1. Introduction
According to subjectivist theories of well-being, an individual's degree of well-being depends on the extent to which their preferences or desires are satisfied.Footnote 1 Such theories are attractive because they closely tie a person's well-being to what that person cares about, or at least what they would care about if they were rational and well-informed. So, unlike hedonism, subjectivism allows that someone's well-being might depend on more than just their pain, pleasure, or other affective experiences, if that person cares about more than just their affective experiences. Similarly, unlike objective list theories, subjectivism holds that whether allegedly objective goods such as knowledge or artistic achievement contribute to someone's well-being depends on whether that person cares about those goods. There is no objective list of achievements that is intrinsically good for each person regardless of what they care about: each person has their own subjective list, depending on their desires or preferences.
Like any initially attractive theory, subjectivism faces several challenges. Perhaps the most serious is the problem of interpersonal comparisons of well-being: subjectivists have a notoriously hard time making sense of claims about some individuals having a greater or lesser degree of well-being than others, or some events benefiting or harming some to a greater or lesser degree than they benefit or harm others. While most of the literature on this problem focuses on preference-satisfaction theories of well-being, I have argued elsewhere that a more promising avenue is available: subjectivists can bypass the problem simply by switching from a preference- to a desire-satisfaction theory of well-being (Barrett Reference Barrett2019; compare Greaves and Lederman Reference Greaves and Lederman2018, Van der Deijl and Brouwer Reference Willem and Brouwer2021).
To summarize, the basic difficulty with interpersonal comparisons of preference-satisfaction is conceptual. Given their essentially comparative structure – I cannot “prefer x” but can only “prefer x to y” – no simple analysis is available of the claim that my preferences are better satisfied by x than yours are by y. But desires are monadic and, I have argued, come with primitive strengths: I can indeed “desire x” full stop, and with some strength that is psychologically basic rather than reducible, say, to facts about my preferences over prospects. This suggests the following straightforward analysis of the claim that my desires are better satisfied by x than yours are by y: the strength of my desire for x is greater than the strength of your desire for y. Although various further questions no doubt remain about the nature and measurement of desire strength, the desire-satisfaction theory therefore at least avoids the conceptual problem of interpersonal comparisons long plaguing the preference-satisfaction theory. Unlike with preferences, no fancy footwork is required to understand, conceptually, what interpersonal comparisons of desire-satisfaction might be.
If this is right, then the move to a desire-satisfaction theory seems quite welcome. However, since the importance of the distinction between desire- and preference-satisfaction theories of well-being has only recently been appreciated, there have been few attempts to work out how, exactly, individuals’ degrees of well-being depend on the extent to which their various desires are satisfied. Any psychologically realistic picture must acknowledge that we have desires not for complete descriptions of states of affairs, but rather for specific objects or features of states of affairs. The question is how the satisfaction or frustration of one's various specific desires combine to yield a verdict about one's overall degree of well-being.
A natural answer is that one's degree of well-being depends on one's total desire-satisfaction. But as an alternative to this totalist approach we might adopt a proportionalist approach, on which one's degree of well-being depends on the proportion of one's desires that are satisfied. Unfortunately, critics of subjectivism have recently argued that neither approach provides a plausible account of individuals’ degrees of well-being, at least in cases involving the gain or loss of desires (Van der Deijl and Brouwer Reference Willem and Brouwer2021). If these critics are right, then subjectivists are in serious trouble. Although they may avoid the conceptual problem of interpersonal comparisons of well-being by adopting a desire-satisfaction theory, this leaves them unable to provide plausible comparisons of degrees of well-being in either the intrapersonal or interpersonal case.
My goal in this article is to defend the desire-satisfaction theory of well-being, and so subjectivism, from this criticism. I begin by elaborating the totalist and proportionalist approaches and explaining why the latter must be rejected. I then defend totalism from two objections due to Van der Deijl and Brouwer (Reference Willem and Brouwer2021): first, that it is inconsistent with the “resonance constraint” that motivates subjectivism, and, second, that it cannot account for the “Epicurean intuition” that having unsatisfied desires is bad for people. As we will see, totalism distinctively claims that gaining and satisfying new desires always increases one's well-being and that having unsatisfied desires typically decreases one's well-being. I conclude by briefly sketching how totalism may therefore have the resources to respond to another perennial objection to subjectivism: the problem of adaptive desires.
