, judging to be good, wishing, etc. A third option, then, is to say that only some of these psychological kinds are well-being-relevant. Along these lines some authors state that ‘mere-wishes’ are well-being-irrelevant.Footnote 21 This sounds plausible enough, and I think it marks an important data point that any solution to the scope problem needs to respect. However, it is rarely stated what such pro-attitudes lack that justifies the ‘mere’ moniker.Footnote 22 According to Fred Feldman's ‘attitudinal hedonism’, by contrast, only the attitude of ‘taking pleasure’ is relevant for both happiness and well-being. Feldman seems to think of this as a sui generis type of attitude. But his account, too, lacks any detail about the psychological profile of this attitude, or why it is just this type of pro-attitude that counts.Footnote 23
The most explicit recent attempt at spelling out well-being-relevance in terms of psychological kinds is Dorsey's judgement-based subjectivism about welfare (JS).Footnote 24 Dorsey's suggestion is that what is good for a person is getting what she values, and that the relevant form of valuing is a belief or judgement that something is ‘good for me’.Footnote 25 This apparently handles the famous examples of the scope problem. I may have pro-attitudes towards all kinds of things intuitively too far removed from my life to affect my well-being. But this is typically not the case with beliefs about what is good for me. Neither Parfit hoping for the recovery of the stranger, nor Sumner desiring the end of apartheid in South Africa, thinks that the events they favour are good for them – on the most natural reading of these examples. However, it is not hard to imagine variations of the cases in which the protagonist does have that thought. To form the belief that an ailing stranger's recovery would be good for me, while eccentric, is not impossible. The problem for Dorsey's view is that adding this psychological quirk to Parfit's scenario does not change the intuitive verdict: it still seems implausible that the stranger's recovery is well-being-relevant for the protagonist.Footnote 26
Even more to the point, however, mere cognitive judgements are simply not the kind of pro-attitude that is plausibly connected to well-being in the way that ST* claims.Footnote 27 Suppose you learn about some event that you believe is good for you, and so you come to believe that you are now better-off. Usually such a belief will be accompanied by some kind of positive emotional reaction – people get good feels when they receive good news. But now imagine that any such reaction is absent. All that happens is that you coldly judge that your life has improved. It seems to me that this fact does not in itself make you better-off. This is not to say that your judgement has to be mistaken. Maybe what happened is good for you instrumentally, or maybe what happened is good for you independently of your pro-attitudes. What is implausible is the idea that an emotionless judgement, all by itself, could confer (prudential) value on an event. In fact, I take this to be another data point that every plausible account of the subjectivist element of well-being has to respect. It is, in this regard, just like the claim about mere wishes. That you have an idle wish that some event occurs does not make this event well-being-relevant for you; neither does a mere cognitive judgment that an event is good (for you).Footnote 28
IV. GOING SCALAR
Naturally, the preceding survey of suggested solutions to the scope problem is incomplete. However, that there is no solution to the scope problem that is considered successful by even a respectable minority should be sufficient to raise serious doubts whether we have been digging in the wrong place. I would like to suggest that the solution has been hiding in plain sight. Even after separating out well-being-irrelevant pro-attitudes, we would still need an account of how much each (relevant) pro-attitude contributes to well-being. Why not start with this task and then say that well-being-irrelevant pro-attitudes are simply those that score a zero on the well-being-relevance scale?
Above we encountered Dorsey's claim, in response to attempts to sort pro-attitudes by their object, that it is possible for the same event to be the object of a well-being-relevant pro-attitude, or a well-being-irrelevant pro-attitude. This point is worth expanding. The same event cannot just be irrelevant or relevant for a subject holding a pro-attitude towards it, but it can be relevant to different degrees. To see this, consider a number of variations of Parfit's ill stranger case.
Variations on Ill Stranger: I meet a stranger, learn he is sick and form a genuine desire that he recovers. Later he does indeed recover. As for me,
(i) I never think of him again.
(ii) I read of his recovery in the local news and am mildly pleased.
(iii) Though I never learn whether he recovers, I had a hand in his recovery, as I found a way to secretly donate some money towards his medical bills.
