Desire-Satisfaction Theories of Welfare

---

Key Concepts

Basic or intrinsic desire

Instrumental or extrinsic desires

Defective desires (all-things-considered versus intrinsically defective)

An excerpt from Christopher C. Heathwood, “Desire-satisfaction theories of welfare.” (2005). Doctoral Dissertations 1896 – February 2014. 2395.

Why Some Are Attracted to a Desire Approach to Welfare (Ch. 1, 1.3)

Roughly, the desire-satisfaction theory of welfare says that one’s life is going well to the extent that one is getting what one wants. The desire-satisfaction theory may currently be the most popular view of well-being among philosophers, political scientists, and economists… 

Many philosophers are liberal minded. They value personal autonomy; they think people should be allowed to do what they want with their lives, to be free to pursue their own idea of the good life.” Some theories of well-being, however, seem rather paternalistic. These theories say: “I know what’s good for you, whether you like it or not. It’s pleasure, or it’s acting in accord with your proper function, or it’s the life of philosophical contemplation.” Some like the desire- satisfaction theory because they see it as unlike this. “You don’t want pleasure?,” the theory asks. “Well, that’s ok; you can still do well without it. In fact, why don’t you just pursue whatever you want. Who are we to say what’s good for you?” The thought is that desire satisfactionism is “in tune with the liberal spirit of the modern age, which tends to see human agents as pursuers of autonomously chosen projects” (Sumner 1996, p. 123).

***

The Problem of Defective Desires (Ch. 2)

The desire-satisfaction theory of welfare says, roughly, that one’s life goes well to the extent that one’s desires are satisfied. On standard “actualist” versions of the theory, it doesn’t matter what you desire. So long as you are getting what you actually want – whatever it is – things are going well for you. There is widespread agreement that these standard versions are incorrect, because we can desire things that are bad for us – in other words, because there are “defective desires.”

The Problem of Defective Desires

Cases abound in which it is bad for a person to get what he wants. We have ill-informed desires. Suppose I’m thirsty and therefore desire to drink from the river. Suppose also that the river is polluted and that drinking from it would make me sick. (Carson 2000: 72-73). Clearly, it would be bad for me to drink from the river, even though it is something I want to do. If only I had more information, I would no longer desire to drink from the river, and would consequently be better off. 

We have irrational desires. Suppose I need to see a dentist for a procedure that will require drilling. And suppose this time lack of information isn’t my problem: I know it is in my long-term interest to see the dentist. But still I want no part of it. Clearly, it would be bad for me not to go, even though it is something I want. If only I weren’t weak-willed, I would come to desire to see the dentist, and would consequently be better off. (Schwartz 1982, p. 196).

We have base desires. Suppose my strongest desires would be satisfied by a perpetual indulgence in bestiality (Moore 1993, pp. 146-147). Some venture to say that any theory that implies that the satisfaction of such desires is good is as false as it is paradoxical. Actualist desire theories are thought to have this implication.

We have poorly cultivated desires. Suppose I prefer Muzak over Mozart, mashed potatoes over masterpiece cuisine, and pushpin over poetry. It is thought that I would be better off if I desired the better things, and got them, than if I desired and got the mediocre things I now desire (Schwartz 1982, p. 196).

These last two objections are similar. The idea behind them might be (though need not be) that certain things are somehow inherently more worthy of desire than other things, and this desire-worthiness translates into enhanced welfare when the worthy things are gotten. The objects of base desires are at one end of the spectrum, such that getting them is positively bad; the objects of noble desires are at the other end, and getting them is super; somewhere in the middle are the mediocre things, and getting them isn’t so great. That we put effort into cultivating a taste for certain things we think worthy, and that we encourage some desires over others in our children and other loved ones, suggests that the idea that the objects of desire can vary in worth is common.

Yet another variation of this same idea is found in the possibility of pointless desires. Suppose my only desire is “to count blades of grass in various geometrically shaped areas such as park squares and well-trimmed lawns” (Rawls 1971, p. 432). All I want is to count the blades of grass, all day, every day. It is thought that I would be better off if I desired to do something more worthwhile, and did that, than if I did the utterly pointless thing I now desire to do. 

