1. Introduction
Desire satisfactionists are united by their belief that what makes someone well-off is the satisfaction of their desires. But this commitment obscures a number of underlying differences, since there are several theoretical choice points on the way to making this commitment precise. This article is about two of the most important choice points. The first concerns an epistemic requirement on well-being. Suppose that one's desire that P is satisfied. Must one also know (or believe, or justifiably believe) that one's desire that P is satisfied in order to benefit from P? If so, there is an epistemic requirement on well-being. The second concerns the time at which one benefits. Well-being is a temporal phenomenon: given that one benefits from the satisfaction of one's desire that P, when does one benefit? Perhaps one benefits at the times at which one desires P, or the times at which P obtains, or both.
In this article I will argue for a strong epistemic requirement that I call the awareness requirement, and for concurrentism, the view that one benefits from the satisfaction of one's desire for P only at times at which both one desires P and P obtains. My view combines these two structural conditions: one must be aware that the desire is satisfied while it is satisfied – if one only becomes aware that one's desire that P was satisfied after one loses the desire or after P ceases to obtain, then one does not benefit. In this article, I will not investigate whether these conditions are sufficient as well as necessary. We might want to exclude pointless, ill-informed, irrational, base, or malicious desires from those that could benefit one if satisfied, or we might instead hold that the above conditions are sufficient for benefit – I will remain neutral on this question here.Footnote 1 I call this view Concurrent Awareness Desire Satisfactionism (CADS). According to CADS, one is benefitted at t by the satisfaction of one's desire that P only if one is aware at t that one's desire that P is satisfied. Awareness at t that one's desire that P is satisfied has two components:
1. Awareness at t that P.
2. Awareness at t that one desires at t that P.
I view the latter condition as relatively weak. People sometimes lack awareness of their own mental states,Footnote 2 but the vast majority of cases are ones where people are aware of what mental states they are in, and in particular, aware of what they desire. I include the latter condition primarily for the sake of completeness; I will focus most of my attention on the former condition since, typically, it is much easier to be aware of one's own desires than it is to be aware that the content of those desires obtains. Awareness as I will conceive it is a factive mental state, so these conditions imply, respectively, that at t P obtains and at t one desires P. I will defend the epistemic and temporal components of the view separately, then explain the overall picture of well-being that emerges.
Here is how the article will proceed. In section 2, I argue for an epistemic requirement. Merely believing that one's desire is satisfied is not sufficient for benefit, since the actual satisfaction of one's desires is necessary. I also give a more comprehensive argument for the stronger awareness requirement that I favor. In section 3, I set aside the awareness requirement to consider the time-of-benefit question – when does a satisfied desire benefit one? The premise motivating debates about this question is temporalism about well-being: a welfare benefit benefits a welfare subject only if it benefits them at some interval or moment in time. I argue for concurrentism, the view that one benefits only at times at which both one has the desire and its contents obtain. In section 4, I put the two parts of the view together and discuss the comprehensive view of well-being that emerges: CADS. I argue that concurrentism and the awareness requirement are complementary theses, and I examine issues related to the intensionality of awareness and desire that emerge only when the theses are considered together. My claim is that CADS is the best version of desire satisfactionism.Footnote 3 I conclude by noting that if my view of well-being is correct, then well-being is an emergent property that obtains only when the mind and the world stand in the right kind of relation with one another. One must have a mind-to-world state (desire) to a content and must have a world-to-mind state (awareness) to the same content in order for that content to benefit one. Well-being is an organic unity that exists and is valuable only when these parts come together in the right way.
2. The Awareness Requirement
Following Parfit (Reference Parfit1984, 494) I call the view one is benefitted when one desires P if and only if P the Unrestricted Desire-Fulfillment Theory (UDFT). UDFT faces compelling objections. Recall Parfit's stranger on the train:
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, this 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.
Kagan (Reference Kagan1998, 37) presents a similar case. I desire that the number of atoms in the universe is prime, and in fact, the number of atoms is prime. According to UDFT, I am better-off in virtue of this fact. But Kagan (rightly) intuits that I do not benefit from the fact that the number of atoms is prime, for this fact does not affect my lived experience at all.Footnote 4
In response to these cases some theorists have proposed restricting UDFT by counting only the satisfaction of desires which are about one's own life as contributing to one's well-being.Footnote 5 One's desire about the stranger and one's desire about prime numbers are not about one's own life, so their satisfaction does not benefit one. Call this view restricted desire-fulfillment theory (RDFT). RDFT avoids some of the problems facing UDFT, but we can set up examples of a similar form which suggest that RDFT's diagnosis of the problem is off-track:
Oppressed Author: An author works on his manuscript for years, desiring nothing more than for it to be published and read by others in his field. However, he lives in a country with a repressive regime, so he entrusts his completed manuscript to a confidant who smuggles it out of the country to be published abroad. The confidant publishes the manuscript, which is very well received, but word of the manuscript's success never reaches the author.
It seems like the author does not benefit from the book's publication in this scenario. Though the matter cannot be settled on intuitive grounds alone, it seems that people who intuit that one does not benefit from the stranger's good health should also intuit that the author does not benefit from the book's publication. A unified explanation of what goes on in both cases is available: the object of one's desire obtains but one does not know about it. One's desire concerning the stranger is not about one's own life but the author's desire is about his own life, but this difference does not make a difference concerning the well-being of the people in question. The identification of this unified explanation obviates the need to draw a deep distinction between the things that benefit you if you desire them and the things that could not benefit you even if you desire them. This is advantageous because there will be hard theoretical questions about such a distinction in any case. For example, RDFT's “about one's own life” condition has to be spelled out in a satisfactory way, and it seems that, however it is spelled out, there will be a vague frontier between desires that are clearly about one's own life and desires that are clearly not about one's own life. The restricted theorist needs to give an account of how the satisfaction of borderline “about one's own life” desires affects one's well-being.