(iv) Though I never learn whether he recovers, I had a big hand in his recovery, as I started a research foundation dedicated to finding a cure and devoted most of my time to it.
(v) I had a big hand in his recovery, as I started a research foundation dedicated to finding a cure and devoted most of my time to it. And I also learn of his recovery through the records of my foundation and am elated.
I submit that the favoured event is well-being-irrelevant in variation (i), somewhat well-being-relevant in variations (ii) and (iii), even more relevant in variation (iv), and very much so in variation (v).
Moreover, I see no reason to think that the difference between (i) and the rest of the field is different in kind from the differences between variations (ii)–(v). This suggests that to solve the scope problem we may be able just to make do with an account of a scale (or scales) that measures how much of a difference the satisfaction of a given pro-attitude makes to our well-being. The kind of cases we worry about when we think about the scope problem would simply be cases where the answer to that question is ‘zero’. In the next section, I will sketch a scalar model of ST*.
V. A SCALAR MODEL
What is needed, then, is a dimension along which pro-attitudes can be ranked and which functions as a (at least partial) measure of how much of a benefit to the individual the satisfaction of each pro-attitude would constitute. I will suggest such a dimension presently. However, because I will suggest that there are in fact two separate dimensions playing the role outlined above, each of them applicable to a certain class of pro-attitudes, I first need first to introduce a distinction between two categories of pro-attitudes.
As mentioned above, there are many kinds of pro-attitudes. The class includes rich psychological concepts that differ from each other in all kinds of ways. However, we can draw a distinction that sorts all of them into two mutually exclusive categories. For every event that we have a pro-attitude towards we can either have or lack a belief that the event obtains. I will use the phrase ‘taking pleasure’ as a technical term for all pro-attitudes towards events we believe to obtain; and I will use the term ‘desire’ as a technical term for all pro-attitudes where we lack such a belief. With this distinction in hand, here is my suggestion for specifying ST*:
The Subjectivist Thesis
ST: If subject S has a pro-attitude towards event E and E obtains, this contributes to S's well-being to the degree that the pro-attitude involves conative commitment (in the case of desire), or hedonic tone (in the case of taking pleasure).
According to ST, all pro-attitudes are well-being-relevant to a degree (though for some this degree is 0). That degree is determined by one of two scales. The relevance of desires (in my technical use of the term) is determined by the scale of conative commitment, while the relevance of taking pleasure (again in the technical sense introduced above) is determined by the scale of hedonic tone. While a full defence of ST will have to wait for a future occasion, I will make an initial case for it here. This will involve a schematic explanation of how the two scales work (sections V.1 and V.2) and why they should matter (section V.3).
As we will see, the distinction between desire and taking pleasure, that is the distinction between pro-attitudes towards events we believe to obtain and events where we lack such a belief, divides pro-attitudes into those that come with varying levels of conative commitment, i.e. an intention or plan by the subject to bring about the desired event, and those that come with varying levels of hedonic tone. Thus, at any given time, every pro-attitude will only be measurable on one of the two relevant scales, and hence the question which of the scales to use will not arise.
V.1. Desire and conative commitment
That desires, and only desires, come with varying levels of conative commitment should be fairly uncontroversial. For every event that I favour but do not believe to obtain there is a level to which I am committed to bringing it about. At the low end of this scale are the aforementioned idle wishes, at the high end are goals and aspirations that I invest a lot of effort or other resources in, or are resolved to do so should the need arise. By contrast, if I believe an event to already obtain, the idea of a commitment to bringing it about to any degree does not make sense. Thus, instances of taking pleasure do not come with any level of conative commitment (I may, of course, be committed to protecting or prolonging an event I am taking pleasure in – but the prolonged event that I am then committed to bring about is not one that I already believe to obtain).