Continuing the litany, we have artificially aroused desires. Suppose as a result of a late-night indulgence in infomercials, I want nothing more than to own a Flowbee Vacuum Haircut System. Suppose the desire is artificially aroused: I have it only due to the in-your-face marketing campaign’s effect on my insomniatic mind. It is tempting to think that the satisfaction of such desires does not benefit their subject.

Finally, we desire to be badly off.  Suppose that, guilt-ridden by past crimes I seek to punish myself by taking an arduous, boring, and insignificant job (Kraut 1994).  Suppose I land one and become pretty miserable. Clearly, it is bad for me to get the job, even though it is what I want. If I were less susceptible to guilt, I would have no desire for such a job, and would consequently be better off…

The Actual Desire-Satisfaction Theory

What is the theory thought to be so refuted by the problem of defective desires? …

All parties to the debate seem to agree that any desire-satisfaction theory must count only basic (or intrinsic) desires (Brandt 1979, p. Ill; Parfit 1984, p. 117). If I desire to turn on my CD player only because I desire to hear the Pixies, I am made better off only if the latter, basic desire is satisfied. My desire to turn on my CD player is merely instrumental (or extrinsic): I have it only because I desire something else, and I think the thing instrumentally desired will lead to that something else. The theory is therefore interested only in “basic desire satisfaction” – that is, the satisfaction of one’s basic desires. Intuitively, a basic desire is a desire for something for its own sake, not merely for something else that it will lead to, or otherwise realize. 

We can take this theory to be a paradigmatic version of an actualist desire-satisfaction theory, and the target theory of the problem of defective desires.

The theory is actualist rather than idealist, since it counts towards welfare the satisfaction of one’s actual desires, not the desires one would ideally have. The theory is summative rather than global : it implies that the amount of welfare in a life is obtained by summing over all the satisfactions in the life, not just the satisfactions of one’s “global desires” (the desires about one’s whole life, or about a largish part of one’s life). ⁴ The theory is a satisfaction version rather than an object version of the desire theory, since it is desire satisfactions rather than the objects of desire that are intrinsically good and bad for a subject (Rabinowicz and Osterberg 1996; Bykvist 1998). Importantly, I mean the theory above to require concurrence: in order for a state of affairs to count as a genuine instance of desire satisfaction, the state of affairs desired must obtain at the same time that it is desired to obtain. If I desire fame today but get it tomorrow, when I no longer want it, my desire for fame was not satisfied. A desire of mine is satisfied only if I get the thing while I still desire it, and continue to have the desire while I’m getting it… It doesn’t matter, welfare-wise, how long I desire something before I get it. All that matters are the following: that I still desire it while I’m getting it; how long the concurrent desiring and getting last; and how intense the desire is while it is being satisfied. 

The Form of The Argument 

We can abstract away from the particulars of the types of argument outlined in Section I and be left with the general form of the argument from defective desires: 

1. There are defective desires: desires whose satisfaction does not make their subject better off.
2.  But if the actualist desire-satisfaction theory is true, then there are no defective desires.
3.  Therefore, the actualist desire-satisfaction theory is not true._. 

If these premises are true and univocal , actualist forms of the desire-satisfaction theory are done for. Call this the ‘Main Argument’. 

Though the Main Argument captures the structure of all the arguments suggested above, the arguments subdivide reasonably well into three categories: (i) arguments from all-things-considered defective” desires, (ii) arguments from “intrinsically defective” desires; and (iii) a third category to be named later. 

Arguments from All-Things-Considered Defective Desires 

Some things are intrinsically bad for people. It is the job of a theory of welfare to identify these things. According to hedonists, episodes of pain are intrinsically bad for people. According to actual desire-satisfaction theorists, desire frustrations are intrinsically bad for people. In each case, the allegedly bad items are bad in themselves, rather than for what they lead to. The items would still be bad even if they led to nothing else bad.

