I. ENUMERATIVE AND EXPLANATORY THEORIES

At least since Parfit's discussion,Footnote ¹ theories of well-being have been divided thus: hedonism, desire-fulfilment theory, objective-list theory. Even allowing that taxonomies are interest-relative, we might wonder whether this is the best taxonomy of theories of well-being. One reason for thinking otherwise is that it does not take account of a distinction, highlighted by Crisp, between enumerative and explanatory theories of well-being. I will use this distinction as follows.Footnote ² Enumerative theories of well-being specify which things enhance well-being. Explanatory theories aim to explain why something enhances well-being.

Whilst hedonism and objective-list theories are enumerative – they specify an informative list of contributors to well-being – the desire-fulfilment theory is an explanatory theory. The desire-fulfilment theory tells us the conditions under which something is good for someone – namely when it fulfils one of the person's desires. Thus, unlike hedonism and objective-list theory, desire-fulfilment theory does not provide an enumerative account of which things are good for us. This is understandable because providing such an account would be difficult and unnecessary. It would be difficult because the enumerative account would be a list of everything that we desire (or desire under certain circumstances). It would be unnecessary because, even though the desire-fulfilment theory is explanatory, it tells us how to find out what is good for someone; we just need to find out what they desire.

Those who hold the desire-fulfilment theory should not be perturbed by the claim that their theory is an explanatory one. This is no objection to it and I expect desire-fulfilment theorists to be happy with their view being classified this way. Evidence for their likely agreement comes from their presentations of the view. A standard characterization is thus:

This is clearly an expression of an explanatory theory. And I take this to be representative of how desire-fulfilment theorists regard their view.

It is worth pausing briefly to tackle the likely suggestion that we could collapse this distinction between enumerative and explanatory theories. Someone might try to do this by arguing (for example) that desire-fulfilment theories make an enumerative claim, namely that the things that are good for someone are all and only the things that fulfil their desires.Footnote ⁴

Of course, this is a possible way of presenting a view. But we should resist this attempt to collapse the distinction for a number of reasons. An enumerative version of the desire-fulfilment theory (namely the claim that all and only the things that a person desires are good for her) gets us no nearer to truth about which things are good for someone than the desire-fulfilment explanatory theory does on its own. If we know that the things that are good for someone (whatever they may be) are so because they fulfil her desires, we learn nothing new from the so-called enumerative theory that all and only the things that fulfil her desires are good for her. This purported enumerative theory is a spare wheel.

The only way to make this enumerative version of the theory informative would be to add a claim about which things we desire. For example, if the person claimed that we only desire (e.g.) knowledge, we can use this to home in on the truths about what would enhance someone's well-being – because we know that all and only the things that he desires are good for him and that he only desires knowledge. But all of the work in making the enumerative element informative is being done by the separate claim that we only desire knowledge.

A second reason not to collapse the distinction between enumerative and explanatory theories is that the distinction is one with which we are perfectly comfortable, and can see the value of, elsewhere. The same distinction distinguishes the Ten Commandments (an enumerative list of the actions that are wrong) from the divine-command theory (an explanatory theory of what makes something wrong). I take it that no-one thinks it improves our thinking about these issues to group these theories together by treating the divine command theory as a theory of which things are wrong.

A third reason why we should not collapse the distinction between enumerative and explanatory theories is that one might hold an enumerative theory whilst holding that there is no true explanatory theory. An example of this would be a hedonist who held that all and only pleasure is well-being enhancing but who thought that this was a fundamental fact about well-being, one not (non-trivially) explained by any further fact such as pleasure's fulfilling the human function, or desires, or anything else.Footnote ⁵

Finally, there is also a pay-off to treating desire-fulfilment theories as purely explanatory, namely that doing so creates a parallel with desire theories of reasons. These are not enumerative accounts that specify the reasons that people have but, rather, theories of what explains why someone has a reason to ϕ (or of what must be the case for someone to have a reason to ϕ). Such theories claim that in a case where A has a reason to ϕ, A has a reason to ϕ because ϕ-ing fulfils one of A's desires.Footnote ⁶

Noticing that the desire theory of reasons has been construed as a purely explanatory theory should undermine any residual sense that classifying the desire-fulfilment theory of well-being as explanatory is any weakening of the view. And if we construe desire-fulfilment theories of well-being as explanatory theories we create a helpful isomorphism between desire theories of reasons and of well-being.Footnote ⁷

