Value Pluralism Research Paper

The metaphysical plurality of values has been thought to be important in two main ways. First, if values are plural, any theory that relies on value monism is mistaken. So, for example, hedonistic utilitarianism and most forms of preference-utilitarianism must be rejected. Second, the plurality of values is thought to raise problems for rational choice. If justice, for instance, is irreducibly distinct from mercy, how can there be rational choice between them? Without reduction, it seems there is no common ground that justifies choosing one value over another.

On closer inspection, however, it turns out that metaphysical or ‘reductive’ value pluralism does not have the implications for rational choice that many have supposed. A charitable interpretation of value pluralist writings suggests a second, ‘nonreductive,’ form of value pluralism. Nonreductive pluralism is neutral on the metaphysical question of plurality but insists that in the context of choice, there are differences between values—whether or not those values reduce to a single supervalue—that have important implications for rational choice. Since the differences claimed to hold between values vary from author to author, nonreductive value pluralism is not itself a particular view about values but merely a convenient rubric under which a loose collection of different views about values may be grouped.

This research paper examines the main arguments for reductive value pluralism, argues that reductive value pluralism does not have certain implications it is widely thought to have, and outlines three forms of nonreductive value pluralism.

1. Reductive Value Pluralism

Value pluralists maintain that whatever values are, there are ultimately many of them: they do not all reduce to a single ultimate value. Exactly how this metaphysical thesis is to be understood depends on how the notion of reduction is to be understood.

Reduction in the context of values is best understood as an explanatory relation: if one value reduces to another, what it is to bear the one value is fully explained by what it is to bear, promote, or respect the other value. This reduction is neutral on the general ontological question of whether there ‘really’ are any values at all. The pluralist maintains only that there are many values, whether or not they are to be regarded as entities in their own right.

Two paradigmatic relations of explanatory reduction are ‘is merely instrumental to’ and ‘is wholly constituted by.’ (Others include ‘is merely symbolic of,’ ‘is merely contributory to,’ ‘is merely a part of’). If one value is merely instrumental to another, there is nothing more to having the one value than promoting the value it is a means to. For example, if beauty is merely instrumental to pleasure, what it is to be beautiful is fully explained by the pleasure it brings. If one value is wholly constituted by another, there is nothing more to having the one value than being a way in which the other value is borne. For example, if the evening’s pleasure is wholly constituted by the thrill at the gaming tables, what it is to have the evening’s pleasure is fully explained by the thrill it is constituted by.

If all values reduce to a single value, that value is the only ultimate value, and value monism is correct. If however there is something more to two or more values than the values they are instrumental to, constituted by, etc., then those values are irreducibly distinct, and value pluralism is correct.

2. Three Arguments For Reductive Value Pluralism

Although there has been no agreement on which is the one value to which all others reduce, the view that there is such a value has two main attractions. If all values reduce to one, then values can be neatly systematized as instruments to, constituted by, etc., a single supervalue. Value monism appears to ensure a simple and elegant axiology, with a supervalue at the trunk of a structure that branches out to the other values that each derive from the supervalue in some way.

Moreover, if value monism holds, it seems that all conflicts between values are only apparent. For if there is ultimately only one value, then options for choice can be tidily arrayed according to how much of the supervalue they bear, promote, or respect. Any evaluative choice would ultimately be a choice between two amounts of the supervalue. Choosing between values would always be like choosing between two lumps of coal or three. Thus, value monism seems to dissolve the threats to practical rationality posed by tragic choices, moral dilemmas, and ‘incommensurable’ options.

Despite these attractions, most contemporary value theorists assume that value pluralism is true. There are three main arguments for pluralism. First is the intuitive implausibility of value monism: given the apparent diversity of values, how could there be a single value ‘common to’ all valuable items? Second is the thought that akrasia can be explained only if values are plural: how can it make sense for someone to choose something that she believes is worse overall unless there is something attractive about the worse option that is not ‘included’ in the better one? A closely related third argument maintains that some choice situations involve unavoidable loss: no matter which alternative one chooses, something valuable will be forgone, and thus, there must be plural values at stake in the choice.