It is worth noting that this way of looking at things assumes a somewhat narrow notion of conative commitment. According to this notion conative commitment may be constituted either by doing something to bring about the desired event or by resolving to do so. A wider notion of conative commitment would also allow that one is conatively committed simply by being disposed in a way that one would do something to bring about the desired event under the right circumstances. According to this wider notion of conative commitment, many instances of taking pleasure would also come with conative commitment. If I believe that there is ice cream in the freezer and take pleasure in that fact, I might very well be disposed to make it the case that there is ice cream in the freezer under the right circumstances. Part of what would make circumstances right, in this case, is that I did not have the belief that there was already ice cream there.Footnote 29
I will return to the wider notion of conative commitment below (section VII.1). ST intends the narrower notion. Conative commitment, then, is a dimension along which we can rank different desires. On the one end of the spectrum are mere hopes or wishes which come with no, or almost no, conative commitment; on the other end are goals that the agent firmly resolves to bring about, or puts lots of effort into bringing about.Footnote 30
V.2. Taking pleasure and hedonic tone
It is a lot more controversial to say that instances of taking pleasure vary in hedonic tone. That is because, first, some people question the very notion of hedonic tone; and, second, it is usually employed in the context of sensory pleasures, not as a property of pro-attitudes.
Let us begin with the question whether there even is such a thing as hedonic tone. It is a well-known fact that sensory pleasures have wildly different phenomenologies and that it is anything but clear what they all have in common in virtue of which we group them together in the category ‘pleasure’. The idea that all sensory pleasures contain an isolatable phenomenological element – the sensation of pure pleasure, if you will – is a non-starter; what is the phenomenological commonality of listening to a symphony and eating ice cream, for example? Recognizing this, people have moved towards one of two kinds of theories of sensory pleasure. The first is that a pleasant sensation is any sensation that we desire or appreciate having. These are attitudinal theories of pleasure.Footnote 31 The second kind of view, hedonic tone views, insists on a shared phenomenology but modifies this commitment.Footnote 32 Such views hold that when we eat ice cream, for example, it is not that we taste sugar and vanilla, and then also have the experience of pleasure. It is rather that the taste of sugar and vanilla is pleasant and that pleasantness is a dimension of sensations such as taste and sound. Shelly Kagan's influential analogy with the volume of sound is helpful here. While it is true that a violin sounds completely different from a drum, they both have volume.Footnote 33 And that does not mean that, in addition to the way a violin sounds to us, we also experience volume – volume is simply a dimension of sound.
In a recent paper, Murat Aydede has suggested a way to reconcile these two types of views.Footnote 34 Aydede is sympathetic to the idea that hedonic tone, as a dimension of experiences, is something that has a certain kind of phenomenology. But he suggests that, when we look at the neurological basis for this phenomenology, we find something that looks somewhat similar to a pro-attitude. Before we ever get the chance to think consciously about sensory inputs, he explains, such inputs are being processed at the sub-personal level in ways that ‘will causally influence . . . the subject's motivational, cognitive, and behavioural priorities’.Footnote 35 The idea is that there is a sub-personal machinery that filters, biases, enhances and amplifies these inputs. This machinery makes us prone to form certain attitudes about these inputs on the conscious level, but it does not fully determine our attitudes there. Now, what people call hedonic tone is simply the phenomenology accompanying the operations of this sub-personal machinery. The picture is that when you eat ice cream, there is a sub-personal endorsement of the sensory inputs (a proto pro-attitude, if you will). That means that you are likely to be motivated to have another scoop, to judge that eating ice cream is good for you, to see ice cream vendors in a positive light, and so on. On one way of looking at it, this kind of biasing simply is what it is to experience something as pleasant. This is the truth in attitudinal theories of pleasure. But being so biased also feels a certain way – that is the truth in hedonic tone views.
