A unified theory of intrinsic value.
There are a variety of candidates for the ground of intrinsic value. Different theories posit that the ground consists of some or all of the following: types of experiences, desire-satisfaction, virtue, meaningful relationships, true beliefs, desert-satisfaction, etc. The ground can be local or global depending on whether it grounds value of a fact-specific part of the universe (e.g., Jones enjoying this ice cream) or all facts considered. In this essay, I argue that the single ground of global intrinsic value is the total amount of desert-adjusted well-being. I begin by providing a sketch of moral desert. I then argue that desert-adjusted well-being alone satisfies our intuitions about global intrinsic value.
b. A brief overview of desert
I begin with the structure of desert. (1) Roughly, desert is a relation that determines the intrinsic goodness of things. More specifically, it is a function that converts a ground (e.g., a person having a certain degree of virtue) and an object (e.g., that person's level of well-being) into a level of intrinsic value. The complex formulation leaves open whether desert-satisfaction is by itself intrinsically valuable or whether it affects intrinsic value only by changing the value of some other factor (e.g., well-being).
I shall make a number of assumptions that will not be defended in this essay. First, I will assume that the event or state that is deserved is something that affects the deserving person's well-being. This in part makes the range of objects of desert (i.e., that which is deserved) depend upon the kinds of things that can affect a person's well-being. So, for example, the issue of whether a dead person can deserve things, such as remembrance, depends in part on whether a person's well-being is affected by post-mortem events.
Second, I assume that the subject of positive or negative desert is a morally responsible agent rather than an abstract entity, say, a musical score, or a non-responsible agent, say, the spitting cobra. In this essay, I focus on moral rather than non-moral desert. One who thinks that creatures that are not morally responsible agents who can nevertheless morally deserve things will be able to accept much of what is commonly said. When someone makes statements such as "Liu's musical score deserved the Oscar," or "The spitting cobra deserves its fearsome reputation," the truth of these statements does not depend on moral desert. Perhaps in cases like these it depends on some kind of non-moral desert.
Third, even though I maintain that a deserving being is a morally responsible agent, I will not assume that a person has to be morally responsible for the ground of her desert claim. On certain accounts, desert is sensitive to a particular type of intrinsic value, namely, that which accrues to an agent in virtue of some act or characteristic for which she is morally responsible. Other accounts do not make such a claim. (2) One case in which these two kinds of account will disagree involves compensation for an unjust injury that the injured person is not responsible for receiving. Accounts of the first kind imply that the person cannot deserve compensation in such a case, while accounts of the second kind imply that such compensation can be deserved. Moreover, accounts of the first kind deny that a person can deserve treatment simply in virtue of being a person, where this is understood in terms of having certain capacities (e.g., rationality), since a person is not responsible for her own possession of such capacities. Accounts of the second kind, on the other hand, may entail that somebody can deserve something simply in virtue of being a person. In this essay, I remain neutral with respect to these differing views.
Fourth, I distinguish desert from merit. The person who is the most meritorious is the one who ought to receive something under certain conventional rules (i.e., rules that result from a social understanding). This might rest on attributes for which a person is not responsible or for which he is not the most deserving. Consider Al, who wins a race because he is the fastest sprinter in it. However, another competitor, Bob, worked harder, sacrificed more, and would have won the race because he was faster than Al. Unfortunately, racist thugs beat Bob up minutes before the start of the race. In such a case, Al merits but does not deserve the title.
Fifth, I do not think that desert satisfaction requires a causal connection between the ground and object of desert. (3) I consider desert satisfied when a person receives the event or state he deserves, even if this is not caused by his act or character. For example, if a criminal escapes punishment but receives the suffering he deserves through a crippling car accident, this intuitively seems as intrinsically good as if he had received the suffering through a reliable legal process. Of course, there may be other intrinsic goods that accompany a close causal connection and that account for our preference in favor of such a connection between the ground and object of desert. For example, intrinsic value might accompany the victim's pleasure in learning of the suffering of her brutal attacker, and this intrinsic value may be independent of that which accompanies the attacker's getting what he deserves.
2. Desert-Adjusted Well-Being Alone Satisfies the Criteria for Global Intrinsic Value
a. Theories of global intrinsic value
In this essay, I shall assume that intrinsic value exists. Global intrinsic value is the intrinsic value of all facts. In contrast, local intrinsic value is the intrinsic value that something would have if it were the only thing that existed. This is analogous to the notion of a prima facie duty, which is a duty that obtains if it is not undermined. For example, some promises (e.g., to murder a child) might on some accounts produce no duty at all, rather than merely being overridden by another duty. Like an undermined prima facie duty, local intrinsic value is not an actual instance of value simpliciter since it can be enhanced, modified, or transvalued by the presence of other factors. This change in value indicates that local intrinsic value is not an actual value that by itself contributes to the value of a larger whole, but rather a mere tendency that need not be actualized.
There are several different theories of global intrinsic good. The first class of theories identifies the global ground in terms of what makes persons' lives go well (i.e., well-being).
(1) Welfarism: The ground of global intrinsic value is well-being. One plausible variant of this theory asserts that the sum of well-being is the ground of global intrinsic value.
(1a) Arithmetic Welfarism: The ground of global intrinsic value is the sum of individuals' well-being.
The term "arithmetic" is designed to capture the notion that the sums are arrived at via addition and subtraction. There can be other variants that focus not on the sum but on some other relation between amounts of well-being. For example, one might adopt a multiplicative theory that asserts that the ground is the product of individuals' well-being. For example, if two persons both have three units of well-being, a simple multiplicative theory might entail that the state of affairs including the two has nine units of intrinsic value. Alternatively, one might adopt an analogue to the difference principle whereby the ground is equal to the well-being of the worst-off individual.
A second plausible variant identifies the ground as a function of well-being, but it allows that some other factor might adjust the value of well-being. The adjusting factor might include such things as desert or autonomy.
(1b) Factor-Adjusted Welfarism: The ground of global intrinsic value is a function of the amount of well-being adjusted by some function.
Desert-adjusted welfarism is a variant of (1b) that identifies desert as the sole adjustment factor.