2. Totalism versus proportionalism
There are many things I desire, and these desires have different strengths. For example, at a given time, I may strongly desire to stay with my partner, somewhat less strongly desire to maintain my job, and much less strongly desire to win a board game. Any plausible version of the desire-satisfaction theory of well-being should claim that satisfying stronger desires improves my well-being more than satisfying weaker ones. But this falls short of a full account of how the satisfaction or frustration of my various desires contributes to my overall degree of well-being.Footnote 2 Here, two approaches seem available (Van der Deijl and Brouwer Reference Willem and Brouwer2021).
According to totalism, my overall degree of well-being is my total desire-satisfaction: the sum of all my satisfied desires, weighted by their strength.Footnote 3 So, for example, if my desire to stay with my partner has strength 0.5, my desire to keep my job has strength 0.4, and my desire to win the game has strength 0.1, and – being a very simple sort – I have no other desires, then my total well-being in the case where I stay with my partner, keep my job, but lose the game is 0.9: I have one desire of strength 0.5 satisfied, one desire of strength 0.4 satisfied, and one unsatisfied desire of strength 0.1 that has no effect on my well-being.
The alternative to totalism is proportionalism. On this approach, what matters to my well-being is not my total desire-satisfaction, but the proportion of my desires that are satisfied: my total desire-satisfaction divided by the total strength of all my desires that are either satisfied or unsatisfied. So, in the above case, my degree of well-being is again 0.9, but for a different reason: the total strength of my satisfied desires is 0.9, the total strength of all my desires including my unsatisfied desire (to win the game) is 1, and 0.9 divided by 1 is 0.9. The key difference between the two approaches, then, comes out if we imagine that we remove my unsatisfied desire. On the totalist approach, this makes no difference to my degree of well-being, which remains 0.9: I still have one desire of strength 0.5 satisfied and another of strength 0.4 satisfied. But according to proportionalism, losing this desire increases my well-being from 0.9 to 1, since my total desire satisfaction remains constant at 0.9, but the total strength of all my desires decreases from 1 to 0.9, and 0.9 divided by 0.9 is 1.
Although I have so far focused on desires for ease of exposition, it is worth noting that people have not only (positive) desires, but also (negative) aversions. I may both desire to win a game, and be averse to not winning, and these may be of different strengths: anecdotally, for example, it seems that some people strongly desire to win games without being very averse to not winning, while others have only weak desires to win but strong aversions to not winning (where the former tend to be better sports). We should therefore follow Van der Deijl and Brouwer (Reference Willem and Brouwer2021) in extending the above two approaches to accommodate aversions.Footnote 4 For example, according to totalism, my degree of well-being should be understood as the sum of all my satisfied desires minus the sum of all my realized aversions – each weighted for their strength.
Both totalism and proportionalism seem initially plausible. However, Van der Deijl and Brouwer (Reference Willem and Brouwer2021) argue persuasively that proportionalism yields untenable verdicts in cases of depression. The problem is that depression typically induces a loss of desire, such that a depressed person might stop caring about much of what they previously cared about, and so find themselves with a very meager set of desires. If these few desires are satisfied, then proportionalism implies that the depressed person is very well off. For example, if, in my state of depression, I am left with only a desire to “sit on the couch all day and watch YouTube,” and I do this, then proportionalism implies I am maximally well off, since all my desires are satisfied (Van der Deijl and Brouwer Reference Willem and Brouwer2021: 12). Yet this cannot be right: depression paradigmatically involves not only a loss of desire, but also a loss of well-being. Indeed, if, coming out of my depression, I develop a wide range of strong desires, many of which I satisfy, this intuitively renders me better off. Totalism implies this much, since it holds that gaining and satisfying new desires increases my total desire-satisfaction and so my well-being. But proportionalism instead says that gaining these new desires makes me worse off if even one is unsatisfied, since this upsets my perfect ratio of satisfied to unsatisfied desires – namely, having my one and only desire to sit on the couch watching YouTube satisfied. This is highly implausible.