Von Kriegstein (Reference von Kriegstein2018) proposes an alternative to UDFT designed to avoid this problem. He distinguishes two states that are relevant to a person's well-being: someone takes pleasure in P when they have a pro-attitude to P and they believe that P obtains; they desire P when they have a pro-attitude to P and they lack a belief that P obtains. Desiring something that obtains increases one's well-being in proportion to the degree of one's conative commitment to it. Though the scalar notion of conative commitment is not susceptible to vagueness like the binary “about one's own life” condition, it remains vulnerable to a variation of the prime number example. Consider a deranged physicist, who is strongly conatively committed to there being a prime number of atoms in the universe, and performs experiments in his lab to try to bring this about. If there are actually a prime number of atoms in the universe (something that could not be known by the physicist), von Kreigstein's account would suggest that the physicist greatly benefits from this, because strength of conative commitment is the scale that determines how much one benefits from a satisfied desire. The problem is not that a weak desire for a prime number of atoms does not benefit one though a strong one could, it is rather that it seems that no satisfied desire pertaining to such a remote state of affairs, by itself, could influence how one's life is going.
These problems suggest a more perspicuous diagnosis of why one does not benefit in the prime number and train cases: there is an epistemic requirement on well-being that is not met.Footnote 6 To see this, consider variations of the cases in which you learn later that the stranger has been cured, and in which you learn that the number of atoms in the universe is indeed prime. In these modified versions of the cases, it seems as if you do become better-off by learning that your desire is satisfied, even though in both versions of the cases the desires in question are not about your life. Think of the satisfaction that you would experience upon learning the fundamental fact about the universe that you hoped was true, or your feeling of relief about the stranger's condition. This shows that an epistemic requirement can do the same work as the more obscure “about one's own life” condition, and more.
These cases demonstrate the need for an epistemic requirement of the form: an agent benefits from the satisfaction of her desire that P only if the agent stands in relation R to P. Here, R can be filled in by any epistemic relation: believing, knowing, having high credence that, etc. Some philosophers have suggested that the epistemic requirement is very weak. Heathwood (Reference Heathwood2006), for example, requires only that the agent believe that their desire is satisfied, in order to benefit. But importantly, he also thinks that mere belief is sufficient for benefit. He calls this view subjective desire satisfactionism (SDS), which he regards as a form of hedonism, as well.
If mere belief that one's desire is satisfied is sufficient for benefit, then the actual satisfaction of one's desire is not necessary, since one can have false beliefs. Hedonists will agree with the desire theorist who endorses an epistemic requirement about the revised prime number and stranger cases presented above: the feelings of satisfaction and relief (due to one's belief that one's desires are satisfied) are necessary for one to benefit. But SDS and hedonism imply that they are sufficient as well: the feelings of satisfaction and relief are all that matters, and they matter independently of whether the number of atoms is actually prime or the stranger is actually cured. This is not my view, for I think that one's recognition that one's desire is satisfied and its actual satisfaction are both necessary for one to benefit. I will argue against SDS, first by arguing against the sufficiency clause, and second by considering the necessity clause. Then I will defend a stronger version of the epistemic requirement.
According to SDS, what benefits one is not the satisfaction of one's desires, but rather the belief that one's desires are satisfied. Because of this, SDS has quite different implications than ordinary desire satisfactionism across a range of cases. For example, one motivation of desire satisfactionism vis-à-vis hedonism is to vindicate the idea that one is not rationally required to plug into the experience machine. But if what makes one well-off is the belief that one's desire is satisfied, then this is a good that the experience machine can provide – plug in and you can believe that all of your desires are satisfied even if none are!Footnote 7 Heathwood happily accepts this implication, though, since his project is to identify what he regards as the best version of desire satisfactionism with the best version of hedonism. Most other theorists, however, including myself, will regard the implication that plugging into the experience machine makes one better-off as a mark against SDS.
Consider one further argument against the sufficiency of belief for benefit. If SDS is true, then one becomes better-off by acquiring false beliefs that one's desires are satisfied, rather than by performing the actions that would actually satisfy them. According to SDS, when forming a false belief that P is easier than bringing about P, an agent who desires P will find that the most straightforward way to become better-off will involve forming a false belief that P, rather than trying to bring P about. There are many cases where self-deception might be a more efficient means to the end of acquiring a belief that P than actually bringing about P, even if this is not the norm. Indeed, SDS entails that the state in which I falsely believe that my desire for P is satisfied is better for me than the state in which my desire for P is actually satisfied but I do not believe that it is satisfied. This implies that the safer course of action for an agent who wants to be better-off is simply to try to get themselves to believe that their desires are satisfied, since the actual satisfaction of one's desires does not always cause one to have a belief that the desires are satisfied. The person who has many false beliefs that their desires are satisfied has a high level of well-being according to SDS. But this is ersatz well-being, not genuine well-being. When we take actions aimed at improving our lives, they are directed toward the world; the primary goal is not to shape our own doxastic states. Massive self-deception or being irrationally optimistic are not ways to become better-off; one must instead do the harder work of actually bringing about the things one desires. SDS cannot explain this fact.