(1bi) Desert-Adjusted Welfarism: The ground of global intrinsic value is a function of the amount of well-being adjusted by desert.
Note that the adjustment here is done individual by individual, since the global value depends on the value of individuals' lives. There are also arithmetic and other versions of this type of welfarism, depending on whether the adjusted values of individuals' well-being is summed or combined in some other manner.
Desert-adjusted welfarism looks at individuals' well-being with the value of each individual's state being a function of his well-being enhanced, discounted, or transvalued by his desert. For example, imagine that a state containing a person who deserves and receives +10 well-being units has a value of +20 units of intrinsic value. The state with two such persons, assuming an arithmetic function, has +40 units of intrinsic value. Other factor-adjusted welfarist theories might look at factors such as need or autonomy, rather than desert, in determining the value of a level of well-being in an individual.
This class of theories is compatible with different theories of well-being. In particular, it is compatible with theories that identify well-being in terms of experiences, desire-satisfaction, objective-list elements (i.e., features that are independent of experiences and desire-satisfaction), or combinations of these. (4) Objective-list elements include things such as virtue, knowledge, and meaningful relationships. This class of theories is also compatible with different theories as to whose well-being grounds intrinsic value (e.g., God, autonomous agents, sentient creatures).
The second class of theories identifies the ground as being something other than well-being.
(2) Non-Welfarism: The ground of global intrinsic value is independent of well-being.
These grounds are separate from well-being but still might affect it. This class of theories might focus on factors such as community, aesthetic value, desert-satisfaction, or the relation between God and his creatures. One plausible variant is that the ground is desert-satisfaction.
(2a) Desert-Satisfaction: The ground of global intrinsic value is a function of desert-satisfaction.
This theory can be in arithmetic or in some other form depending on how the degrees of desert-satisfaction across individuals are combined. An attractive version of this theory looks at the sum of various persons' desert-satisfaction. Both classes of global intrinsic good are compatible with theories that focus on average amounts of the ground (averagism) or ones that focus on the total amounts of the ground (totalism). (5)
On (2a), the intrinsic value of something is a function of the degree to which desert is satisfied. For example, consider when one person, Jane, deserves -10 units of well-being and receives it, whereas a second, Kate, deserves +10 units of well-being and receives it. On (2a), both scenarios are equally valuable since they contain an equal degree of desert satisfaction. This differs from Desert-Adjusted Welfarism, which asserts that well-being, properly adjusted, rather than desert-satisfaction, grounds intrinsic value. On (1bi), the state involving Kate is likely more valuable than that of Jane since the enhanced value of Kate's positive well-being is likely to be greater than the value of Jane's negative well-being.
Other combinations of theories are also possible. There are also hybrid theories that identify the ground with elements of both theories. A disjunctive hybrid theory asserts that well-being (adjusted or not) and other factors separately ground global intrinsic value. A conjunctive hybrid asserts that well-being and the other elements together ground global intrinsic value.
b. Metaphysical criteria
The successful theory of intrinsic value must meet certain metaphysical conditions as well as satisfying a range of intuitions about particular cases. The distinction between the two criteria is not a sharp one since the metaphysical conditions are justified at least in part on the basis of intuitions.
The ground of global intrinsic value must adequately rank different scenarios and satisfy certain plausible principles. Specifically, it can be used to produce a real-number ranking of every scenario. A real-number ranking has a true zero point and equal intervals between units of measurement. This is based on an analogy to other quantity rankings. Both classes of theories can satisfy this requirement.
The notion that intrinsic value involves a real-number ranking is supported by several reasons, albeit weak ones. First, it is difficult to generate scenarios that intuitively seem to be intransitively ranked. (6) Second, intrinsic goodness rankings have a true zero point. This is the point at which the states of the objects of the world do not make the world any better or worse than a vacuum. Third, levels of well-being seem to allow for interpersonal real-number ranking, and to the extent that some version of Welfarism is true, this would appear to support such rankings. Fourth, another area of morality, rightness, seems to allow for real-number ranking and one might think that adjacent areas of value theory admit of similar analyses. It intuitively appears that we can provide a real-number ranking of things such as just punishment and compensation. For example, consider when a particular wrongdoer, for example, a rapist, serves three years rather than six months in prison. Intuitively, this isn't merely more likely to satisfy the victim's (or perhaps the citizenry's) claim to punishment, but more likely to some particular degree to satisfy the claim. Here, also, there is a true zero point (i.e., no punishment). The specific degree will depend on the rapist's utility function and the degree of harm that his act caused the victim (or perhaps could reasonably be expected to cause the victim). The three years might also be equivalent to a specific amount of torture, isolation, lost wages, or other penalty. These comparisons seem to involve the sort of fine-grained rankings that typify real-number rankings. In fact, it is the fine-grained commensurability of wrongdoings and punishments that on some accounts explains our intuition that justice provides a precise ceiling on permissible punishment. (7) Similar fine-grained rankings and equivalency judgments also intuitively seem to characterize the status of compensatory payments. Here the fine-grained rankings help to explain the notion that justice provides an exact floor for permissible compensation and that this floor is commensurate with both a specific amount of money and the degree to which the victim was harmed. Thus, the notion that the ranking of different states of affairs should be a real-number one fits nicely with the difficulty of generating clear examples of intransitively ranked states of affairs, the idea that in assessing levels of intrinsic value there is a true zero point, real-number rankings of well-being, and real-number ranking in an area of morality adjacent to intrinsic goodness.
There are also certain fundamental metaphysical principles about the good that are often thought plausible. Among them is the following.
(3) Principle of Organic Unities: The intrinsic value of a sum is not necessarily equal to the sum of the intrinsic value of its parts.