One final clarification: whilst I mostly talk in terms of a contrast between enumerative theories and explanatory theories, there are of course theories of well-being that have both enumerative and explanatory elements. For example, Aristotle's theory of well-being contains an enumerative element that lists the constituents of human well-being and an explanatory element – namely the claim that there is a necessary connection between human well-being and the human function.Footnote ⁸ Nothing I say here should be taken to suggest that theories of well-being are exclusively explanatory or enumerative. Rather, some theories of well-being are best thought of as making enumerative claims, some are best thought of as making explanatory claims, whilst others make both kinds of claim.

II. HEDONISM, OBJECTIVE-LIST THEORIES AND ATTITUDE-DEPENDENCE

So far I have argued that desire-fulfilment theory is an explanatory theory of well-being and that hedonism and objective-list theories of well-being are enumerative. This moves us from the standard taxonomy to the following:

Seeing that the objective-list theory is an enumerative theory helps to undermine some of the reasons why objective-list theories are rejected, reasons that stem from thinking that there is a deep difference between hedonism and objective-list theories as a class.Footnote ⁹

Having made this first change to the well-being taxonomy by introducing the enumerative/explanatory distinction, we should now ask whether it is correct to follow the old taxonomy in distinguishing between hedonism and objective-list theories. The distinction between hedonism and objective-list theories cannot plausibly be a distinction between monistic and pluralistic theories, for example, as then monistic theories would end up being classed as forms of hedonism independently of their content. Nor is the distinction between hedonism and objective-list theories a distinction between theories that meet the experience requirement and theories that do not.Footnote ¹⁰ Whilst some objective-list theories do not meet the experience requirement, others do. So we cannot differentiate hedonism and objective-list theories on this ground.

In fact we should not distinguish between hedonism and objective-list theories in our taxonomy of theories of well-being. Hedonism just is an example of a monistic enumerative theory and the label ‘objective-list theories’ just picks out all of the other enumerative theories. A similar claim – that hedonism just is an objective-list theory with a short list – has been made before, but some may still be sceptical. For this reason, I will take the time to show that there is no deep distinction between hedonism and objective-list theories.

Someone who thought that we should distinguish between hedonism and objective-list theories will probably claim that there is a significant difference between the two along the following lines: objective-list theories, unlike hedonistic theories, are attitude-independent, elitist or autonomy-violating. Though this is prima facie plausible, it is mistaken. These considerations cannot distinguish between hedonism and objective-list theories, for they apply to both or neither. The first task is to understand exactly what the objection is.

One might think that the objection to objective-list theories is that they are elitist or autonomy-violating. But this is a mistake. An objective-list theory is simply a list of which things contribute to well-being. What we should do with respect to those who do not care about or desire some element from the objective-list is a separate issue, one on which all theories of well-being are agnostic.Footnote ¹¹

The best way to understand the attitude-independence thought is as claiming that objective-list theories are distinct from hedonism because objective-list theories hold that whether something is good or bad for someone is independent of the attitudes that the person takes towards the thing. We can formulate this as the claim that hedonism and objective-list theories differ with respect to:

Someone might claim that whilst objective-list theories are inconsistent with AD1, hedonism is consistent with it.

However, hedonism violates AD1 also. Thus AD1 cannot be used as a way of distinguishing hedonism from objective-list theories. To see why, suppose that we take an objective-list theory that claims that knowledge is the only thing that contributes to well-being. Clearly on this theory whether something is good or bad for someone is independent of the attitudes that the person takes towards the thing. The same is true of hedonism, however. Take an ascetic who seeks to avoid feeling pleasure for fear of corrupting his soul or angering his deity. Hedonism claims that all and only pleasure is good for someone and is thus committed to claiming that if this person were to experience pleasure then this pleasure is good for them, despite the fact that the person actively does not want this thing.Footnote ¹² Both hedonism and objective-list theories violate AD1. They both hold that whether something is good or bad for someone is independent of the attitudes that the person takes towards the thing.

Someone might respond: ‘hedonism combined with the desire theory of pleasure makes well-being attitude-dependent in a way that objective-list theories do not. Thus there is a real distinction between hedonism and objective-list theories.’ But this is too quick. Even if combined with a desire theory of pleasure, hedonism still violates AD1.