Other things are bad for people in a different way. They are bad not in themselves but for what they lead to, or what they prevent.  Many desire satisfactions are like this – e.g., many ill-informed desires and irrational desires of the sorts described above (and likewise for some of the other kinds of desire catalogued above, but more on that later). Recall my ill-informed desire to quench my thirst by drinking from the polluted river. There is nothing intrinsically or necessarily bad about having my thirst quenched. The desire is defective merely accidentally: its satisfaction just happened to be bad overall, in the long run, due to the fact that the river happened to be polluted.

Let’s say that if something is bad in this way, then it is all-things-considered bad for the subject… Given this simple distinction, it is not difficult to see how actualist desire- satisfaction theories have the resources to deem ill-informed and irrational desires defective. Despite the fact that the theories declare all desire-satisfactions to be intrinsically good for their subject, the theories can nevertheless also declare some desire-satisfactions to be bad — all-things-considered bad. Since intrinsic badness and all-things-considered badness can come apart, the theory is consistent with – indeed, it entails – the claim that many desire-satisfactions are all-things-considered bad. To make this more clear, let’s take a closer look at the examples. 

The Argument from Ill-Informed Desires 

Assume my desire to quench my thirst is a basic desire…We all agree that this ill-informed desire is defective because, were I to satisfy it, I would be worse off than I would be if I did not satisfy it. The future I would have if I did satisfy this desire would start out pretty good – my thirst would be quenched. But it would quickly turn bad as sickness set in. The future I would have if I chose not to satisfy the desire to quench my thirst would start out not so great – I would be thirsty. But that’s about as bad as it would get (at least as far as this case goes), since I would avoid illness. All told, the life I would lead were I not to quench my thirst then and there – call it ‘L1’ – is better than the life I would lead — call it ‘L2’ — were I to quench my thirst then and there. 

These are the facts. And actualist forms of the desire-satisfaction theory fully accommodate them. The actualist theory entails that L1 is better than L2 because, due  to the sickness suffered in L2 (and despite the extra desire satisfaction (the thirst- quenching within the hour) in L2), L1 contains a more favorable balance of satisfaction over frustration. It avoids those desire frustrations that accompany being ill, whose negative value outweighs the positive value of satisfying the desire to be quenched within the hour. Far from being a counterexample to actualist theories, this case of an ill-informed defective desire actually provides confirmation to actualist desire-satisfaction theories of welfare. 

So we say that my desire to quench my thirst in this example is all-things-considered defective. If the concept of all-things-considered defectiveness is the one being used in the Main Argument above, then premise (2) is false: according to the actualist desire-satisfaction theory, there are indeed defective desires. 

The other concept was that of intrinsic defectiveness, the idea of something being bad in itself for a subject, whether or not it happens to lead to anything else bad or prevent anything good. So long as we attend only to basic desires (as any desire-satisfaction theory requires), it is not plausible to say that merely ill-informed desires are intrinsically defective. If I intrinsically desire to quench my thirst right now, and the desire is satisfied, this is intrinsically good for me. Saying this is consistent with saying that it is not in my best interest to quench my thirst, that I ought not to quench my thirst, and that it would be all-things-considered bad for me to quench my thirst. Once we see what is meant – and what is not meant – by saying that the satisfaction of the desire is intrinsically good for me, there is no compelling reason to deny it. 

And there are compelling reasons not to deny it. Consider what it is to say that my desire to quench my thirst is intrinsically defective. To do this, we need to compare initial segments of two possible lives. First imagine the initial segment of LI, the life in which I do not quench my thirst. These last few minutes of SI are spent with a parched throat and a longing for liquid. Now imagine the corresponding initial segment of L2, the life in which I quench my thirst. This segment - call it ‘S2’ - is just like S1 except that I choose to satisfy my desire to be quenched. The last few minutes of S2 are spent with a wet whistle and little longing. Importantly, S2 ends before any pollutant-induced illness sets in.