This argument can be extended to show that one must bear an even stronger epistemic relation than true belief to the contents of one's desires in order to benefit. People usually have some satisfied desires which they do not believe are satisfied. If truly believing is the relation one must bear to the contents of one's desire in order to benefit, then the thing to do to make oneself better-off, if one can, is to form beliefs that all of one's desires are satisfied. This will increase one's well-being because it will take some desires from being merely satisfied (which does not give one well-being) to being truly believed to be satisfied (which does give one well-being on the present proposal). This too is ersatz well-being – one cannot improve one's lot by irrationally forming beliefs or self-deception, even if one is lucky and the contents that one is forming beliefs about happen to be true. The right kind of causal connection must obtain between P and one's belief that P in order for one to be better-off. The right kind of causal connection, whatever it is, is also what makes a belief count as knowledge. (This is not to fully endorse a causal theory of knowledge or awareness, but only to concede that there are some causal constraints on knowledge and awareness.) In order to benefit from the satisfaction of a desire one must be sure that the desire is in fact satisfied, and only knowledge can properly provide such assurance.
The problem with naïve desire satisfactionism is that it ignores the importance of one's felt-mental states for one's well-being, leading to the stranger and prime number objections. The problem with SDS is that it ignores the importance for one's well-being of the world's matching one's conative states, abandoning the original motivation for desire satisfactionism and leading to the experience machine. Introducing the awareness requirement allows us to get the best of both views. A match between one's desires and the world is important, but so are one's felt mental states. I discuss this matter further in the final section.
Now consider the necessity claim in SDS: one benefits from one's desire that P only if one believes that one's desire that P is satisfied. The stranger case and the prime number case support this claim. This is indeed a necessary condition for benefit, but only because it is entailed by a stronger necessary condition on benefit: one benefits from one's desire that P only if one is aware that one's desire that P is satisfied. This is the awareness requirement, the version of the epistemic requirement that I defend. Awareness, like knowledge, is a factive mental state: if one is aware that P, then P. Awareness, again like knowledge, entails belief: if one is aware that P then one believes that P. Indeed, awareness is a kind of knowledge: one is aware that P when one's knowledge that P is presented under a phenomenal mode of presentation. Awareness implies phenomenal consciousness. Philosophical zombies and advanced computers could perhaps know things, but since they are not conscious, they are not aware of anything. On my view, even if they can form desires, zombies and computers are not welfare subjects, because they cannot satisfy the awareness requirement. A weaker knowledge requirement would be unable to secure this intuitive result.Footnote 8
I have just argued that the epistemic requirement must be stronger than a true belief requirement, and that consciousness is necessary for being a welfare subject. Now, I will give two additional arguments in favor of the awareness requirement, then consider an objection. The first is based on intuitions about how we compare the well-being of similar subjects in similar cases. Consider two twins who have exactly the same desires, Ann and Beth. Suppose that Ann and Beth desire that their town's park be refurbished, but only Ann is aware that the park is being refurbished. I have the intuition that Ann is better-off than Beth. Her awareness that the park is being refurbished contributes positively to Ann's mental life, but Beth's lack of awareness that the park is being refurbished contributes negatively to Beth's mental life. Given the intuition that Ann is better-off than Beth and the setup of the case, the only possible explanation of the intuition is that an awareness requirement holds.
If you do not have this intuition in the case where there is a small difference between Ann and Beth, another way to see what is at issue is by taking the thought behind this example to its logical extreme. Consider someone who believes that all of their desires are unsatisfied, when in fact, all of them are satisfied. Such a person is radically deceived about how good the world is, according to their desires. It appears to them as if they have a very poor quality of life because they believe that nothing is the way they would like it to be. Every day for them is miserable. It seems far off the mark to say that this person is well-off, and indeed perfectly well-off since all of their desires are satisfied. Yet, one who denies the awareness requirement has to say, implausibly, that the person is perfectly well-off, despite the fact that things appear to the person themselves quite to the contrary. Facts about a person's well-being and what they believe about how their life is going can come apart, but not this far apart: one cannot believe that one is very poorly-off and in fact be perfectly well-off.Footnote 9 The awareness requirement explains why the person in this case is not well-off. Indeed, drawing a parallel between two cases seems appropriate here:
Case 1: All of one's desires are satisfied but one believes of each desire that it is unsatisfied.
Case 2: None of one's desires are satisfied but one believes of each desire that it is satisfied.
We should say the same thing about these two cases. In case 2 (analogous to entering the experience machine), one is not well-off because one's desires are not satisfied. In case 1, one is not well-off because one does not satisfy the awareness requirement.
Consider a variation on the above cases in which a person truly believes but is not aware that all of their desires are satisfied. The weaker true belief requirement judges the person to be perfectly well-off, but the stronger awareness requirement judges them not to be well-off. I think this is the right result. Since they are not knowledge the beliefs in question are unsafe, and as such are not reliable enough to secure the person's well-being. The person has something good – beliefs that they are getting the things they want. But these beliefs are not causally connected to the things that matter to the person in the way required to secure their well-being.Footnote 10 Considering this case also reveals an asymmetry in my view: one cannot be well-off but believe one is badly-off, but one can be badly-off but believe that one is well-off. This seems exactly right: one of the good things about being well-off is having accurate beliefs about how one's life is going, and one of the bad things about being badly-off is having false beliefs about how one's life is going.
The second argument I shall consider concerns rationality. Divide one's satisfied desires into two classes: those of which one is aware that they are satisfied and those of which one is not aware that they are satisfied. Surely, at most times, most of us have satisfied desires in each category. If the awareness requirement holds, then, it is rational to invest some time, energy or other resources into learning whether one's desires are satisfied, since one does not benefit from a satisfied desire until one learns that one's desire is satisfied. In particular, when we are inquiring about something, our goal is to gain knowledge about that thing, not merely a true belief. So, for one who accepts an awareness requirement, it is rational to pay to learn that one's desires are satisfied, or to buy insurance against forgetting that one's desires are satisfied, etc. This is the right result. Contrast this to what is rational for one who rejects the awareness requirement. They should be indifferent to learning whether or not their desire is satisfied, but in addition, they must be indifferent to losing all knowledge about whether their desires are satisfied. One who denies the awareness requirement should gladly accept a deal where they are paid one dollar in exchange for forgetting that all of their desires are satisfied. But this is the wrong result, and the awareness requirement explains why.