The argument for this principle is that it is necessary to account for a number of intuitions. Specifically, certain states can be bad despite having only a (locally) good part. For instance, Smith's being pleased at the thought of Jones's suffering intuitively seems bad. This intuition remains even though the state contains pleasure (in Smith) and no pain (since it does not entail that Jones is suffering). (8) Similarly, this principle accounts for the comparative rankings of various states. We intuitively think, for example, that Smith s being pained at the thought of Jones's pleasure is worse than Smith's being in pain. This principle is also necessary to account for the intuition that in some cases a person's character or past acts affect the value of his current pleasures. For example, consider a scenario in which, at the end of his life, Adolph Hitler lives alone on a tropical island, although everyone thinks that he died in a bunker at the end of World War II. It intuitively seems better that Hitler's life goes poorly than that it goes well. The intuition here is that the value of pleasure is reversed due to Hitler's character or past acts. This principle, if true, rules out the straightforward types of Welfarism, such as Arithmetic Welfarism, because the latter has no basis for adjusting the value of Smith s pleasure on the basis of its object or relation to past acts.
Next consider the following principle:
(4) Principle of Universality: The intrinsic value of a part is independent of the whole of which it is a part.
This principle accounts for the intuition that if something grounds intrinsic value, then it will continue to do so unless undermined or overridden. If pleasure grounds intrinsic value in a local state, then it will continue to do so regardless of the state with which it is conjoined. The underlying idea is that the intrinsic value of a thing depends on, and only on, its intrinsic properties. (9) So Smith's being pleased intuitively seems to have the same intrinsic value regardless of whether it is conjoined with Jones's being in pain. It might seem that this conflicts with the Principle of Organic Unities since Smith's being pleased is intrinsically bad if its object is the thought of Jones's suffering. However, this is not a contradiction since a specific state can be a local intrinsic good (i.e., if it were an isolated state, then it would be good) without this value contributing to the value of a larger state in an arithmetic manner. Both Welfarist and Non-Welfarist Theories can satisfy this principle.
It might be objected that the Principle of Organic Unities doesn't apply to the sort of states with which we are concerned (e.g., Smith's being pleased at the thought of Jones's suffering). This is because the object of the thought, say, Jones's suffering, is a property of Smith's pleasure rather than a part of it. The underlying idea is that only particular objects, not properties, are parts of conjoined objects and, perhaps, a state that focuses on conjoined objects. The content of a pleasure is a feature of it rather than a part of it. If this is correct, then an analogue principle of the Principle of Organic Unities would also seem to be true. On this analogue principle, the intrinsic value of a state of affairs is not necessarily equal to the sum of the values of its exemplified properties. An analogue principle of the Principle of Universality also seems to be true.
Some theorists claim that there are higher-order goods. (10) Higher-order goods are ones that are intrinsically better than an equal quantity of other goods. The quantity of a good is a product of its intensity and its duration. (11) I am assuming here that goods have intensities or come in degrees. The idea behind higher-order goods is that attitudes (e.g., pleasure and true belief) can be quantified and these quantities compared in a way that is distinct from the attitudes' intrinsic importance. Such goods are necessary to account for certain intuitions. For example, they can explain why it is better to live as a fully functioning adult in ecstasy for a century than as a contented infant for 10,000 years despite the fact that the latter contains a greater amount of valuable experiences and desire-satisfaction. Alternatively, this type of example might be accounted for by the notion that objective-list elements (e.g., virtue, knowledge, and meaningful relationships) are necessary factors for a person's life to go well. This in turn is supported by intuitions that suggest that when pleasure is held fixed, a person's life goes better with the addition of these elements. (12)
Both Welfarism and Non-Welfarism can allow for higher-order effects by introducing either a plurality of intrinsic goods or a higher-order account of what promotes someone's welfare. On a monistic account of intrinsic goodness, such as that found in Desert-Adjusted Welfarism, there can't be any higher-order goods since there isn't a plurality of intrinsic goods. However, there can be higher-order functions with regard to well-being. For example, my life might go better if I experience a century with a high intensity of pleasure and objective-list elements rather than an eternity of constant mild pleasure and none of the objective-list elements. This then has effects similar to that of an account with higher-order goods since it allows that a state of affairs with considerably less pleasure than a second has more value than the second. The higher-order function of well-being might be a multiplier function (e.g., well-being is equal to the product of two or more components of well-being) or a lexical-ordering function (e.g., a life with all of the elements necessary for flourishing is better than a life with any amount of only some of them). It is not clear, however, whether this latter mechanism allows for real-number ordering of either well-being or intrinsic value.
Another metaphysical notion is that the bearer of global intrinsic value is a state of affairs that obtains. The argument for this is that intuitively an individual (e.g., Smith) is not intrinsically valuable, for a world with Smith but where he has no level of well-being would not be intrinsically valuable. Nor is the attribute (a specific level of well-being) intrinsically valuable where it is unexemplified. Rather, it is a state of affairs that is valuable because it contains the individual (Smith) exemplifying a level of well-being. This state of affairs must obtain since an individual like Smith exists only in actual scenarios. Again, both Welfarism and Non-Welfarism can account for the nature of the bearer of intrinsic value. This account of the bearer of global intrinsic value is consistent with the notion that intrinsic value is essential to its bearer. This account is also consistent with statements about the intrinsic value of counterfactual states of affairs since these statements should be understood as asserting that the states of affairs would have a certain amount of intrinsic value were they to obtain.
Kantians will strenuously disagree with this claim. They will claim that if you have to choose between running over a person and a horse, assuming no danger to yourself, you are morally required to run over the latter. The best explanation of this is that a person, independent of his level of well-being, has more intrinsic value than a horse. This type of argument is unconvincing since this requirement can be explained in terms of deontological constraints rather than in terms of bringing about the best results. For example, if the Kantian has to choose whether to kill a person who will live for another day in order to transplant his organs to other persons all of whom will live for more than thirty years, he has to avoid doing so. From this, nothing follows about the value of the two results. Kantians might still argue that the well-being of non-autonomous beings has no value. However, this assertion intuitively seems mistaken.
In addition, certain variants of Desert-Adjusted Welfarism are compatible with the Kantian claim that horses don't count. In particular, some variants assert that the states of creatures that don't deserve anything, which is different from deserving a well-being level of zero, have no value. That is, if one viewed a being's level of well-being as having some intrinsic value only if she has the capacities to morally deserve something (whether positive, negative, or zero), then a horse's pleasure would not count. This type of theory is unappealing if one intuits, and I do, that animal pleasure is intrinsically good.