Turning to the second of the two potential problems I started with, this account of sensory pleasure can be very naturally extended to the more general idea of taking pleasure in my technical use of the term. If there is a sub-personal machinery that springs into action to present a sensory input, say an itch, in a certain quasi-evaluative light, then it seems plausible that this same machinery could do this with an input that is not a sensation but a thought. I also think that phenomenology bears this out. It seems to me that I can introspectively rank different instances of taking pleasure in an event along the very same dimension of hedonic tone that I can use to rank sensations. Taking pleasure in Germany's 2014 world cup victory, for example, is something that comes with a lot of hedonic tone for me. Whereas, when I take pleasure in the fact that life expectancy has increased drastically over the course of the last hundred years, this does not have quite as much hedonic tone. That is despite the fact that I (obviously) judge that the latter is a much more valuable event (and if I were to learn that only one of them actually happened, I would very much hope that it would be this one). It is just that this judgement is made at the conscious (rather than the sub-personal) level and not as immediately felt. What is going on here, I think, is that when I contemplate my beliefs regarding the world cup and life expectancy, these thoughts are subconsciously prepared for evaluation in much the same way that sensory inputs are. At the level of sub-personal evaluation, I care more about soccer results than increased life spans, and so contemplating the world cup induces the warm glow of hedonic tone. It also seems to me that I can introspectively and without any stretch compare sensory and non-sensory pleasures in terms of hedonic tone. Eating a fresh mango usually has more hedonic tone for me than thinking about Germany's world cup victory, which has more than eating chocolate, which has more than contemplating the recent increases in life expectancy.Footnote 36
Hedonic tone, then, is a dimension along which we can rank instances of taking pleasure. On one end of the scale are cool cognitive judgements that an event lives up to a standard I embrace; on the other are times when the contemplation or experience of an event leaves me elated, euphoric, in ecstasy.
V.3. Why these scales?
I opened this article by claiming it to be fairly uncontroversial that the best theory of well-being will include a suitably refined version of ST*. In other words, there is something deeply attractive about the idea that a person's well-being is, at least partly, anchored in their own attitudes. Sumner, describing the initial attractiveness of desire-based theories of well-being, remarks that such theories portray every person as ‘a shaper of his own destiny, a determiner of his own good’.Footnote 37 As Feldman points out, however, this kind of language can be misleading. For we cannot plausibly maintain that our opinions about our own well-being are infallible, nor that we can freely decide what kinds of things we will need for a good life.Footnote 38 Rather, the idea captured by subjectivist theories of well-being is that what kind of things are good for a person – what are the sources of their well-being – depends on the kind of person they are. ST* is attractive, then, in virtue of providing a (part of a) general theory of well-being that accommodates the obvious fact that different people can derive well-being from very different sources.Footnote 39
If this is what makes (partial) subjectivism about well-being so attractive, the scales measuring how strongly a pro-attitude's satisfaction is well-being-relevant should be sensitive to how closely the pro-attitude in question is related to the subject's identity as a welfare subject. We saw earlier that many authors grappling with the scope problem have tried to distinguish between pro-attitudes towards events respectively inside and outside the subject's life. Having moved to a scalar model, we should ask a related question that allows for more finely grained answers: how central to the person's identity as a welfare subject is a given pro-attitude? While some of the terms employed in it may need further specification, this is the question we need to ask when approaching the scope problem.
The scales employed in ST were chosen with this very question in mind. If you not only desire an event but are committed to bringing it about, then it is a project of yours and this defines who you are as a being that can be benefited or harmed.Footnote 40 And if your attitude of taking pleasure in an event involves a lot of hedonic tone, you care about it in such a way that a positive evaluation begins at the sub-personal level. Again, it seems to me that this is a good way of measuring how central this pro-attitude is to your identity in the sense important for well-being.
It may be worth noting, at this point, that neither of these scales is intended as an overall measure of the strength of the subject's pro-attitudes towards an event. Pro-attitudes, as has been mentioned, are a diverse class and there are probably numerous plausible ways to assess their strength. In my discussion of hedonic tone, for example, I mentioned that if I were given the choice between my favourite team winning the world cup and an increase in human life expectancy I would choose the latter. This seems to imply that there is a sense in which my pro-attitudes towards increased life expectancy are stronger than my pro-attitudes towards soccer triumphs. But that notion of strength is not plausible as a measure of how much I am benefited when the favoured events obtain. I propose that the two scales used by ST are plausible as a measure of how much the obtaining of a favoured event affects my well-being. I make no further claims as to whether these scales are plausible as a measure of strength of pro-attitudes, intensity of caring, or some such notion.