It is appropriate here to note an implication of my view concerning the value of knowledge. Theorists of well-being have traditionally tried to capture the thought that knowledge is intrinsically good for the knower. And only objective list theorists have had success in this endeavor.Footnote 11 Hedonists cannot account for the value of knowledge: two people can be phenomenologically alike but differ as to whether they know a proposition; knowledge has no distinctive phenomenological character. Aristotle says that man by nature desires to know, but he was probably not right about this, leading subjectivist theories to difficulty in explaining the value of knowledge for those who do not care about it. But if my view is correct, then knowledge is something like a transcendental condition required for the very possibility of well-being – no one can benefit from anything at all unless they have knowledge of it. Knowledge is not intrinsically good for the knower or good because the knower desires knowledge, but rather because knowledge (in the form of awareness) enables the things which one desires to benefit one. Without knowledge, desire would lack the capacity to contribute to one's well-being.
There is an alternative way to explain the data that I have appealed to in support of the awareness requirement. That is by positing that, at least typically, when someone has a desire for P, they also have a desire to know that their desire that P is satisfied. Call such a desire an epistemic meta-desire (EMD), and call this the EMD theory. According to the EMD theory, in the prime number or stranger cases, you are badly-off because you have a frustrated EMD, not because there is an awareness requirement. Likewise, the EMD theorist can claim that the cases as I have set them up above are inconsistent. If there are EMDs, it is impossible for all of one's desires to be satisfied though one believes of each desire that it is not satisfied. And it is likewise impossible for Ann and Beth to be alike in their desire satisfactions but unalike in what they are aware of.
The EMD theory is not correct, for at least two reasons. First, the view gets the wrong results about a person's net level of well-being in the cases that it is designed to handle. If Parfit desires that the number of atoms is prime, it is prime, he desires that he knows the number of atoms is prime, and he does not know that the number is prime, then his net level of well-being, on the EMD theory, is the intensity of his (satisfied) desire that the number of atoms is prime, minus the intensity of his (unsatisfied) desire that he know the number of atoms is prime. Thus, in this case, according to the EMD theory, Parfit has a middling level of well-being, yet the natural intuition is that he has no well-being. Yu (Reference Yu2022), an EMD theorist, claims that it is easy for us to mistake a middling or small amount of well-being for no well-being in thinking about such cases. But considering well-designed thought experiments can help us distinguish these options. Consider two duplicates of Parfit (in different worlds), both of whom desire that the number of atoms in their world is prime. In Parfit1's world the number of atoms is prime but in Parfit2's world the number of atoms is not prime. Neither Parfit1 nor Parfit2 know anything about the number of atoms. So, their respective EMDs are both unsatisfied. Yet, the EMD theorist must say that Parfit1 is better-off than Parfit2 in virtue of the fact that in Parfit1's world there are a prime number of atoms and in Parfit2's world there are not. This comparative judgment holds whatever the relative strengths of their first-order desires and their EMDs, but it becomes easier to see the stronger their first-order desires are relative to their EMDs. If the first-order desires have strength 1 and the EMDs have strength 10, then it will appear as if Parfit1 and Parfit2's levels of well-being are approximately equal (−9 for Parfit1 and −11 for Parfit2). But if the first-order desires have strength 10 and the EMDs have strength 1, then it will be clear that Parfit1's level of well-being (9) is higher than Parfit2's level of well-being (−11). Hence, Yu's conjecture that we mistake small amounts of well-being for no well-being in these cases is only plausible when EMDs are much stronger than first-order desires. This does not seem to be true in general: most – perhaps all – cases are ones where a first-order desire is much stronger than the desire to know that the first-order desire is satisfied.
Second, the EMD theory faces a dilemma. Either, necessarily, each desire is accompanied by an EMD, or EMDs merely contingently accompany most desires that most people have. The former view leads to an infinite regress: second-order desires to know that first-order desires are satisfied are accompanied by third-order desires to know that second-order desires are satisfied, and so on. This means that if we have any desires at all then we have an infinite number of desires. On this view almost all of our well-being would derive from the satisfaction and frustration of EMDs, rather than first-order desires. This is just incorrect – most of our well-being comes from the way the world is, not from introspection of our own epistemic states. If we try to resist this implication by saying that EMDs have a very small magnitude, and that their magnitudes decrease exponentially as we ascend to ever-higher orders of desire, then this undermines the response to the prime number cases that motivated the EMD theory in the first place (that response requires EMDs to be strong, not weak) and would thus leave the EMD theorist open to my first objection. Trying to cut off the regress at the first level would be unmotivated, and will not work in any case, because for that we need a criterion of what makes a desire first-order. Perhaps a desire is first-order if it is not about one's own epistemic states. But this is not right: I can desire to know whether a certain team won last night without desiring that they won. This desire is first-order, not higher-order.