***

Arguments from Intrinsically Defective Desires 

Sometimes critics of actualist desire theories have in mind an objection that is different in kind from those so far discussed, despite the fact that we can classify it under the same name. Some critics think that some desires are “intrinsically defective.” The claim is that, for some desires, a person can be made worse off by having them satisfied, not because the satisfaction leads to a lower net balance of satisfactions, but simply because the satisfaction is bad (or not very good) in itself. The case of base desires is perhaps the most common example.

The Argument from Base Desires 

In a colorful passage, G.E. Moore presents an argument from base pleasures in an attempt to refute a hedonist theory of value (1993, pp. 146-7). But the point can easily be modified to apply to actualist desire-satisfaction theories of welfare. Modified Moore has us imagine a scenario in which people have no desires for the refined – no desires for the “contemplation of beauty” or for “personal affections” – but in which instead the strongest possible desires are satisfied “by a perpetual indulgence in bestiality” (1993, p. 147). The upshot is that the advocate of an actual desire-satisfaction theory is forced to maintain, absurdly, that “such a state of things would be heaven indeed, and that all human endeavors should be devoted to its realization” (1993, p. 147).

Another provocative case (due to Feldman 2002) provides the basis for a similar argument. Suppose a terrorist hates children and wants nothing more than to see them suffer; suppose he devotes his life to an ongoing terror campaign, and succeeds. None of us is inclined to think the terrorist’s life is worth emulating, but an actual desire view must judge it to be a very good one. 

An actualist reply begins by examining the cases more closely. In the Moorean example, a series of claims is made about the desire-satisfaction theory. It is claimed that (a) the theory entails that the scenario of perpetual indulgence so described would be very good, and in fact much better than the current state of the world. It is also claimed that (b) according to the desire-satisfaction theory, we should see to it that such a scenario is actualized. Finally, it is suggested, though not explicitly stated, that (c) the people in such a scenario would be very well off according to the desire-satisfaction theory. 

It is important to distinguish these claims because, in my view, only one of them – (c) – is true. The desire-satisfaction theory, as a theory of welfare, itself implies nothing about the value of complex scenarios containing many welfare subjects. Theories of welfare do have implications for how well things are going for each subject in Moore’s indulgence scenario. But it is a step to go from the claim that everyone is some scenario is doing well to the claim that the scenario itself is good. The step must be bridged by a principle about how to compute the value of “situations,” or possible worlds. …Since the desire-satisfaction theory of welfare contains no such principles, claim (a) is false. An argument based on it therefore won’t get off the ground.

Likewise, the desire-satisfaction theory of welfare does not imply that we ought to devote all of our efforts to the actualization of the indulgence scenario. Even if we stipulate that the indulgence scenario is the best scenario available to us, a theory of welfare still does not imply that we ought to actualize it. If theories of welfare have any implications about behavior, they imply only what we prudentially ought to do. But since sometimes I prudentially ought to do what I morally ought not to do, and sometimes I morally ought to do what I prudentially ought not to do, theories of welfare on their own imply nothing about what we morally ought to do. So claim (b) above is false, and any argument against an actualist desire-satisfaction theory based on it will be unsound.

Theories of welfare do indeed have implications about how well off a perpetual indulger in bestiality would be. Since it is possible for a person to have strong desires in that direction, and have them satisfied, and satisfied without concomitant frustrations, actualist desire-satisfaction theories of welfare do indeed imply that things can be going very well for a perpetual indulger in bestiality. But I maintain that the actualist desire-satisfaction theories deliver the correct verdict here. 

To see why, we need to get very clear on what we’re imagining vis-a-vis the thought experiment, and on what we are – and are not – saying in claiming that the perpetual indulger has a high level of welfare. For the Argument from Base Desires to work, it would need to establish that the desire (in humans, at least) for sex with animals is intrinsically defective. If the claim is merely that such a desire is all-things-considered defective (because, say, the indulger would get bored or lonely, or would come to feel extreme shame or guilt, or would come down with some barnyard diseases), then the actual desire-satisfaction theory will agree. To be sure our judgement is not clouded by such considerations, we should imagine a perpetual indulger who would suffer no such maladies. We need also to imagine a person who has genuine desires to have sex with animals, not a person who finds it disgusting but is forced to do it. So it’s probably no good to imagine yourself in the barnyard, with all of your current desires. The subject of our Moorean thought experiment must want and get sex with animals, and without large amounts of ensuing desire frustration. 