Note that ST naturally accommodates the fact that neither the satisfaction of idle wishes, nor mere cognitive endorsements of the world, make us better-off. That is not because there is anything special about these kinds of pro-attitudes, but simply because they score a zero on the relevant scales of conative commitment and hedonic tone. If you merely wish that E but never do, nor resolve to do, anything about it, then E coming about does not make you better-off. Unless, of course, once E obtains you take pleasure in it – with hedonic tone. If instead you just coolly judge that E is good then, again, you are not better-off. Suppose for example, that I wish there would be peace in Ukraine. But I do not have any intention to do anything about it. Then peace in Ukraine comes about, I hear about it, and soberly think ‘well, that's a good thing’, but do not feel the least bit happy or elated. In this case, despite having sincere (and potentially quite strong) pro-attitudes towards peace in Ukraine I am not made better-off by it.Footnote 41 I take this to be the right result and it directly falls out of the view proposed here.
VI. THE PROBLEM OF SELF-SACRIFICE
Some readers may wonder why I have not discussed cases involving self-sacrifice so far. After all, Darwall coined the very term ‘scope problem’ in the context of an argument that unrestricted pro-attitude-based theories of well-being make self-sacrifice a conceptual impossibility. Overvold's solution grows out of the very same concern. In his now classic discussion, he presents the example of a father who sacrifices his life so that his children may prosper. Since this man valued the prospering of his children more than his own life, he gets what he had the strongest pro-attitudes towards when he dies and his children prosper. And thus, according to an unrestricted pro-attitude-based theory, he is faring as well as he could have under the circumstances. Thus, giving his life for the prospering of his children turns out not to be a sacrifice after all. At most it was a case of ‘cutting his losses’, i.e. he gave up something good (his continued life) for something better (the prospering of his children), when he realized that he could not have both.Footnote 42 More generally, the problem of self-sacrifice is that, whenever a person appears to sacrifice their own well-being to further some other cause, we may think that that other cause is the object of their strongest pro-attitude. But since, according to pro-attitude-based theories, getting the objects of our pro-attitudes is what benefits us, such theories threaten to render every case of putative self-sacrifice a case of acting in one's own self-interest.Footnote 43 Overvold takes this to be absurd and concludes that desires like the one of the sacrificing father must be well-being-irrelevant.
A closer look at the example, however, reveals that Overvold's contention that the father's desire for the welfare of his children is well-being-irrelevant is too strong (even though Overvold may be right to insist that the father does in fact perform an act of self-sacrifice).Footnote 44 It is plausible that the father's desire is well-being-relevant for him to some degree. We may ask whether, once the father has made his sacrifice, it makes a difference to his well-being whether his children actually do fare well. It seems to me that it does. If his plan succeeds and his children have a better life because of his sacrifice, he may not be as well-off as if he had not made the sacrifice (it is a real sacrifice after all); but he seems clearly better-off than in the scenario where he makes the sacrifice but his children fail to profit from it.Footnote 45 We should also note that Dorsey's objection to object-focused ways of accounting for well-being-irrelevance applies to Overvold's example: we can imagine different fathers who desire the welfare of their children in different ways and only some of those would be well-being-relevant. Or, more precisely, we can imagine different variations of this scenario in which the desire is well-being-relevant to different degrees.
By contrast with views, like Overvold's, that entirely discount pro-attitudes towards things that we make sacrifices for, ST provides a more nuanced picture. In Overvold's example, the model implies that, if the father's sacrifice is actually successful, he will be better-off for that fact (and probably quite importantly so, since his sacrifice surely represents a weighty conative commitment). As I just argued, this seems correct, but it invites the worry that self-sacrifice would be impossible. But ST has the resources to dispel this worry. While it departs from most views in treating the father's desire as well-being-relevant, it concurs with them in also placing important goods on the other side of the ledger. First, he forgoes any possibility of seeing any other of his projects through. Second, he forgoes any possibility of future occasions of taking pleasure.Footnote 46 And third, the good that he gets from the satisfied desire that his children succeed needs to be discounted for uncertainty, for his sacrifice does not guarantee this, and so may be made in vain. Without knowing further details of the case, we are in no position to say whether the case will come out as a genuine self-sacrifice (i.e. a case of taking a lesser rather than a greater good for himself). But that is how it should be. After all, as mentioned above, it does seem that the father has something at stake in how his children do after his sacrifice, and so it does seem like their success actually is good for him. If we find out that he was near the end of his life, without any other projects left on his plate, and few chances of taking pleasure ahead, we might think that he did what was in fact best for him (though he did not do it for that reason). If, on the other hand, he gave up a whole lot of opportunities for future goods, ST would correctly imply that his action did amount to genuine self-sacrifice.