Grasping the other horn implies that there could be creatures that had no EMDs whatsoever, though humans simply happen to have EMDs about most or all of their first-order desires. If this is the case, then all of my arguments above can be run for such creatures. Non-human animals may be such creatures. Suppose that a dog wants its owner to be home, but unbeknownst to it, the owner is hiding in the bathroom. The dog does not have the concept of knowledge, so it would be impossible for it to form the relevant EMD: I want to know that my desire for my owner to be home is satisfied. But it seems clear that the dog is not made better-off when its owner is hiding in the bathroom, but only becomes better-off when it learns that its owner is in fact home. The awareness requirement explains this but the EMD theory does not. Overall, the awareness requirement does the same work as EMD theory, but better, and with fewer new ontological commitments.Footnote 12
3. Concurrentism
In this section I will set aside the awareness requirement to focus on issues about the time of benefit. We will bring the two discussions together in the next section. Well-being is a temporal phenomenon. People have well-being at times not atemporally. Coarse-grained inter-temporal judgments about a person's level of overall well-being such as “Jim is better-off this month than he was last month” make sense, just like interpersonal judgements like “Jim is better-off than John.” Likewise for fine-grained judgments about particular welfare benefits. The things that benefit one do so at some times and not others: I'm now benefitting from the satisfaction of my desire to have a cup of coffee, though I did not benefit from this when I was a baby. Indeed, the premise structuring this discussion is that all welfare benefits are temporal: one benefits from P only if one benefits at some time(s) from P.Footnote 13
Discussions among theorists of well-being usually concern which kind of things make one's life go well or poorly. But they must also take up the question of when the things that make one's life go well do so. Hedonists have a natural extension of their theory to offer here: one's well-being in an interval of time T is determined by all of the pleasures and pains one experiences in T. A pleasure benefits one at the times at which it is experienced. Perhaps desire theorists could say something similar: one's well-being in an interval T is determined by all of the desire satisfactions one experiences during T.
This is not a very natural extension of desire satisfactionism in its own right, because it neglects the fact that there are two structural constituents of desires whose persistence conditions can vary independently of one another: the having of the desire and the obtaining of the contents of the desire. One can desire P at times when P does not obtain, and P can obtain at times when no one desires it. Desires and their contents are both temporally delimited entities: they have a beginning, exist for some duration, and then have an end. So, should we say that the person benefits from the satisfaction of their desire for P at the times when they have the desire in question, or at the times when the content of the desire obtains, or something more complicated? There are four answers to this time-of-benefit question which deserve consideration. One benefits from the satisfaction of a desire at:
1. The times at which the contents of the desire obtain (time-of-object view).
2. The times at which one has the desire (time-of-desire view).Footnote 14
3. Between the earliest time at which one both has had the desire and its contents have obtained, and the latest time at which either one has the desire or its contents obtain (asymmetrism).Footnote 15
4. The times at which both the person has the desire and the contents of the desire obtain (concurrentism).Footnote 16
I will present some arguments defending concurrentism vis-à-vis other views.
One argument concerns the relationship between the time at which one benefits and the amount that one benefits. In addition to the time-of-benefit question, there is also the magnitude of benefit question: given that one benefits at t from the satisfaction of one's desire that P, how much does one benefit? A benefit function BD(P) (t) for a given satisfied desire for P, D(P), denotes how much one benefits at t. Note that the benefit is relativized to a desire: it remains an open question, one that I will not address here, how the multitude of desire-relative benefit functions are aggregated to arrive at one's overall benefit at a time. The standard benefit function holds that the degree to which a person benefits at time t from the satisfaction of their desire for P is given by the strength of the person's desire at t. Formally, for all t: BD(P) (t) = SD(P)(t), where SD(P)(t) denotes the strength of the person's desire for P at t.
Desire satisfactionists have to answer the magnitude of benefit question in addition to the time-of-benefit question. The two questions are related.Footnote 17 The standard benefit function is by far the most natural way for desire satisfactionists to answer the magnitude of benefit question. But it is inconsistent with two answers to the time-of-benefit question: asymmetrism and the time-of-object view. To illustrate this, consider a case where the content of someone's desire begins to obtain only after the person has stopped having the desire: there is no overlap between when one has the desire and when its contents obtain. (Note that asymmetrism behaves like the time-of-object view when the times at which the contents of a desire obtain are after the times at which one has the desire.) Not having a desire that P at a time is equivalent to having a desire that P with strength zero at that time. If either the time-of-object view or asymmetrism is true, then one benefits at times during which the strength of one's desire is zero, the times when the contents obtain. This is inconsistent with the standard benefit function, since the strength of the desire and the degree of benefit are out of proportion.
Which should we reject, the standard benefit function or the answers to the time-of-benefit question incompatible with it? We should reject the latter. This is because the standard benefit function is the way to precisify desire satisfactionism that is most faithful to its original motivations. The informal thought behind many desire theories is the resonance constraint, the view that in order for something to benefit one, one must at least be disposed to positively value the thing in question. We can formulate a synchronic version of this constraint too: in order for something to benefit one at a time, one must at least be disposed to positively value the thing in question at that time.Footnote 18 Suppose I wanted to be an astronaut as a child, but not now. If I become an astronaut now, the synchronic resonance constraint (correctly) indicates that I do not benefit from this because being an astronaut does not resonate with me now, even though it used to. Further sharpening this constraint, we can say that the degree to which a thing benefits you at a time must be proportional to the strength of your evaluative dispositions or pro-attitudes toward the thing at the time. If one modestly valued a thing at a time but greatly benefitted from it, or if one greatly valued a thing at a time but only modestly benefitted from it, then this violates the kind of precisified resonance constraint I have in mind. Benefit cannot come apart from something's resonance with you, in time or in degree. That, I contend, is the most natural way to spell out the resonance constraint, and thus, the motivations behind subjectivism.Footnote 19 If degree of resonance came apart from benefit at any time (or in their magnitudes at a time), then we would have to say what factors cause them to be out of proportion. Since desire satisfactionists are subjectivists about well-being, they should be hesitant about positing such factors, which would have to be independent of the person's own evaluative dispositions. The standard benefit function is the only one that ensures that degree of benefit and degree of desire remain in proportion, and that is why desire satisfactionists should not give up the standard benefit function, but rather should reject views about the time of benefit inconsistent with it.Footnote 20
Second, many, including myself, will see the fact that concurrentism immediately rules out posthumous and preconceptional benefits and harms as a key advantage of the view. Concurrentism rules out these possibilities, of course, only on the assumption that one cannot have desires after one dies or before one is conceived, either because one does not exist at these times or because one does not have the psychological capabilities necessary for having desires at these times. Indeed, inquiring into the time-of-benefit question helps us to see the conceptual mistake made by fans of such harms and benefits. Since anything which harms or benefits someone must do so at a time, events that go on after your death or before your birth must benefit or harm you at times during your life, if they benefit or harm you at all. Something cannot benefit or harm you at a time at which you do not exist. Suppose that, as an adult, I desire P, but P began to obtain before I was born. I do not benefit from the satisfaction of this desire before I was born. At best, the first time I benefit is when I first have the desire. Someone cannot benefit or be harmed at a posthumous or preconceptional time, because they do not exist to be benefitted or harmed at such a time, and in order for one to be a welfare subject at a time (indeed, to have any properties or stand in any relations at a time), one must exist at that time.