2.1 Ordinary Intuition

Value monism seems to run afoul of common sense. Two intuitions suggest that any monistic account will be vulnerable to counterexample. First, how could there be just one value that runs through valuable items as diverse as, for example, achieving philosophical insight and eating a slice of delicious cheesecake? While perhaps the value of both options is instrumental to or constituted by some value (e.g., pleasure), it is hard to believe that bearing, promoting, or respecting that value is ultimately all there is to their being valuable. Second, if there were a single value that exhausted the value of all valuable things, the evaluative difference between things could always be given by some amount of the supervalue. But how could the difference in value between achieving philosophical insight and eating delicious cheesecake be a matter of quantity of some one thing? It is incredible to think that evaluative differences among diverse goods are just a matter of more or less of a single value.

To these charges, monists offer two replies. The first is to suggest that the troublesome intuitions rely on an uncharitable view of monism, one according to which what it is to bear the supervalue is to bear or bring about a kind of feeling or experience like pleasure or happiness. It is, indeed, hard to believe that all valuable things involve having or producing a pleasurable or happy experience. But monism need not be so crude. Some monists propose instead that the ultimate value is given by the satisfaction of one’s ‘fully-informed’ or ‘rational’ desires or preferences. The one property that runs through all valuable items, then, is the property of satisfying constrained desires, and it is the value of that that is the supervalue. Evaluative differences between items are a matter of the strength or number of desires they satisfy.

Whether desire-satisfaction accounts of the supervalue are plausible depends on what it is for a desire to be ‘fully-informed’ or ‘rational.’ ‘Full-information’ is usually understood as ‘having all the relevant facts and being free from logical error.’ ‘Rational’ desires might be procedurally rational (passing certain evaluatively-neutral tests) or substantively rational (passing certain evaluative tests). Insofar as ‘full-information’ or ‘rational’ is a value-neutral constraint, there is no guarantee that such constrained desires or their strengths will track what is intuitively valuable or more valuable: it is perfectly possible that everyone have fully-informed or procedurally rational desires to perform evil and malicious deeds, and that the desires for those deeds be stronger than desires for angelic ones. If, in an effort to secure this tracking, one places evaluative constraints on desires, as one might do by insisting that desires be substantively rational, then one must give up monism. For the values that operate as constraints on desires must be distinct from the value they constrain, and so there are ultimately many values, not one.

Monists have a second, more promising, reply. This reply was introduced by Mill in an effort to defend utilitarianism against counterintuitive consequences of Bentham’s quantitative hedonism. There are, Mill insisted, not only different quantities of pleasure, but different qualities of pleasure. The pleasure of philosophical insight is a ‘higher’ pleasure than the pleasure of eating delicious cheesecake. Thus, while there is ultimately only one value, there are different qualities of it that explain the seeming diversity of values. Moreover, the evaluative differences between valuable items are not simply given by some amount of the supervalue; achieving philosophical insight and eating cheesecake may differ in the quality of the supervalue they bear or instantiate—one pleasure is ‘higher.’

This reply gets to the heart of the dispute between monists and pluralists. The monist insists that the common-sense belief in multiple values is in fact a belief about different qualities or aspects of a single value. Whether this position is plausible depends on how values are to be individuated. What makes two considerations two distinct values as opposed to two qualities or aspects of a single value? Unfortunately, this question has received almost no philosophical attention. Until the question of how values are to be individuated is settled, ordinary intuitions cannot provide good reasons for thinking value pluralism is true.

2.2 Akrasia

Some philosophers (Wiggins 1978, Nussbaum 1986) have argued that for akrasia to be explained as a coherent phenomenon, plural values are required, and since akrasia is a coherent phenomenon, it follows that there are plural values. This argument is often presented as establishing the ‘incommensurability’ of values, but ‘incommensurability’ is used as synonymous with or as entailing plurality.