Factor-Adjusted Welfarism satisfies these metaphysical relations. It is an application of the Principle of Organic Unities because it is simply a specific theory of how the intrinsic value of a whole can differ from the sum of the local intrinsic value of its parts. It allows for the Principle of Universality since this type of Welfarism is a theory of global, not local, value. Indeed, it is likely the fact that it is unfitting for undeserving persons to receive local goods that explains why factor adjustment is relevant to the global good. This type of Welfarism is consistent with the bearer of intrinsic value being a state of affairs that obtains since it is only in such a state that there are actual individuals who have some level of welfare. This theory type is also compatible with either the presence or absence of higher-order good-type effects since it doesn't address whether ranking of states of well-being might occur merely via quantities of a single property (e.g., pleasure) or via some other ranking system.
A stronger conclusion should be drawn. The cases in which organic-unities effects occur are limited to those in which persons have positive or negative desert. The organic-unity effects kick in where there are moral goods or evils on top of the natural ones. Moral goods occur where persons are virtuous or where they receive components of well-being that fit with their being virtuous or having done virtuous acts in the past. Moral evils function in a parallel manner. The virtuousness of persons is probably not a global intrinsic-good-making factor since a person's virtue contributes to the intrinsic value of a whole state of affairs only when it is linked to a particular level of well-being. For example, we can't infer whether doubling the number of saints in the world makes it a better place until we know whether the added saints are flourishing or suffering terribly. A similar set of observations holds with regard to moral evils.
On some accounts, there is a function that makes additional units of well-being add diminishing amounts to intrinsic value. (13) This explains why we intuitively think it would be better to improve the lives of suffering animals rather than flourishing ones. The same intuitions apply to persons. This would seem to present a problem for my theory since it would seem to add a factor to intrinsic value besides well-being and desert that explains the diminishing effect. However, in the case of creatures with zero desert, the desert function might account for the diminishing value. That is, the desert function for a creature with zero desert need not be a linear one. Alternatively, for creatures with zero desert, there might be a linear relation between well-being and intrinsic value but additions of pleasure might make diminishing contributions to well-being. This latter account fits nicely with the notion that pleasure is a distinctive type of feeling. (14)
In addition, the purported cases of organic-unities effects with factors unrelated to well-being and desert are not convincing. Some persons claim that a universe is better if it has a diversity of types of goods (whether moral or natural) or an increasing amount of good over time. (15) I find these examples implausible. For example, why should it matter whether a given amount of well-being is spread evenly over a period of time or whether that same amount is distributed in an increasing amount over time? The difference might be explained in terms of the universe feeling better to observers, but this is an additional welfarist good that conflicts with the notion of a fixed amount of well-being. A similar thing is true of moral good since improvement would suggest that persons were more vicious early on and it is not clear why this doesn't nullify any effects of moral improvement. Similarly, why value a diversity of natural or moral goods unless it somehow makes someone's life go better or his being more virtuous? These improvements go to the amount of good, not the forms in which it occurs.
If this is correct, then the most plausible form of Welfarism is the desert-adjusted kind because it uniquely accounts for organic-unity effects when and only when they occur. If this is correct, then Desert-Adjusted Welfarism is better suited to account for organic-unity effects than other types of Welfarism and types of Non-Welfarism that don't focus on desert.
Other variables such as knowledge, aesthetic goods, and meaningful relationships are components of a person's well-being rather than a factor that adjusts the value of that well-being. One intuition behind this is that these things affect how well a person's life goes. A second is that we can't judge whether it is intrinsically better that a person have more of these things until we know about the degree to which he is virtuous or vicious. For example, whether it is intrinsically better for a person to enjoy a beautiful Hawaii sunset and have a great marriage than not depends on whether he spent the previous year helping troubled orphans or coordinating the slaughter of Tutsis.
To the extent that autonomy enters the picture, it is also an element of well-being. The idea here is that even when we hold other factors fixed (e.g., pleasure, knowledge, and meaningful relationships) a person's life goes better to the extent that it is self-shaped, rather than the result of forces that he didn't endorse. There is an issue as to whether the contribution of autonomy can be cashed out in terms of other elements of well-being, specifically, pleasure and knowledge, the idea being that what persons really want is the experience of making decisions and contact with reality. (16) Since nothing in this essay rests on this issue, I will sidestep it.
An objector might claim that virtue and well-being independently ground global intrinsic value. He might concede that we can't read off whether additional virtue or well-being by itself makes a state of affairs a better place without knowing the value of the other factor. However, he might argue that if we hold fixed a positive level of virtue, whatever this level turns out to be, the state is intrinsically better the higher the level of well-being. The underlying assumption is that if other factors are fixed and a factor independently varies with the global intrinsic value of a state, then this factor should be seen as intrinsically valuable. More specifically, this factor should be seen as a distinct intrinsic-value-making factor. Thus the objector concludes that virtue and well-being are intrinsically valuable. (17)
The problem with this is that holding the other factors fixed and positive indicates that they are part of what makes additional well-being or virtue valuable. If these were independent factors, then there would be no need to hold the other factor fixed and positive. The need to hold the other factor fixed and positive indicates that it is the relation between virtue and well-being that makes the world a better place. This is analogous to the way in which the number of molecules, heat, and volume determine the pressure of a gas. When we hold fixed a given quantity of gas molecules and heat, volume varies with the pressure. However, from this we wouldn't conclude that volume alone determines the pressure that a gas places on the walls of a container.
c. Substantive intuitions about intrinsic value
Other intuitions rule out Non-Welfarism. Consider the following worlds.
World 1: A physical world without sentience: This world has many physical objects such as mountains, rivers, and plant life, but no sentient creatures.
World 2: The vacuum world: This world consists of nothing but space.
It intuitively seems to me that the two worlds both have no intrinsic value. I realize that others don't share this intuition. (18) Similarly, it intuitively seems to me that if God had to choose between creating the two worlds, he would be indifferent between them (unless he was going to enjoy the beauty of the former). The underlying idea here is that beauty and life are valuable because of what they do for someone, rather than valuable in and of themselves. If this is correct, then one explanation for this is that intrinsic value focuses on some element of well-being since this is what is missing from both worlds.