Unfortunately, both the time-of-object view and asymmetrism entail that you can benefit or be harmed at a posthumous time, since the object of your desire can still obtain at a posthumous time. The only way to avoid this commitment is by adopting an ad hoc modification of either view to restrict the times at which one can benefit from the satisfaction of a desire to the times at which one is alive.Footnote 21 Concurrentism requires no ad hoc stipulations to avoid such an implication, and is therefore the better view.
We have argued thus far against the time-of-object view and asymmetrism, but what can be said against the time-of-desire view? The best arguments against the time-of-desire view will require a somewhat nuanced understanding of the nature of contents of desire. I have been assuming throughout, and will continue to assume, that the contents of desires are propositions. But there is a debate about whether propositions can change truth value over time – tensed propositions can change their truth value over time, tenseless propositions cannot.Footnote 22 <The cat is on the mat> is a tensed proposition, since it changes its truth value as the cat moves about the room, but <the cat is on the mat at t> is a tenseless proposition, since it is true at all times if the cat is in fact on the mat at t.
Most discussions in this domain would be less interesting if we thought of the contents of desire as tenseless. For example, the time-of-object view would amount to the claim that one benefits from the satisfaction of a desire that P at all times, for if P is true at any time, it is true at all times. This view is clearly false: I did not benefit from my desire to drink a cup of coffee one second after the big bang. Likewise, if propositions are tenseless, then the time-of-desire view is equivalent to concurrentism: since propositions are true at all times or none, if a desire is satisfied then the interval of time during which one has the desire is identical to the interval of time during which one has the desire and its contents obtain. And if the time-of-desire view collapses into concurrentism, I have no arguments against this view, since it is true.
However, if propositions are tensed, then the time-of-desire view is false, for it faces compelling counterexamples. Suppose that P obtains at all times before t, and ~P obtains at all times after t. Suppose further that one desires ~P at all times before t, and desires P at all times after t. Both one's desire that P and one's desire that ~P are satisfied (recall that satisfaction, as I am using the notion, is a tenseless matter: a desire is satisfied if its contents obtain at some time). According to the time-of-desire view, the times at which one benefits from the satisfaction of one's desire that ~P are the times at which P obtains, and the times at which one benefits from the satisfaction of one's desire that P are the times at which ~P obtains. At no time in the case does one get what one wants when one wants it. One wants ~P at all times up until t, and is disappointed because one does not get it at any of these times, and just when one does get ~P, one ceases to want it. This is an very disappointing turn of events. One does not have what one wants, then gets it when one does not want it. At no time does it seem like one benefits in this case. One's disappointment will not be alleviated by being told by the time-of-desire theorist that since ~P will obtain sometime in the future, this future state of affairs retroactively benefits one now, even though it does not obtain now when one wants it. Similarly, it will not help you to be told, when you desire P, that you are actually benefitting now because P once obtained in the far past, long before you desired it. Well-being is about getting what you want when you want it. Getting what you want at t at some time other than t at which you are indifferent or even hostile to it is no substitute. Eating when one is full, having sex when one is not aroused or socializing when one feels like being alone do not benefit one, even though one has desires to eat, have sex, or socialize at other times. These events do not benefit one when they occur (as the time-of-object view suggests) or when one wants them at other times (as the time-of-desire view suggests), for they do not benefit one at all.Footnote 23 Thus, the time-of-desire view is mistaken. Having argued against the other plausible views in the literature, I conclude that concurrentism is the best view.
4. Concurrent Awareness Desire Satisfactionism
The two issues that I have considered in this article – epistemic requirements on well-being and the time-of-benefit question – may seem quite independent of one another. This is not the case. In this section, I will present the two issues together. First, I will restate my official view, further specifying three aspects of the view. Then I will explain why the awareness requirement and concurrentism naturally go together, and argue that views which accept one but not the other are implausible.
Two conditions must obtain at a time t in order for one to benefit at t from the satisfaction of one's desire that P.
1. One is aware at t that P.
2. One is aware at t that one desires P.
When these two conditions obtain, I will say that one is aware that one's desire is satisfied. Given the factivity of awareness, the first condition implies that P obtains at t, and the second condition implies that one desires P at t. Three things need to be clarified about this position, in light of the preceding discussion.