So, Van der Deijl and Brouwer are right to reject proportionalism. Indeed, even setting aside their analysis of depression, it seems obviously false that someone who loses all their desires but one very weak satisfied desire (say, to stay alive) is thereby rendered maximally well off.Footnote 5 This suggests that what matters to our well-being is not the proportion of our desires that are satisfied but our total desire-satisfaction. In other words, it suggests that we should embrace totalism.
3. The resonance constraint
On first glance, totalism can easily handle the case of depression. However, according to Van der Deijl and Brouwer (Reference Willem and Brouwer2021), things are not so simple. The trouble is that while totalism gets the intuitively right answer in this case, it does so in a way that apparently abandons the motivation “at the heart of subjectivism” (2021: 4). This motivation is well-captured by Railton in the following famous passage:
Is it true that all normative judgments must find an internal resonance in those to whom they are applied? While I do not find this thesis convincing as a claim about all species of normative assessment, it does seem to me to capture an important feature of the concept of intrinsic value to say that what is intrinsically valuable for a person must have a connection with what he would find in some degree compelling or attractive, at least if he were rational and aware. It would be an intolerably alienated conception of someone's good to imagine that it might fail in any such way to engage him. (Railton Reference Railton1986: 9)
Railton's “resonance constraint” states, in other words, that “an object, event, state of affairs, etc., φ is good for an agent x only if x takes a valuing attitude (of the right sort) towards φ” (Dorsey Reference Dorsey2017: 687). Specifically, in the case of the desire-satisfaction view, the idea is that something can be good for me only if I desire it. This, recall, is what motivates subjectivists to abandon objective list theories, which hold – contrary to the resonance constraint – that knowledge or artistic achievement, say, might be good for me, even if I don't care about them.
Van der Deijl and Brouwer argue that totalism violates the resonance constraint. To see what they have in mind, recall the case of depression. If, emerging from my depression, I generate new desires that I satisfy, this intuitively benefits me, and totalism agrees – if, say, I gain two desires of strength 0.5, and satisfy both, then my well-being increases by 1. But note that, when I am depressed, I might have absolutely no desire to form new desires and satisfy them. So totalism implies that something can be good for me even though I don't hold any pro-attitude toward it; in particular, it implies that gaining new desires and satisfying them is good for me, even when I don't desire that this occur. This, Van der Deijl and Brouwer claim, violates the resonance constraint (2021: 14–16).Footnote 6
Although this is an intriguing argument, it appears to rest on an ambiguity. The resonance constraint says that something can only be “intrinsically valuable” for me if I have some pro-attitude toward it, such as a desire (Railton Reference Railton1986: 9 [emphasis added]; Dorsey Reference Dorsey2017). This is the motivating insight at the heart of subjectivism. But this constraint is perfectly compatible with the idea that certain things that are not intrinsically good for me may nevertheless be good for me in the sense that they produce an increase in my well-being – even when I don't have pro-attitudes toward them. The most familiar way this can occur involves an instrumental relation. For example, if I desire to be in the sun but don't desire to walk outside, then walking outside may increase my well-being without being intrinsically good for me because it promotes something that, according to totalism, is intrinsically good for me: being in the sun. But another way something can produce an increase in my well-being without being intrinsically good for me involves an enabling relation: something may enable an increase in my well-being by making something else intrinsically good for me when it otherwise would not be (compare Dancy Reference Dancy2004). For example, if I am currently in the sun, then, according to totalism, having a desire to be in the sun increases my well-being. But this is clearly not because having the desire is intrinsically good for me. Rather, what is intrinsically good for me is being in the sun, and having a desire to be in the sun only increases my well-being because it makes being in the sun intrinsically good for me when it otherwise would not be.