One issue is whether these two worlds have zero intrinsic value or no level of intrinsic value, zero or otherwise. On my account, these worlds have zero value. This is because a being that considered creating these worlds (e.g., God) should rationally be indifferent between the two worlds and a third world where beings deserve zero units of well-being and receive that amount. The underlying idea is that the addition to the world of such individuals doesn't make the world better or worse. An objector might claim that such a world doesn't have any level of intrinsic value since there are no beings at all and hence no levels of well-being or desert, whether zero or otherwise. The idea is that a vacuum has no level of well-being because a level of well-being (zero or otherwise) is not something that is present without something in which that property is instantiated. However, it is not clear why there must be a being whose life goes better or worse in order for a world to have zero amount of well-being. A vacuum has zero units of air and mass and zero objects and it is not clear why a similar thing isn't true with regard to units of well-being.
Aesthetic values are probably tied to well-being. First, there is a real issue as to whether aesthetic values are mind-independent. Realism about aesthetic values arguably receives less support from convergent customs, attitudes, and behaviors than moral realism does. Second, even if aesthetic values are mind-independent, it still seems that they make the world a better place only when noticed by sentient creatures. Consider a completely dark planet strewn with boulders that contains no sentient creatures. Such a planet does not intuitively seem more valuable when its boulders are arranged in a beautiful manner as compared to when they are not so arranged.
Similarly, life unaccompanied by sentience lacks intrinsic value. Consider, for example, an isolated human being who lies in a permanent vegetative state (biologically alive but without conscious awareness) for a year before dying as opposed to a scenario in which she dies immediately. Intuitively, she is no better off in one scenario than the other. More to the point, leaving aside the effects on others, the world intuitively seems no better with one scenario than the other. If human life unaccompanied by sentience is not intrinsically good, it is not clear why other forms of life should be.
The centrality of well-being is further borne out by the observation that values such as community, aesthetic value, and relationship to God are likely explained in welfarist terms. For example, if we had to choose between two worlds with equal levels of well-being but one had greater community and the other had other offsetting increases in well-being, perhaps due to increased meaningful marriages, it intuitively seems that the two worlds are equally good. Here "community" refers to features of a meaningful relationship between persons who are part of the same society but are not family. This pattern further supports Welfarism since it can incorporate the considerations that might be cited by a proponent of Non-Welfarism.
Furthermore, we can rule out a plausible variant of Non-Welfarism that presupposes the presence of well-being but is independent of it. We can rule out the Desert-Satisfaction Theory, since desert-satisfaction alone does not seem capable of picking out what is intrinsically valuable in a state of affairs. To see this, consider the following two worlds.
World 3: Bad guys: This world has two very evil persons who deserve a negative level of well-being (-20 units) and receive this amount.
World 4: Good guys: This world has two good persons who deserve a positive level of well-being (+10 units) and receive this amount.
Both worlds have an equal amount of desert-satisfaction. Since Desert-Satisfaction judges the value of a state of affairs in terms of desert satisfaction, the theory entails that the two worlds should be of equal intrinsic value. Yet it seems that World 4 is intrinsically preferable. This result is bolstered by the fact that it is hard to see how a mere abstract relation such as desert, as opposed to a mental state, is capable of being intrinsically valuable. It seems as if abstract relations are not the sort of thing that by themselves make the world a better place. It is worth noting that Worlds 1 and 2 also have the same amount of desert-satisfaction as Worlds 3 and 4 (namely, perfect desert-satisfaction). The view that World 4 is superior to the first two provides yet another reason to reject Desert-Satisfaction.
A deeper argument for Desert-Adjusted Welfarism is also available, but before turning to it, we have to explore briefly an issue surrounding the nature of desert.
d. Comparative desert
An important issue is whether desert is fundamentally comparative or non-comparative. If desert is in part comparative, then it is concerned with the well-being of one person in comparison to that of another. (19) On a non-comparative account of desert, one person's desert is independent of another's. A comparative theory of desert can be incorporated into both Welfarism and Non-Welfarism, depending on whether the degree of comparative satisfaction affects intrinsic value by affecting the value of someone's well-being. Since I shall argue that desert is not comparative, there is no need to address which type of theory better incorporates this factor.
Because I think that there are some possible situations in which a being who is the sole person on the earth may have desert, I think that there must be non-comparative desert. For example, if the only person on earth spends his time torturing dogs and apes, it intuitively seems that he has negative desert (e.g., he deserves more pain than pleasure).
To see the difference between comparative and non-comparative desert, consider a case in which person A non-comparatively deserves the same as B, but A enjoys greater well-being. On a comparative account, it follows that B is not getting what he deserves because his desert is the same as A and he is getting less. On a non-comparative account, whether or not B is getting what he deserves depends only on facts about B. No reference to the other person is made. I think that an account that contains comparative desert should be rejected, because it leads to there being some value in leveling down. (20) Imagine that there are two saints, Sebastian and Thomas, each of whom deserves 20 units of well-being, and that Sebastian gets 8 units (a moderately enjoyable life) and Thomas gets 4 (a minimally enjoyable life). Note that the assignment of these numbers assumes that non-comparative desert is in effect and that both are not getting what they non-comparatively deserve. From the point of view of comparative desert, lowering Sebastian from 8 to 4 units of well-being improves the situation in one respect. This is because the equally deserving persons are then no longer experiencing different levels of well-being. My intuition is that this does not improve the situation in any way that contributes to intrinsic value, and so the comparative notion should be rejected.
Others reject this intuition. Thomas Hurka argues that such a change does make the world in one respect better, regardless of whether the two saints are getting more or less than they deserve. (21) To see why Hurka is incorrect, let us begin by modifying the above hypothetical by assuming that these two saints have (via plane crashes) landed on separate and uninhabited Pacific islands and will live there for the rest of their lives. Sebastian loses part of his hand to infection and this lowers his total well-being from 8 to 4 units. Overall, this seems to be a bad thing; but Hurka can admit this and argue that the badness of his decreased well-being outweighs the goodness of the greater degree of comparative desert. However, imagine a similar scenario in which Thomas does not even exist. Is the overall badness of Sebastian's loss any worse? I think not. If this is correct, then comparative desert does not affect intrinsic value.