If one judges that one has most reason to choose one alternative but instead chooses the other, one is weak of will. Suppose one must choose between going to a party and staying home to work. Although one believes that staying home to work has the greater value—and therefore that one has most reason to stay home—one chooses to go to the party. How can such a choice be explained? If monism is correct, it seems that weakness of will must be an incoherent phenomenon. For suppose that there is a single ultimate value, say, pleasure. It seems, then, that there is nothing about the lesser option that could possibly attract one to it. For everything the lesser option has going for it—a certain amount of pleasure X—is included in what the greater option has going for it—X plus more pleasure to boot. One cannot choose to stay home for a reason, no matter how bad, for there is no reason for choosing the one option that is not already a reason for choosing the other.

The argument, however, does not succeed in establishing value pluralism. For akratic choice can be explained simply by attributing to the akrates a belief that there are plural values at stake; there need not actually be plural values. One might mistakenly believe that the value of going to the party is irreducibly distinct from the value of staying home and thus be attracted by the allure of the former. One’s reason for going to the party, while based on a false belief, could nevertheless be a reason that makes one’s weakness of will coherent.

But perhaps for akrasia to be coherent in an ‘objective’ sense, that is, coherent for agents with, inter alia, relevant beliefs that are true, plural values are required. This version, however, also fails. One possible difficulty is that what grounds the akrates’ attraction to the lesser option may be not the plurality of the values at stake but rather some contingent feature of the circumstances in which the value of the lesser option is instantiated or realized. So, for example, although there might ultimately be only pleasure at stake, the fact that the lesser pleasure of going to the party occurs in a seedy part of town has a special allure for the akrates. There are arguably no plural values here, only circumstances extrinsic to the supervalue whose special appeal to the agent might provide a reason for choosing the lesser option.

A more significant problem is the fact that attraction to the lesser option can be explained by appeal to different qualities or aspects of the supervalue. Take, for example, a choice between two glasses of wine, one laced with pepper and one not (Stocker 1990). Al-though one judges that the unadulterated wine will provide the greater pleasure, one is charmed by the particular pleasure of the wine laced with pepper and so chooses the lesser pleasure for the reason given by its particular charm. Similarly, one might choose to go to the party while judging that staying home is better for the reason that the party option provides a particular quality of the supervalue which staying at home lacks.

The success of this reply depends on an account of the individuation of values. Is what attracts about the lesser option a value distinct from the value of the greater option or some distinctive quality of an ultimate supervalue?

2.3 Unavoidable Loss

In some choice situations, it seems that no matter which alternative one chooses, something valuable will be lost. If, for example, one must choose between attending a lecture on Kant’s ethics and having cheesecake with friends, one is bound to lose out on value no matter which alternative one chooses—and not simply because one cannot have both. If one chooses to attend the lecture, one will forgo the gustatory pleasure of eating cheesecake; if one chooses to go out for cheesecae, one will forgo the philosophical insight one would have gained at the lecture. The fact that any choice in such situations entails value loss shows that the choice involves a conflict between distinct values. Therefore, there are plural values. (Essentially the same argument is sometimes put in terms of the possibility of rational regret over having foregone a lesser good.)

Monism, it seems, must deny that such cases exist. For if monism were true, then choice would always be a matter of choosing more or less of a single value. If one chooses the option with more value, there is no value loss. If one chooses the option with the lesser value, then there is value loss, but it is not unavoidable.

Both the argument from akrasia in its ‘objective’ form and the argument from unavoidable loss rely on the conviction that values are sometimes incompatible with one another: pursuit of (respect for, instantiation of, etc.) one excludes pursuit of (etc.) the other. It is the incompatibility of values that grounds the akrates’ attraction to the lesser option and that ensures that no matter which value one chooses, there will be value loss. This incompatibility of values might be conceptual, as Isaiah Berlin believed was true of value concepts like justice and mercy, or an intrinsic feature of values given the circumstances in which they arise. If values are incompatible, it seems that they cannot be reduced to a common value.

The monist reply here is already familiar from the previous two arguments. Sophisticated monists of the Millian variety might insist that there can be incompatibilities among different qualities or aspects of a single value. In cases of unavoidable loss, there need not be more than one ultimate value. The languorous pleasure of basking in the sun might be incompatible with the piquant pleasure of hearing unexpected good news: having the one pleasure now rules out having the other now though they are nevertheless instances of a single value, pleasure (cf. Stocker 1990). Similarly, the qualitative value of achieving philosophical insight may be incompatible with the qualitative value of eating delicious cheesecake. Choosing between the lecture and the cheesecake involves a loss in value no matter how one chooses, but there is only one ultimate value.