With this in mind, we can now see why totalism's claim that gaining and satisfying a new desire always increases my well-being doesn't further imply that gaining and satisfying a new desire is intrinsically good for me even when I don't desire that this occur, as would indeed be a violation of the resonance constraint. Suppose again that I am so depressed that my only desire is to sit on the couch watching YouTube. Sitting on the couch watching YouTube, therefore, is the only thing that is intrinsically good for me. But now suppose further that, as my mental health improves, I obtain a new desire to be in the sun, and I walk outside on a sunny day. Developing and satisfying this new desire is good for me in the sense that it produces an increase in my well-being. But, crucially, developing and satisfying the new desire is not intrinsically good for me, since it is not something that I desire to occur. Instead, developing the new desire enables this increase in my well-being by making it the case that being in the sun is intrinsically good for me, such that when I walk outside and find myself in the sun this is indeed intrinsically good for me – because I now desire it.
At no point in this story, then, does totalism violate the resonance constraint. Although the prospect of gaining and satisfying new desires might not resonate with me in my current state of depression, it is a strength rather than a weakness of totalism that it can explain how my well-being may nevertheless increase if, once I emerge from my depression, I develop and satisfy new desires. The view is not that generating and satisfying new desires is intrinsically good for me even when I don't desire that this occur, but rather that generating new desires makes more things intrinsically good for me – such that I can increase my well-being by achieving those things that are rendered intrinsically good for me by my new desires. Totalism is therefore consistent with the resonance constraint at the heart of subjectivism.
4. The Epicurean objection
There remains another problem for totalism. Although we have seen that depressed people who lose almost all their desires are not very well off, it might nevertheless seem like losing one's unsatisfied desires can at least sometimes increase one's well-being. Van der Deijl and Brouwer call this the “Epicurean intuition”: “A person may sometimes be directly benefitted by a reduction in unsatisfied [desires]” (2021: 12). Totalism violates this Epicurean intuition, claim Van der Deijl and Brouwer, since it implies that losing unsatisfied desires never has any effect on one's well-being, as this never affects one's total desire satisfaction. So if the Epicurean intuition holds, totalism must be rejected.
To motivate the Epicurean intuition, Van der Deijl and Brouwer contrast the cases of Bill and Will (2021: 10–11). Bill is very rich and so is able to satisfy all of his desires, leaving him with 50 satisfied desires of strength 1. Will is just like Bill, except that he has more desires than Bill and can't satisfy all his additional desires – he has 51 satisfied desires of strength 1, and 9 unsatisfied desires of strength 1. According to totalism, Bill has 50 well-being, and Will has 51 well-being, leaving Will better off than Bill. But Van der Deijl and Brouwer think this implausible: gaining ten additional desires and only satisfying one of them should make one worse off. Following Epicurus, who held that the path to greater well-being was to eliminate desires for what one cannot obtain, they suggest that “unsatisfied [desires] may be a burden on someone's wellbeing” (Van der Deijl and Brouwer Reference Willem and Brouwer2021: 12).
My response to this objection is that it again turns on an ambiguity between what is intrinsically good (or bad) for someone and what increases (or decreases) their well-being in some other way. It is certainly true that having unsatisfied desires typically decreases one's well-being. But there are various ways to explain this without appeal to the claim that having an unsatisfied desire is intrinsically bad for someone, and these explanations are perfectly consistent with totalism. In other words, the observation that people are often benefited by a reduction in unsatisfied desires is surely correct, but to refute totalism, Van der Deijl and Brouwer need the stronger claim that people are intrinsically benefited by the reduction of unsatisfied desires. And this stronger claim is much less plausible.
There are at least three ways having unsatisfied desires might decrease one's well-being without being intrinsically bad for one. First, desires have a close connection to affect or feeling, such that having an unsatisfied desire typically feels bad: when a desire is unsatisfied, this often leads to a feeling of frustration, of longing or loss. People are generally averse to bad feelings, and so according to totalism, their well-being is set back in cases where they have unsatisfied desires that give rise to them. This much is commonplace.
Second, desires also have a close connection to motivation, such that when one desires something, one is typically motivated to pursue it. When one fails to achieve the satisfaction of a desire one pursues, this takes up time, energy, and resources that one would have otherwise spent elsewhere, satisfying one's other desires. There is therefore an opportunity cost involved in having unsatisfied desires that one unsuccessfully pursues. Further, many seem to have a specific aversion to wasting their time, energy, and resources in this way, which might partially explain (and partially rationalize) the widespread tendency to engage in “sunk cost” reasoning.