Hurka provides a positive argument in favor of comparative desert. He posits a situation in which the death penalty is given to all and only blacks who deserve it, but to only some of the whites who deserve it. He then argues that the badness of this situation is not explained by the injustice of allowing white murderers to escape the death penalty. (22) In this case, Hurka is correct insofar as our intuitions do appear to support the notion that the unfair pattern of death penalty distribution is an additional bad element. However, our intuitions in this case can be explained away. Suppose that the death penalty is unfairly distributed among both white and black murderers, with a disproportionately high percentage of good-looking murderers getting the death penalty and a disproportionately low percentage of ugly murderers getting it. This occurs because of the halo effect, whereby good-looking people are judged more favorably for their good acts and more harshly for their bad ones. However, nobody is aware of the effect in this context since jurors, the press, prosecutors, etc. never see enough of the defendants to recognize the pattern. The notion that this discriminatory pattern should support the abolition of the death penalty is considerably weaker than in Hurka's case. This suggests that what drives our intuitions in the racial case is the badness of persons applying racist beliefs or attitudes, and the harm caused when others notice it, not the frustration of comparative desert.
e. The deeper argument for Desert-Adjusted Welfarism
The deeper argument for Desert-Adjusted Welfarism rests on its connection to fittingness theory. On one attractive account of desert, desert is the fitting state of affairs for persons-with-certain-values. This is part of a more general fittingness theory. If desert is the highest-order instance of fittingness, and I argue below that it is, then it should not be surprising that desert plays such a central role in intrinsic value. To see that desert is part of a more general fittingness relation, let us examine the latter. Fittingness theory is committed to the following three claims.
(5) Fittingness: It is fitting that locally intrinsically good things be connected to (for example, received by or be directed at) other intrinsically good things and that intrinsically bad things be connected to other intrinsically bad things.
(6) Degree of Fittingness: The degree of fittingness is a function of the proportionality of the connection specified in (5).
(7) Fittingness and Value: This fittingness relation affects intrinsic goodness and does so in accordance with the degree of fittingness.
Note that the first two propositions concern local intrinsic value, the third either local or global. Relations that are not fitting either lack fittingness or are unfitting; a parallel set of principles applies to them. It should be noted that fittingness theory does not rule out other intrinsic-good-making properties.
The main advantage of the fittingness theory is that it explains the unified structure of virtue and desert in terms of a more fundamental relation. This unified account comes out in Hurka's theory of virtue and desert. On Hurka's theory, it is intrinsically good that persons have a pro-attitude toward intrinsically good things and a con-attitude toward intrinsically bad things. This theory of virtue addresses attitudes toward first-order goods (e.g., pleasure and knowledge), second-order virtues (e.g., love of others' pleasure and knowledge), and still higher-order goods (e.g., love of others' love of pleasure and knowledge). On this account, then, the virtues are those attitudes to goods and evils that are intrinsically good. Virtue is thus a higher-order good focusing on the conjunction of two intrinsic goods: a first-order good (e.g., pleasure and knowledge) and a higher-order good (attitudes toward first-order goods or other attitudes). This account is recursive because the system of valuation applies to successively higher-level goods. On this recursive theory, desert is the intrinsic good that is one level higher than virtue. Specifically, desert is the relation between a person's degree of virtue and level of well-being. (23) A fittingness theorist might explain this relation in terms of the notion that it is fitting that virtuous persons receive a level of well-being proportionate to their virtue (or viciousness). Thus the fittingness theory can provide a unified account of the structure of virtue and desert.
This fittingness principle also seems to be a good candidate for a primitive principle of goodness since it or something like it underlies other plausible principles of goodness. It is capable of explaining the Principle of Organic Unities. This principle asserts that the intrinsic value of a whole is not necessarily equal to the sum of the intrinsic values of its parts. For example, the intrinsic value of Smith's taking pleasure at the thought of Jones suffering is not merely the aggregate of Smith's pleasure (which has positive value) and his thinking of Jones suffering (which has no value since this does not entail anything about Jones). Fittingness explains this principle by accounting for the way in which the value of a whole is in part a function of the relation between its parts. Fittingness is also compatible with the Principle of Universality. This principle asserts that the local intrinsic value of a part is independent of the larger whole of which it is a part. For example, a virtuous person's enjoying ice cream is locally intrinsically good regardless of the larger state of affairs of which it is a part. Fittingness is compatible with this principle since the local intrinsic-good-making feature is present regardless of whether a fittingness relation makes the smaller state part of an intrinsically bad whole.
Fittingness further explains the nature of the bearer of intrinsic value. Ramon Lemos and Noah Lemos have argued that it is states of affairs that obtain that are intrinsically valuable. (24) Their argument consists of thought experiments that show that while states of affairs that obtain intuitively seem intrinsically valuable, the same is not true for mere states of affairs, individuals, or attributes. For example, the mere possibility of a person receiving pleasure is not intrinsically good, nor is the isolated person or the abstract universal of pleasure. Fittingness explains the nature of the bearer by asserting that it is only states of affairs that obtain that contain the fittingness relation, which holds when an individual exemplifies certain attributes (e.g., virtue). It might be objected that it is fitting that the virtuous flourish even if no one is virtuous, and that my theory can't account for this because this would concern a state of affairs that doesn't obtain. However, my theory can account for this claim if it is understood in the counterfactual sense, that is, if there were a virtuous person, then his flourishing would be intrinsically valuable.
Fittingness might seem an unsatisfying principle given its rather simple and primitive nature. However, the explanatory work done by the relation suggests that it is not trivial. In fact, one might expect that if there is a unified theory of value, some relation, basic or not, would recur throughout the different levels of value.