VII. TWO LOOSE ENDS
ST is a rather bold model and a full defence would go beyond the scope of a single article. However, in this section, I will begin this task by discussing two issues that arise with regard to my use of the scales of conative commitment and hedonic tone respectively: whether my notion of conative commitment is too narrow (section VII.1); and what ST implies for the well-being contributions of so-called false pleasures (section VII.2).
VII.1. Broad conative commitment
In section V.1, I claimed that conative commitment cannot be co-present with a belief that the favoured event already obtains. Thus, it is a dimension only of instances of desire, not of taking pleasure. As I pointed out, in my brief sketch, this assumes a somewhat narrow notion of conative commitment according to which it may be constituted either by doing something to bring about the desired event or by resolving to do so. A wider notion of conative commitment would also consider an agent conatively committed if, and to the degree that, she is disposed to try to bring about the desired event under the right circumstances.
It may be tempting to think that this wider notion of conative commitment gives us a better measure of a pro-attitude's well-being-relevance than the narrow notion employed by ST. Consider cases in which there is no (narrow) conative commitment simply because the subject believes there is nothing they can do to bring about the desired event. Imagine, for example, that a friend of yours has an incurable disease. You may desire very much that they recover but you know there is nothing you can do about this. Such a desire seems to be more well-being-relevant than a desire to own a new pair of jeans, which you are planning to buy next week. Yet, the latter easily beats the former in terms of (narrow) conative commitment, simply because you know that you can actually do something about it.
I do not deny that (unless circumstances are rather strange) your friend's recovery is more important for your well-being than owning a new pair of jeans. But ST can account for this fact without appealing to the wider notion of conative commitment, i.e. the idea that, if circumstances were right, you would do whatever it would take to help your friend recover. After all, you are likely to take deep displeasure in your friend's illness as well as in your inability to help them. And while extrapolating a theory of ill-being (or negative contributions to well-being) from a theory of well-being is anything but a straightforward matter, it does seem reasonable to say that taking displeasure in an event makes a negative contribution to one's well-being according to ST, and that the same scale of hedonic tone applies to such attitudes of displeasure (though we should perhaps rebrand the scale hedonic-doloric tone in light of that fact). Thus, your pro-attitudes towards getting the pair of jeans, though carrying a modest amount of conative commitment (and perhaps some hedonic tone once you have them), is likely to be vastly outweighed by the negative value of your dissatisfied pro-attitude towards your friend's recovery.
It may be replied that, while there is no need to include broad conative commitment as a measure of the well-being-relevance of pro-attitudes, there is also no harm in it. And, since it seems plausible that the hypothetical commitment to helping your friend recover seems an important measure of how much you care, why should we not include it in our model? The answer to this question is that we would then count the pro-attitude in question twice. The very strong pro-attitude that you have towards your friend's recovery is already being accounted for by the fact that you are deeply unhappy about the current state of affairs, and by the further fact that you would be absolutely elated if, against all odds, they did recover. To say that there is further value because you would, if things were different, do what you can to help things along, gives this pro-attitude, important as it is, more than its due.