Third, some people seem to have a higher-order aversion to having their first-order desires unsatisfied. For instance, I myself can remember occasions when my life was going well in most respects, but there was one strong desire I was not satisfying – say, a desire for a fulfilling romantic relationship (before I was lucky enough to find my partner). In those moments, I remember feeling not only frustration at my particular unsatisfied desire, but a higher-order aversion to there being an area of my life where I was failing to achieve what I wanted. I am not sure how widespread this aversion is, but it doesn't seem psychologically unusual.
With these explanations in mind, let us return to the contrast between Bill – who has all his desires satisfied – and Will – who has more (equally strong) desires satisfied, but also has some unsatisfied desires. One way to distinguish the hypothesis that people's well-being is intrinsically diminished by unsatisfied desires from the hypothesis that this is merely a typical effect of having unsatisfied desires is to flesh out the case in a way that rules out the three explanations I have offered, and check whether we retain the intuition that Will is worse off than Bill. To fairly test our intuition about the case of Will and Bill, that is, we must stipulate that Will has greater total-desire satisfaction than Bill even after factoring in (or “controlling for”) any negative feelings associated with Will's unsatisfied desires to which he is averse; even after factoring in any failed attempts Will has made to pursue his unsatisfied desires and any aversion he has to the associated waste of time, energy, or resources; and even after factoring in any higher-order aversion Will has to having unsatisfied desires. For example, we might suppose that Bill has several (previously unmentioned) realized aversions that are cumulatively exactly as strong as the aversions associated with Will's unsatisfied desires, though of a different source.
I submit that, once we clarify the case this way, we dissolve any intuitive pressure to think that Will is worse off than Bill. If Will has more (equally strong) satisfied desires than Bill, and if the various realized aversions associated with Will's frustrated desires are exactly counter-balanced by other (equally strong) aversions of Bill's, there is nothing remotely unintuitive about concluding that Will is better off than Bill – as totalism claims. So totalism can explain the intuition that unsatisfied desires often make people worse off by reference to the various aversions that typically accompany unsatisfied desires, or the tendency of one's failed pursuit of them to come at the cost of satisfying other desires, without leaving any residual intuition unexplained. Reflecting on these sorts of cases therefore lends support to, rather than undermines, totalism.Footnote 7
5. Conclusion
Switching from a preference-satisfaction to a desire-satisfaction theory of well-being is a promising way for subjectivists about well-being to avoid the problem of interpersonal comparisons of well-being. However, it is possible that this switch brings further challenges, which ultimately render subjectivism untenable. While Van der Deijl and Brouwer should be applauded for raising this possibility, for distinguishing totalism from proportionalism, and for noting the defects of the latter, their challenges to totalism are unsuccessful. Totalism denies that unsatisfied desires are intrinsically bad for one and allows that one can increase one's well-being by gaining and satisfying new desires, but there is nothing unintuitive or inconsistent with the motivation for subjectivism about either result.
Indeed, I conjecture that totalism's approach to cases of gained or lost desires may even help subjectivists deal with other perennial objections – for example, the problem of adaptive desires (Khader Reference Khader2011). After all, even if individuals in unfavorable circumstances tend to adapt to these circumstances by losing their desires for unattainable objects that seem intuitively good for them, totalism can explain both why this desire adaptation may often be good for someone while the relevant objects remain unattainable – due to the various aversions associated with unsatisfied desires – and yet why, at the same time, obtaining these objects will often be good for someone, not intrinsically, but in the sense that obtaining them increases that person's well-being. In particular, totalism implies that obtaining these objects will increase someone's well-being if the person will redevelop their desires for them after obtaining them, rendering these objects intrinsically good for that person once again.
Unfortunately, I cannot fully explore this issue here. So for now I simply note that, given the advantage of the desire-satisfaction theory over the preference-satisfaction theory when it comes to the conceptual problem of interpersonal comparisons, and given that totalism provides more plausible comparisons of degrees of well-being than proportionalism in cases of gained or lost desires, the totalist desire-satisfaction theory of well-being appears to be the best version of subjectivism available. At least until future arguments suggest otherwise, I recommend that subjectivists embrace it.Footnote 8