Desert satisfaction is the highest-order fittingness relation. Desert consists of the relation between virtue and well-being. This involves a relation between the only two types of local intrinsic goods. The argument for this rests in part on the lack of other plausible candidates for local and global intrinsic value. It also rests in part on the absence of a valuable relation between different individuals' desert (assuming we reject comparative desert) or between desert and some other intrinsically valuable entity. Since fittingness involves the relation between the only two local intrinsic goods and since desert isn't part of a still higher valuable relation, it is the highest-order fittingness relation.
Since desert is the highest-level fittingness relation and since fittingness is what in part makes a state valuable, it makes sense that it would be what in part makes a global state valuable. This adds further support for Desert-Adjusted Welfarism since it suggests that it coheres with the large body of intuitions and principles about value. In particular, it coheres with the local value of well-being and virtue, the relation between the two, and the recursive nature of virtue, and it does so in a way that illuminates the shared relation that explains why these things are valuable.
Desert-Adjusted Welfarism asserts that the ground of global intrinsic value is a function of individuals' well-being adjusted by their desert. I conclude that this is the correct theory of global well-being. One rival, Arithmetic Welfarism, ignores the intuitively attractive cases that motivate the Principle of Organic Unities. Another class of rivals, Non-Welfarism, puts intrinsic weight on things such as aesthetic values, life, and community that are better seen as valuable because they contribute to individuals' well-being. One type of Non-Welfarism, namely, Desert-Satisfaction, doesn't satisfy our intuitions that desert-satisfaction among the virtuous is intrinsically better than similar satisfaction among the vicious. My conclusion further rests on the claim that Desert-Adjusted Welfarism can account for our intuitions about cases where organic-unity effects occur (i.e., cases where persons have positive or negative desert) and cases in which they do not (i.e., cases of zero desert). In addition, Desert-Adjusted Welfarism coheres nicely with the fittingness relation that unifies different types of local intrinsic goods. If this is correct, then desert is central to intrinsic goodness.
This theory has implications for political philosophy, since it suggests that if political and economic systems (e.g., libertarian, welfare-state capitalist, and socialist) economies are to be chosen on the basis of their effects rather than on their tendency to respect side-constraints, then they should be ranked according to the degree to which they maximize desert-adjusted well-being. Elsewhere, there have been arguments over whether desert rests on acts or virtue and, if the former, whether capitalist acts ground desert. (25) There are other complicating issues such as whether we should look at average or total values. My analysis of intrinsic value entails that to the extent that consequences should determine which system to adopt, these issues are of crucial importance. (26)
State University of New York, Fredonia
(1) This structure of desert is different from the standard account found in George Sher, Desert (Princeton: Princeton University Press, 1987), pp. 6-8; Louis P. Pojman, "Does Equality Trump Desert?" in Louis P. Pojman and Owen McLeod, eds., What Do We Deserve? (New York: Oxford University Press, 1999), pp. 283-97; James Sterba, "Justice and the Concept of Desert," The Personalist (1976), pp. 188-97. The standard account asserts that subject, S, deserves object, O, just in case it is intrinsically good (or bad) that S receive O in virtue of having some ground. For example, a woman deserves the love of her son just in case she has been a caring and effective mother to him. The problem with this definition is that it assumes that desert is an independent ground of intrinsic value, and some influential accounts deny this.
(2) An argument against the notion that the ground of desert must be something for which the agent is morally responsible can be found in Fred Feldman, "Desert: Reconsideration of Some Received Wisdom," Mind 104 (1995), pp. 63-77.
(3) This is in contrast to Thomas Hurka, who argues that the added causal connection increases the intrinsic value accruing to a person who receives deserved treatment; see Thomas Hurka, "The Common Structure of Virtue and Desert," Ethics 112 (2001), p. 12 n. 10. Unfortunately, almost any thought experiment in which the well-being and badness of persons is held fixed will produce a variation in another value, e.g., knowledge, and hence it is hard to get a clean thought experiment by which to defend our assumption on causation.
(4) This distinction comes from Derek Parfit, Reasons and Persons (New York: Oxford University Press, 1984), pp. 493-502.
(5) The notion of average well-being itself allows for a wide range of theories depending on what one is averaging over. For example, one might look at averages for different slices of time, persons over their lifetimes, or person-stages. See Thomas Hurka, "Average Utilitarianisms," Analysis 42 (1982), pp. 65-69; Thomas Hurka, "More Average Utilitarianisms," Analysis 42 (1982), pp. 115-19.
(6) A few purported cases of intuitively plausible intransitive rankings occur in various contexts; see Stuart Rachels, "Counterexamples to the Transitivity of Better Than," Australasian Journal of Philosophy 76 (1998), pp. 71-83. Two ways in which these cases are accounted for are via higher-order goods and the diminishing marginal value of different events or states. For an example of the two strategies, see Derek Parfit, "Overpopulation and the Quality of Life," in Peter Singer, ed., Applied Ethics (New York: Oxford University Press, 1986), pp. 145-64, and Thomas Hurka, "Value and Population Size," Ethics 93 (1983), pp. 496-507. My own approach is analogous to the higher-order-good strategy but focuses on the way in which well-being is calculated.
(7) At least one theorist denies that there is such a ceiling; see H. Scott Hestevold, "Disjunctive Desert," American Philosophical Quarterly 20 (1983), pp. 357-63. I find Hestevold's example in support of this notion unconvincing, but a full discussion of it would take us too far a field.
(8) There is an issue of whether these states really have parts as opposed to attributes. In the Smith/Jones case, it seems that the object of Smith's pleasure (Jones's suffering or the representation of it) might be understood as an attribute of a complex mental state rather than a part. If this is correct, then what must be satisfied is an analogue to the Principle of Organic Unities. Such an analogue might state that the intrinsic value of a whole in virtue of all its properties is not necessarily equal to the sum of its values in virtue of having different particular properties.
(9) Shelly Kagan has challenged this in his "Rethinking Intrinsic Value," The Journal of Ethics 2 (1998), pp. 277-97. Kagan claims to have examples that show that it is possible that there are intrinsically valuable things whose value depends on their relation to other things. His argument does not succeed if one identifies states of affairs that obtain as the bearer of intrinsic value since the intrinsic value of such states rests on non-relational attributes.