First, one's awareness that one's desire is satisfied must obtain concurrently with the desire's contents and one's desire itself. Three things must obtain at a time in order for one to benefit at that time: one is aware that one's desire that P is satisfied, one desires P, and P obtains. One does not benefit if one becomes aware that one's desire is satisfied only after the contents have ceased to obtain, or after one has ceased to have the desire, as the following example illustrates:
Promotion: You desire that your friend get a promotion. The friend gets the promotion, but she does not tell you. You and your friend then have a falling out so severe that you no longer desire that she be promoted. Much later, you learn that your friend was promoted (during the time while you desired that she was promoted).Footnote 24
Here, it is implausible to say that you benefit from your friend's promotion at any time. Since we argued for concurrentism above, the time would have to be while you had the desire and while its contents obtain. But at all such times you lack one of the necessary conditions for the desire satisfaction to benefit you: the awareness that your desire is satisfied. If this awareness comes later, after you've stopped having the desire or after its contents have ceased to obtain, then the awareness is useless. You're never aware, when you've obtained what you want when you want it, that you've obtained what you want when you want it, so you do not benefit.
Second, there must be a match between the guise under which you desire P and the guise under which you are aware that your desire that P is satisfied. I think desires (and all intentional states) have coarse-grained contents which are presented under guises – desire and awareness are both three-place relations between an agent, a coarse-grained content, and a guise under which the content is presented. But it is possible to desire that P, be aware that P, yet fail to benefit because the guise under which you desire that P is not the same as the guise under which you are aware that P, as the following examples illustrate:
Lois Lane: Lois Lane desires to kiss Superman, but has a desire of equal strength not to kiss Clark Kent. She kisses Superman (i.e., kisses Kent while he is in uniform). At the time of the kiss, she is aware that she kissed Superman but unaware that she kissed Kent. Later she learns that Kent is Superman, and thus becomes aware that her desire not to kiss Kent is frustrated.
Perry's Cart: Perry is in the grocery store and notices a trail of sugar leading around the aisles, and figures that someone has a leaky bag. He desires that someone makes such a mess, because he hates this store and is in rivalry with its owner. But he desires that he himself does not make the mess. Perry looks down and becomes aware that he is the one making the mess.Footnote 25
Lois Lane desires that P and desires that ~P, and the reason that this is psychologically possible for her is because their contents are presented under different guises. Perry desires an existentially quantified content which is contingently made true by a specific content that he desires not to be the case. If one desires P under guise G, but is aware that P under guise F (where F =/= G), then one does not benefit from the fact that one's desire for P is satisfied. This fact explains how Lane's and Perry's well-being change over time in the stories above: intuitively, they both start well-off, and then become worse-off. Lane has higher well-being at the start of the story because she is aware that she's kissed a particular man under the Superman guise, making her well-off, and then becomes aware that she's kissed that particular man under the Kent guise, making her worse-off. Likewise, Perry is aware under a generic guise merely that someone has spilt sugar, making him well-off, and then becomes aware under a first-personal guise that he's spilt the sugar, making him worse-off. The desire theorist who is not committed to the awareness requirement cannot explain the change in Lane's and Perry's momentary well-being as the stories progress. But once we adopt an awareness requirement and understand the contents of desires and awareness as presented under particular guises, we are able to explain the changes in their well-being over the course of the stories.
Third, being aware that one's desire that P is satisfied may be something over and above the sum of the states of being aware that one desires P and being aware that P (under the same guise). This is because a person may fail to draw the inference from (1) P and (2) I desire that P to (3) my desire that P is satisfied, because they get distracted, etc. Think about it this way: if one person is aware that they desire P, and a different person is aware that P, this is not sufficient for anyone to benefit, because there is not the right kind of causal connection between the state of awareness that P and the state of desiring P. This causal connection between the two states is necessary for benefit. But if the causal connection between these two states could fail to obtain when the states are located in different persons’ minds, it could fail to obtain when they are located in the same person's mind as well (even if this case is less frequent). So, we must add the qualification that one must draw the appropriate inference giving rise to the appropriate causal connection between the two states, in order to be aware that one's desire is satisfied.Footnote 26
Now I will argue that concurrentism and the awareness requirement naturally go together: views which accept one but not the other enjoy weaker motivations than a view which accepts both. Consider the case of the oppressed author again, but slightly modified:
Oppressed Author: An author works on his manuscript for years, desiring nothing more than for it to be published. He lives in a country with a repressive regime, so he entrusts his completed manuscript to a confidant who smuggles it out of the country to be published. There are two versions of what happens next:
• Version 1: The author loses the desire before the manuscript was published, whether because he moved on to other projects, or he was discovered by the regime and subjected to reeducation and brainwashing, etc. Then, as it is published the author becomes aware of the status of the manuscript, but is indifferent to the fact.
• Version 2: The confidant dutifully publishes the manuscript while the author continues to desire that it is published, but word of the manuscript's publication never reaches him. The author believes that the manuscript is not published.
One who denies concurrentism but accepts the awareness requirement will say that the author benefits in the first version of the story but not the second, and one who accepts concurrentism but denies the awareness requirement will say that the author benefits in the second version but not the first. However, it is not plausible to say that the author benefits in either version. Awareness only matters given that there is concurrence, and concurrence only matters given that there is awareness. In version 1, the only thing that the author can be aware of is a mundane state of affairs which has no significance for him, given his loss of desire. He may be aware that his book was published, but this awareness does nothing for him given that he then has no pro-attitude toward the book's publication. In general, whenever there is not concurrence between the content and the desire, one can be aware merely of the content (as an undesired, ordinary proposition), or one can be aware merely of an unsatisfied desire. On a phenomenological level, these two kinds of defective awareness do not seem to be the kind of thing which would contribute to one's well-being, for they would have either a neutral or a negative valence, respectively. A life full of only such mental states would not be a good one. The more robust awareness that one's desires are satisfied enables well-being, but it is impossible for one to have awareness of this kind when there is not concurrence. This is because the mind and the world do not both simultaneously provide their shares – the mind must provide the desire and the world must provide the content in order for awareness of something to promote one's well-being at a time.