What are we not saying if we agree with the desire-satisfaction theory that the perpetual indulger does have a high level of welfare? We are not saying that it is good simpliciter that he is doing what he is doing. The desire theory of welfare says nothing about what is good simpliciter_, only about what is good for a subject. It is coherent to say that every indulger in Moore’s heaven is well off even though it is a very sad state of affairs.

In short, it is perfectly consistent for an actual desire- satisfaction theorist to issue the following judgment about the perpetual indulger: “What a pity! Sure, he’s well off there in the barnyard, happy doing his thing, getting just what he wants, but his life is pathetic: he will achieve nothing; what he does is degrading; and his moral character is woefully underdeveloped. I would not wish this life upon anyone.” 

***

We should acknowledge one final factor that might be judgment-clouding: that we find sex with animals disgusting. But that we are viscerally repulsed by some state of affairs does not mean we have good reason to condemn it. Indeed, I think it means we must be very careful not to jump to prejudicial conclusions without giving the matter some thought in a cool moment. 

I acknowledge that the disfavor commonly felt towards a life like the perpetual indulger’s may survive reflection. But the reason, I submit, that we react strongly against a life of perpetual indulgence in bestiality is that we care about more than just welfare. We don’t merely want ourselves (or those we love) to be well off; we also want (or want them) to do good things, to be good people, to achieve worthwhile goals. A life of perpetual indulgence – in bestiality or whatever – lacks the other elements that we care about. 

We have teased out what is and what is not implied by the desire-satisfaction theory. The theory implies that the desire for a perpetual indulgence in bestiality is not intrinsically welfare-defective. That is, it implies that the satisfaction of this desire, on its own, augments its subject’s welfare. It is consistent with the desire theory that the satisfaction of this desire is…is not in the subject’s long term best interest, and diminishes the subject’s virtue and dignity. Once all this is pointed out, the Argument from Base Desires loses its thunder. The perpetual indulger is pleased to be with the animals, is getting what he wants, is happy about it all. Given this and everything else that we have said, the claim that the satisfaction of this desire contributes positively to his welfare does not seem implausible. 

Let me return briefly to the case of the terrorist, because it has elements that may not be present in the Moorean thought experiment. For instance, just about everyone agrees that the terrorist is a terrible person and that what he is doing is morally reprehensible. A parallel judgment about the perpetual indulger is far more controversial. 

The desire-satisfaction theory of welfare implies that things are going well for the terrorist as he delights in the children’s suffering. Again, I submit that this is the correct verdict. But let’s remind ourselves what this verdict is not. By granting that the terrorist is well-off, the desire-satisfaction theory is not saying that his being well-off is good; it is not saying that we ought to let the terrorist do what he is doing; it is not saying that the terrorist’s life ranks high on the other scales. The theory is just saying that his level of welfare is high as he continues to get what he wants. But isn’t this the correct result? I think one reason we are so outraged at the situation is that the terrorist is well off despite that fact that he’s such a horrible person. We find the injustice of it all maddening. Any theory that delivered the result that the terrorist was not doing well could not explain this injustice (Lemos 1994, pp. 43-44; Goldstein 1989, pp. 269, 271). 

I conclude that if the Main Argument is interpreted as being about intrinsically defective desires (that is, intrinsically welfare-defective desires), then premise (1) is false: there is no such animal. 

Bibliography

---

Question for further thought…

With respect to the bestiality case, Heathwood makes an important distinction between the level of welfare of the subject and the level of goodness of that state of affairs. He argues that it’s coherent to say that “every indulger in Moore’s heaven is well off even though it is a very sad state of affairs.“

Do you think it makes sense to regard something as a sad state of affairs if there isn’t anybody that it is bad FOR?