(10) See, e.g., John Stuart Mill, in George Sher, ed., Utilitarianism (Indianapolis: Hackett Publishing Co., 1979), pp. xii-xiii; W. D. Ross, The Right and the Good (Oxford: Oxford University Press, 1930), p. 150; Derek Parfit, "Overpopulation and the Quality of Life," in Peter Singer, ed., Applied Ethics, p. 161; Noah Lemos, Intrinsic Value (New York: Cambridge University Press, 1994), chap. 4.
(11) The idea for this definition comes from Fred Feldman, Doing the Best We Can (Boston: D. Reidel, 1986).
(12) These intuitions are brought out in Parfit, Reasons and Persons, pp. 493-502, and W. D. Ross, The Right and the Good, repr. (Indianapolis: Hackett, 1988), pp. 134-41.
(13) See, e.g., Fred Feldman, "Adjusting Utility for Justice: A Consequentialist Reply to the Objection from Justice," Philosophy and Phenomenological Research 55 (1995), p. 264; Neil Feit and Stephen Kershnar, "Explaining the Geometry of Desert," Public Affairs Quarterly 18 (2004), pp. 273-98, esp. p. 282.
(14) Fred Feldman criticizes this view because of the lack of a unified type of feeling in different parts of pleasure. Fred Feldman, "On the Intrinsic Value of Pleasures," Ethics 107 (1997), pp. 448-66. He views pleasure as, roughly, an experience that one prefers to have for its own sake.
(15) Two writers who appear to put forth such a claim about the diversity of goods are A. C. Ewing, Value and Reality (London: George Allen & Unwin, 1973), p. 221, and William Frankena, Ethics, 2d ed. (Englewood Cliffs, NJ: Prentice-Hall, 1973), p. 92. However, they might have been viewing the diversity as enhancing well-being. For an assertion of the value of increasing goodness, see Franz Brentano, The Foundation and Construction of Ethics, trans. Elizabeth Schneewind (New York: Humanities Press, 1973), pp. 196-97; Ewing, Value and Reality, p. 219.
(16) The notion that the value of autonomy is ultimately grounded in the value of being connected to reality can be seen in Robert Nozick, Anarchy, State, and Utopia (New York: Basic Books, 1974), pp. 42-45, and Thomas Hurka, "Why Value Autonomy?" Social Theory and Practice 13 (1987), pp. 361-82, esp. p. 372. However, both theorists probably would not accept the claim that value of autonomy can be cashed out in terms of its contribution to pleasure and knowledge.
(17) I owe this objection to Shelly Kagan.
(18) G. E. Moore, e.g., has the opposite intuition; see G. E. Moore, Principia Ethica (Buffalo: Prometheus Books, 1988), pp. 83-84.
(19) The notion that there is comparative desert can be seen in Shelly Kagan, "Comparative Desert," in Serena Olsaretti, ed., Desert and Justice (Oxford: Clarendon Press, 2003), pp. 93-122; Thomas Hurka, "Desert: Individualistic and Holistic," in Serena Olsaretti, ed., Desert and Justice, pp. 45-68.
(20) The idea for this point comes from Kagan, "Equality and Desert," p. 304.
(21) Hurka, "The Common Structure of Virtue and Desert," p. 26.
(22) Ibid., 27. As a side note, black murderers are in fact underrepresented on death row relative to white murderers. For a study showing this from 1929 to 1966, see Gary Kleck, "Racial Discrimination in Criminal Sentencing: A Critical Evaluation of the Evidence with Additional Evidence on the Death Penalty," American Sociological Review 46 (1981), pp. 783-805. There have been later studies that confirmed that this pattern continues, e.g., Sheldon Eckland-Olson, "Structured Discretion, Racial Bias, and the Death Penalty: The First Decade after Furman in Texas," Social Science Quarterly 69 (1988), pp. 853-73, although these studies often attempt to explain away this pattern as due to other factors. However, given the history of racism, one can easily imagine a scenario where this was not the case.
(23) Hurka actually focuses on pleasure and pain but it is more plausible to leave the door open for other things that make a person's life go well; see Hurka, "The Common Structure of Virtue and Desert," pp. 6-31.
(24) Noah Lemos, Intrinsic Value, chap. 2; Ramon Lemos, The Nature of Value (Gainesville: University of Florida Press, 1995), chap. 2. For global intrinsic value, their theories would have to be changed so that they are borne by a maximally comprehensive state of affairs that obtains, where the maximally comprehensive aspect entails that the state of affairs contains or excludes every other state of affairs. This latter point comes from Alvin Plantinga, The Nature of Necessity (New York: Oxford University Press, 1974).
(25) For an argument that virtue grounds desert see, e.g., Thomas Hurka, Virtue, Value, and Vice (New York: Oxford University Press, 2001), chaps. 1-3; Hurka, "The Common Structure of Virtue and Desert," pp. 6-31; Kristjan Kristjansson, "Justice, Desert, and Virtue Revisited," Social Theory and Practice 29 (2003), pp. 39-63; Eric Moore, "Desert, Virtue, and Justice," Social Theory and Practice 26 (2000), pp. 41742. Moore and Kristjansson argue that virtue alone does so. Some recent discussion of whether capitalist acts ground desert can be seen in Stephen Kershnar, "Giving Capitalists Their Due," Economics and Philosophy 21 (2005), pp. 65-87; John Christman, "Entrepreneurs, Profits, and Deserving Market Shares," Social Philosophy & Policy 6 (1988), pp. 1-16; David Schweickart, "Capitalism, Contribution, and Sacrifice," in John Arthur and William H. Shaw, eds., Justice and Economic Distribution, 2d ed. (Englewood Cliffs, NJ: Prentice Hall, 1991), pp. 168-81; N. Scott Arnold, "Why Profits Are Deserved," Ethics 97 (1987), pp. 387-402.
(26) I am grateful to Neil Feit, Thad Metz, and George Schedler for their extremely helpful comments and criticisms.