The desire theory of pleasure is a claim about the nature of pleasure and it can take one of two forms. The first version is the claim that a sensation S is a pleasure when S is desired.Footnote ¹³ The second version is the claim that a pleasure is a compound state constituted by some sensation S and a desire for S. On both views it is true that whether some agent is experiencing pleasure is partly determined by their desires. This means that pleasure is attitude-dependent on these theories. But this does not make hedonism attitude-dependent in a sense that divides it from objective-list theories. On the hedonistic theory, pleasure is good for you even if you have no pro-attitude towards it.

To see why, consider the second version of the desire theory of pleasure, the one that claims that pleasure is constituted by some sensation and a desire for that sensation: <sensation S + desire for sensation S>. The hedonist theory of well-being claims that this composite item, which is a pleasure, is good for someone whether or not she has a pro-attitude towards it. If the ascetic from the earlier example is experiencing pleasure, he is in this type of composite state, according to the desire theory of pleasure, and this state is good for him, according to hedonism, despite the fact that he has a second-order desire not to be in any such state.

For this reason, hedonism is no different from objective-list theories with respect to AD1. Both specify that certain things are good for people, independently of their attitudes towards those things. And this remains the case even if the hedonist also holds an attitudinal theory of pleasure.

Someone might respond that there is a different way in which the hedonistic theory – unlike the objective-list theory – is attitude-dependent, namely that certain counterfactuals where people lack pro-attitudes come out as true on the view. For example, take a person – A – who is experiencing pleasure playing football. A has a combination <S + pro-attitude towards S> and so is experiencing pleasure; on the hedonist theory of well-being, this is good for him. Someone might, rightly, point out that in the situation where A does not have this pro-attitude there would be nothing enhancing his well-being there and she might conclude from this that hedonism, unlike objective-list theories, does not make well-being attitude-independent.

This suggests a different form of attitude-dependence:

Someone might argue that we can distinguish hedonism from objective-list theories on the grounds that hedonism is consistent with AD2 whilst objective-list theories are inconsistent with it.

Again this is too quick. We cannot distinguish hedonism from objective-list theories using AD2 because objective-list theories can also be consistent with AD2. Take an objective-list theory with friendship on the list. Given that friendship is at least partly constituted by positive attitudes, the same kinds of counterfactuals come out as true on this theory as for hedonism. If someone has a friendship then that is something that is good for her, and in the counterfactual situation where the person lacks the pro-attitudes that constitute friendship this good is absent. This form of objective-list theory, by respecting AD2, is attitude-dependent in the same way that hedonism can be. Thus the kind of attitude-dependence suggested by AD2 does not distinguish hedonism from objective-list theories.

For these reasons we should not think that there is a significant distinction between hedonistic theories and objective-list theories of well-being. And in order to draw the most interesting distinction, namely between these theories and desire-fulfilment theories, we should simply classify both as enumerative.

It might help to take a step back from the details and look again at the bigger picture. My suggestion is that hedonism and objective-list theories should not be distinguished fundamentally in the taxonomy of theories of well-being. I imagined someone claiming that there is a deep distinction between the theories, namely that objective-list theories do, and hedonism does not, hold that whether something is good for you is independent of your attitudes in some sense. I then showed that this is not true (even if hedonism were combined with the desire-theory of pleasure). Hedonism and objective-list theories are both attitude-independent, by violating AD1, and both can be attitude-dependent, by respecting AD2.Footnote ¹⁴ Thus, in order to draw the most illuminating and important distinction – that between these theories and explanatory theories like the desire-fulfilment theory – we should classify both as enumerative in the following manner.Footnote ¹⁵

In trying to find the truth about well-being we should have a choice between two routes. The first route is that of establishing that some explanatory theory is correct. If we could establish that (e.g.) the desire-fulfilment theory is true, that would tell us an important truth about well-being and one that would enable us to do without an enumerative theory (though if there was one that would be an interesting bonus). Another approach is to develop enumerative theories.Footnote ¹⁶ My aim in the rest of the article is to make some progress down the second route. I do this by proposing a new, pluralist, enumerative theory and by showing that a major motivation for the competitor theory – the desire-fulfilment theory – is captured by my theory without incurring the costs associated with the desire-fulfilment theory.