VII.2. False pleasures
I take the part of ST concerning attitudes of taking pleasure to capture the thought that pleasure is a good. If something like Aydede's view of sensory pleasures is correct and can be extended to non-sensory pleasures, then the experience of pleasure is the phenomenology associated with a particular type of sub-personal pro-attitude. This provides a subjectivist explanation of the value of pleasure. Nowadays, pleasure is almost universally acknowledged as good for the one experiencing it, but it is frequently thought that no argument can, or need, be given for why this should be so. According to ST, on the other hand, pleasure is good for you because to experience pleasure is to give a mental thumbs-up to whatever it is you are pleased about. Thus, to experience pleasure is one way in which the world conforms to a subject's standards. And to say that this contributes to a subject's well-being is to make a claim that anyone with subjectivist leanings should find attractive.Footnote 47
However, extending Aydede's view beyond sensory pleasures introduces the following complication. If the input that is sub-personally modified is a thought rather than a sensation, it is possible that the thought is mistaken. Thomas Nagel and Shelly Kagan famously illustrate the idea with the example of a businessman who takes pleasure in what he believes are his great personal and professional successes, when he is in fact being despised, ridiculed and defrauded.Footnote 48 Let us assume that the businessman's taking pleasure ranks high in hedonic tone. One would have to be a very committed hedonist to bite the bullet on cases like that and say that the businessman's life is going well for him. And, at least on the face of it, ST delivers the result that these false pleasures do not contribute anything to his well-being. After all, the view requires that pro-attitudes are satisfied – and his are not.Footnote 49
But that result is perhaps too strong. After all, we might think that the businessman's false pleasures do count for something. It seems clear that he is faring less well than he would if the events he is pleased about did actually obtain. On the other hand, it is less clear that he is not doing better than if the facts were what they are but he did not have his delusional pleasures. ‘At least he is happy’ does not seem to be a crazy thing to say about him.Footnote 50 Now, ST can be made to produce the result that false pleasures are good, but not as good as true pleasures, for the person who has them. I am unsure about the plausibility of the move that allows for this. But let me sketch it briefly.
Recall that the notion of hedonic tone I employ rests on the kind of proto-pro-attitudes that Aydede describes. Now, we might wonder what exactly the intentional object of these attitudes is when it comes to non-sensory pleasures. Consider, first, the intentional object of the businessman's pro-attitudes on the personal, or conscious, level. On that level, the object of the businessman's pleasures consists of the events that he falsely imagines to obtain in the world. Since things are not actually as he believes them to be, his pro-attitudes are not satisfied, and thus there is no contribution to his well-being. But things get messier at the sub-personal level. What exactly is the input that is being subconsciously manipulated to appear in a positive light to him? Perhaps at least part of it is his thinking that, say, his children love him. Maybe part of what he takes pleasure in is not the world as he imagines it to be, but the event of him contemplating these events. If so, then part of his taking pleasure is actually veridical rather than delusional, for he does in fact contemplate these events. And thus ST would imply that he is better-off than without his false pleasures, but worse-off than if his pleasures were real.Footnote 51
Again, I am unsure how plausible this move is. But, then, I am also unsure whether false pleasures contribute anything to a subject's well-being. A theory that would provide a clear and easy answer to this question may actually be suspect for that very reason. That it is unclear what ST’s verdict on false pleasures would be, and that the answer appears to depend on a difficult question about the intentional objects of proto-pro-attitudes, may thus well be a badge of honour.
VIII. CONCLUSION
It has been almost twenty years since Richard Arneson declared that the scope problem cannot be solved.Footnote 52 I think that he may well be right as long as we are thinking of well-being-relevant and well-being-irrelevant pro-attitudes as different in kind. This article is a call to change the channel and think of the scope problem in scalar terms. We need a scale telling us how much of a difference each (dis)satisfied pro-attitude makes anyway. I suggest that, if we get this task right, we will get the solution to the scope problem for free. I am confident that this is a fruitful direction for future research, even if ST were ultimately to be rejected in favour of a different scalar view.
The particular scalar model that I proposed will need to be developed more fully. Much remains to be said about both of the scales that it employs. But I think that the general idea is a very promising one: our lives go better for us when our pro-attitudes are satisfied, and to the degree to which those pro-attitudes involve either conative commitment or hedonic tone. This is a moderately pluralistic proposal.Footnote 53 The view postulates two distinct subjectivist elements of well-being, roughly corresponding to the concerns of desire theorists and hedonists respectively, which promises to improve over pure versions of either of these in terms of extensional adequacy. But there is a common rationale for including these elements. Both scales employed by ST are meant to measure how central a pro-attitude is to our identity as well-being subjects. And the more central a pro-attitude is in that sense, the more does its satisfaction make us better-off. The kind of well-being-irrelevant pro-attitudes that motivate the scope problem are, in this picture, simply those pro-attitudes that score a zero on the relevant scales of centrality.Footnote 54