Likewise, if one lacks awareness, then it is impossible for concurrence to matter. If one is unaware of whether one's desire is or will be satisfied, then it would be unimportant to one when the desire is ultimately satisfied. One would have no basis for such a preference: why would it matter to you if P occurred when you desire P as opposed to some other time if you never know whether P obtains? If one is aware, then one would clearly prefer that the contents obtain concurrently with one's desire to non-concurrence, because this concurrence enables the positively valanced mental state of awareness of desire satisfaction. And again, the basis of this preference is absent when one is unaware of whether the content obtains. Here are two other ways to see this point. Consider a case of mild sweetening. Someone offers you the following bargain: (1) have a desire of yours satisfied concurrently with the desire, but you are unaware that it is satisfied, or (2) have the same desire satisfied non-concurrently, but you are unaware that it is satisfied, plus you are given $1. The intuition this case is supposed to elicit is that you should take the mildly sweetened option, implying that you have no reason to prefer that your desires are satisfied concurrently with their contents if you are unaware that they will be satisfied in both cases. Second, think about things in the second person: you know that your friend will always remain unaware of whether or not her desire is satisfied, and you are in a position to decide whether the contents obtain concurrently with her desiring them or non-concurrently. It is clear that you would have no reason to bring about the contents at one time rather than the other, since no one, including your friend, could voice a complaint about one decision or the other.Footnote 27
CADS entails that you have no reason to prefer concurrence without awareness to awareness without concurrence or vice versa. This is the right result – getting what you want at t, at a time other than t when you don't want it and knowing about it is simply no better or worse for you than getting what you want when you want it but not knowing about it. In both cases, to speak metaphorically, the desire's satisfaction is outside of your world – it is too remote to play the role in your life that determinants of your well-being must play. Neither influences the quality of your lived experience in the right way, in the one case because the content is undesired, in the other because it is unknown. By contrast, if concurrentism holds but the awareness requirement does not, then one should prefer concurrence between the desire and its content without awareness to awareness without concurrence. And vice versa if the awareness requirement holds but concurrentism does not. These are the wrong results, for there is no reason to prefer one of these to the other.
5. Conclusion: CADS and the Metaphysics of Well-being
According to CADS, well-being only exists when the right relationship obtains between mind and world – straddling the distinction between mental statist and non-mental statist views in the philosophy of well-being.Footnote 28 Well-being is grounded both in the mental and in the non-mental. To be well-off, there has to be the right kind of match between one's attitudes with two different directions of fit. You desire that P (world to mind), but are not well-off until you also are aware that your desire that P is satisfied (mind to world). But not just any combination of these attitudes will do – the examples that we considered above show that in order to be well-off, the guise under which one desires P must be the very same as the guise under which one is aware that P. And since well-being is a temporal phenomenon, these relations between mind and world – as well as the relata that they relate, the subject's desire and the content of the desire – must all obtain simultaneously in order for the subject to be well-off. So the right kind of intentional, causal and temporal relations must obtain among the contents, one's awareness, and one's desire in order for one to be well-off. This manifests an organic unity. On their own, neither the worldly component (P) nor the mental components (awareness that one's desire that P is satisfied and one's desire that P) of the unity could make one well-off. But when they fit together in the right way, the unity has value though none of the parts do.Footnote 29
It is important to see why my view is not a version of what might be called veridical hedonism. This is the view that it is only pleasure that promotes well-being, but that a pleasure contributes to one's well-being only if it is taken in an actually obtaining state of affairs.Footnote 30 Admittedly, I have used language that elicits hedonist intuitions when describing the state that someone is in when they are aware that their desire is satisfied. Indeed, though I did not argue for this above, since I am an attitudinal theorist about pleasure, I think that pleasure just consists in such a state, and it is impossible to be aware that one's desire is satisfied but not be pleased.
The important difference between CADS and veridical hedonism is a difference in where the views locate welfare value.Footnote 31 Veridical hedonism locates value in the state of awareness that one's desire is satisfied, the site of pleasant sensations and feelings of satisfaction. This state is rendered valuable only when it corresponds to something real – meaning that the relations it bears to other states (states in the world) bestow value upon it. And when the pleasurable sensations fail to stand in the appropriate relations to states of the world, they are not valuable. But I, following Moore (Reference Moore1903), do not think that the value of a thing can depend on the relational properties of that thing. Moore, of course, intended this constraint to apply to value simpliciter, not welfare value, for he was famously suspicious of that notion. Nevertheless, the Moorean constraint should be taken up by theorists of well-being as well.Footnote 32 Unlike veridical hedonism, CADS is consistent with the Moorean constraint. On my view value does not reside in the state of awareness that one's desire is satisfied, but rather in the mereologically complex state of affairs which has as components the state of awareness, the agent's desire and the desire's content. This complex state of affairs is what the subject's well-being consists in, though none of the parts on their own give the subject any well-being. It is the unity which is fundamentally valuable, and the state of awareness, the desire, and the contents are valuable only in virtue of being parts of this unity, not the other way around. None of the components would be valuable if the others did not exist or were not related in the right way. On my view, value is not localized in the state of awareness that one's desire is satisfied (a mental component), nor is it located in the content one desires (the worldly component). Rather, when the parts of the unity are related in the right way, it is distributed homogenously throughout the desire, its content, and one's awareness of it.Footnote 33