III. A NEW ENUMERATIVE THEORY

The enumerative theory that I propose claims that well-being is enhanced by:

This leaves many issues open. For example, it includes no claims about the relative importance of each of the items. This is important because an under-appreciated fact about enumerative theories (perhaps because there are few of them on the market) is that even if we agree on such a list in outline, important questions remain. One could hold that all of these goods contribute equally to well-being and that some degree of any one of them is sufficient for a good life.Footnote ¹⁷

Alternatively, one might think that these goods contribute unequally to well-being such that some of them are necessary for the life to reach the threshold of being a good life for the one that lives it, irrespective of how much one has of the other goods. Dealing with these issues is a job for another paper, so nothing that I say here is supposed to take a stand on these kinds of issues. Another issue that I leave aside here is whether we should think of the list as a list of conditional or unconditional contributors to well-being.Footnote ¹⁸

Note that the theory includes only the basic elements of the positive side of well-being, so this is really only half of an objective-list theory of well-being.Footnote ¹⁹ Though the negative is important, because of constraints of space and because the literature has tended to focus on the positive determinants of well-being, I focus on those here.

Looking at the list, someone might complain that these items are all subjects of detailed philosophical investigation into their nature and that, for this reason, the theory does not tell us enough. Though the first part of this claim is true this does not jeopardize the theory for two reasons. First, though we are unsure of the correct account of the nature of many of these things, it is not the case that the list contains items the existence of which is a subject of philosophical debate and division. So whilst it is true that this theory awaits the results of philosophical inquiry into the nature of these things, it is not a hostage to fortune, or otherwise deficient, for that reason. Second, the same is true of other theories of well-being. The hedonistic enumerative theory awaits the correct theory of pleasure; the desire-fulfilment theory awaits the correct theory of desire; human nature perfectionism awaits the correct theory of human nature.

IV. AVOIDING ALIENATION

At the end of section II, I claimed that a major consideration that motivates the main alternative theory (the desire-fulfilment theory) is better accounted for by my enumerative theory. I move on now to showing that this is the case.

I take it that at least a large part of what attracts people to the desire-fulfilment theory is the idea that these theories have the requisite sensitivity to a person's attitudes (and that enumerative theories do not).Footnote ²⁰ There will of course be other things that explain why people have been attracted to desire-fulfilment views – such views are often favoured by those attracted to metaphysical naturalism for example. But given that one can be a naturalist and a hedonist, or objective-list theorist, I think that this alienation or resonance point is the best such rationale for the desire-fulfilment theory.Footnote ²¹ An influential way of putting the rationale comes from Railton:

Railton's claim helpfully brings out what is wrong with certain enumerative theories. If one held that the only positive contributors to well-being were things such as living in particular kinds of community, serving certain social roles or performing particular kinds of work, then one would be forced to hold that people whose lives contained these things were high in well-being irrespective of how much the person was indifferent to, or even loathed, them.Footnote ²³ Clearly, such an enumerative theory is intolerably alienating and implausible.

Looking again at the enumerative theory that I have suggested – comprised of Achievement, Friendship, Happiness, Pleasure, Self-Respect, Virtue – we should note that all of these goods have pro-attitudes as necessary components.Footnote ²⁴ This enables this view to respect Railton's point and to avoid being an alienating conception of a person's good. For everything on the list involves the person's engagement through her holding various kinds of pro-attitudes such as endorsement, desires and affection. If one is experiencing these kinds of states of engagement, one is doing well (and vice versa).

Desire-fulfilment theory correctly latches onto the fact that how well one's life goes is correlated (albeit imperfectly) with the extent that one experiences pro-attitudes. The mistake that the desire-fulfilment theory makes is that of allowing one's desires and attitudes to play an unrestricted explanatory role, one that goes beyond their being elements of well-being enhancers. The theory does this by treating our desiring something as being sufficient to make something good for us. This is because the desire-fulfilment theory offers more than a biconditional that asserts a necessary connection between what contributes to well-being and our desires. The theory claims that things are good for us because we desire them. As Sobel puts it:

It is this feature of the desire-fulfilment view that gives rise to the ‘scope problem’ for these theories – namely that too many things come out as good for us.Footnote ²⁶ My theory avoids the scope problem whilst accommodating what people have found attractive in Railton's intuition, namely that a conception of well-being that was completely independent of their pro-attitudes would be implausible. In light of this, my enumerative theory is able to capture much of what is attractive about desire-fulfilment theories whilst avoiding the scope problem that afflicts those theories. This is because my theory is attitude-sensitive without giving attitudes a role over and above being necessary constituents of the states that are good for us.Footnote ²⁷ Going back to the two types of attitude-independence I distinguished earlier, we should notice that the fact that my enumerative theory is attitude-dependent in the sense captured by AD2 enables it to capture what we find compelling in Railton's intuition.

I am not claiming that this feature of my view – its attitude-sensitivity – secures an outright success for enumerative theories. Nor do I deny that such theories face other challenges. But it is enough to show that my enumerative theory captures a major attraction of desire-fulfilment theories – attitude-sensitivity – and without incurring the significant costs that desire-fulfilment theories generate.

V. TESTING THE THEORY

How can we determine whether the enumerative theory I have suggested is generally adequate? There are two standard objections. One concerns the details of the theory, namely whether it has all and only the right things on the list. Another is whether the theory is arbitrary in some troubling way. I start with the second objection.

Someone might wonder how it is that we can determine what is on the list and how we can do so without inviting the charge that the theory is arbitrary. Though there is something to this objection, it is a mistake to think that it is damaging for the enumerative theory that I have proposed in particular. As it is sometimes wielded, the objection is overblown because no theory of well-being could avoid it.Footnote ²⁸

If we ask the hedonist why pleasure (and only pleasure) is good for someone (and why pain and only pain is bad for someone) it is not clear what non-trivial explanation they could give for this. Similarly, if we ask the desire-fulfilment theorist why something is good for us if and only if, and because, we desire it, it's not clear what non-trivial explanation they could give of this fact. The same is true for human nature perfectionism and every other theory of well-being. Whilst more explanatory depth is better, other things being equal, all explanation stops somewhere.Footnote ²⁹ For that reason one cannot say that in the absence of a further justification for the things on the list then the theory is troublingly arbitrary, for a similar objection could be mounted against all of the other theories. The truth in the arbitrariness objection is that it is a difficult task to discover what should be on the list. But there are at least things that we can do to test and if necessary improve the list.Footnote ³⁰ This takes me to the first objection.

Moving on to the issue of the specific content of the list, clearly there are two kinds of counterexamples that we must rule out. We must test for absences and we must see if the list includes anything that should not be there. To test for the first, we follow a method like that used in discussing Crisp's anhedonic achiever.Footnote ³¹ We should imagine two people with identical bundles of identical goods and then imagine some addition of some purported extra basic well-being contributor or detractor. If we find it plausible to think that their well-being is thereby made unequal, this is evidence that we should expand the list to include this item.

To test for the second kind of mistake we should take bundles of the goods on the list and imagine two people with identical bundles except for the presence of, or extra, (e.g.) friendship. If in such cases we think that the two people have identical levels of well-being, we have some evidence that we should take that good off the list.

Of course, these tests will only generate defeasible evidence that some putative good does or does not belong on the list. Thankfully, there are other resources that we can bring in. First, one can try to show how this apparent good is in fact catered for by something else on the list. For example, one could argue that knowledge as such does not belong separately on the list of well-being contributors, for doing so would be double-counting because the cases where it looks plausible to think that knowledge is good for someone are, in fact, cases where achievement in their (epistemic) goals or the exercise of their (epistemic) virtue is good for them. Second, we can try to explain our propensity to make such judgements of inequality in some other way, such as by being distracted by instrumental value or something else.Footnote ³² So although the tests mentioned above will not conclusively establish that something belongs on the list it is a useful way of determining of some putative good whether we should investigate it further.

Because of constraints of space, I have not here performed these tests for other putative goods such as autonomy, aesthetic appreciation, love, freedom, etc. But I have suggested how to do so. More importantly, I hope to have shown that doing so would be in the service of a theory that is more plausible than is often thought. For the enumerative theory, at least as formulated here, can accommodate the idea that a conception of well-being that could fail in any way to engage the agent would be implausibly alienating. And it is no more vulnerable to the charge of arbitrariness than other theories of well-being.