All three arguments for value pluralism crucially depend on the answer to the question, How are values properly individuated? The fundamental idea behind the common monist response to pluralist arguments is that there is a nonultimate sense of ‘value’ according to which it can be true that there are distinct ‘values’ and yet those values cannot be values in the robust sense implied by the claim that there are two distinct ultimate values.

Exactly how this is to be worked out remains to be seen. Hurka (1996) has suggested that one pleasure is not a different ultimate value from another pleasure if it does not differ in any of its ‘intrinsic’ features. In explaining what it is for a feature to be intrinsic to a value, Hurka urges that the location of a value is extrinsic to it. Thus, John’s pleasure and Joanna’s pleasure are two instances of the same value, pleasure. Stocker (1990), however, has suggested that if pleasure is located in different people, there are different ultimate values: there is the pleasure-for-John and the pleasure-for-Joanna. Stocker (1990, 1997) maintains that if it is rational to care about any difference between two evaluative considerations, it follows that those considerations are different ultimate values. Thus, there is sharp disagreement about how values are properly individuated.

Monism, if it is to be plausible, must hold that the correct account of value individuation leaves room for a tenable notion of a value quality or aspect that serves to undermine intuitions that values are plural, accounts for weakness of will, and explains away choices that seem to involve unavoidable loss. If there is such a notion, then value monism may yet win the day. But any such victory for monism may be hollow. For if monism is to be cluttered with complex relations among value qualities or aspects, the two chief attractions of monism—in short, the tidiness and simplicity it seemed to offer—fall by the wayside. In the end, then, the debate between value pluralists and value monists may be much ado about nothing.

3. Implications Of Reductive Value Pluralism

Assuming that there are ultimately many values, what, if anything, of interest follows? There are two main implications that are thought to be important.

First, if reductive value pluralism is true, then reductive value monism and any theory that relies on it must be rejected. So, for example, Benthamite utilitarianism has to go. Perhaps more significant is the result that preference utilitarianism, or, more generally, desire-satisfaction theories of value must go. (This is not to say that desire-satisfaction theories of what it is for something to be a value must be rejected. One might hold that every value must satisfy some or other desire to be a value while denying that all values reduce to the value of desire satisfaction.) Many economists and some philosophers favor desiresatisfaction accounts of value, and their accounts at the very least require modification if value pluralism is true. Economists who model rational choice on satisfaction of preferences, for example, must relinquish any claim that goodness is a matter of satisfaction of one’s preferences and instead maintain only that the rationality of choice depends on one’s preferences conforming to certain axioms (cf. Broome 1991). Philosophers who think that individual well-being is the only value there is and that what makes one’s life go best is satisfaction of one’s desires must give up either the claim that well-being is the only value there is or the claim that well-being is simply a matter of desire-satisfaction.

Some have thought that value pluralism shows that any form of utilitarianism is mistaken. But what is essential to utilitarianism is compatible with value pluralism (Sen 1981). Others have thought that value pluralism is the foundation of political liberalism: if values are metaphysically plural, then liberalism is the correct theory of justice. But the metaphysical plurality of values is plausibly neutral between liberalism and nonliberalism. The fact that there are ultimately many values does not entail that a state should not compel its citizens to pursue one value over others, nor does monism entail that a state should not protect its citizens’ choices to pursue nonultimate values in ways that do not best promote or respect the supervalue. As some political theorists have pointed out, liberalism itself might be understood as monistic about value: the ultimate value is the value of permitting people to pursue different nonultimate values. Whether there is, in the end, some indirect way in which value pluralism supports liberalism remains an open question.

Value pluralism has been thought to have a second implication. If value pluralism is correct, then difficulties supposedly follow for rational choice. The difficulties alleged are various, but the reason for thinking that none follows from value pluralism is the same. Many have supposed, for example, that alternatives for choice bearing plural values are incomparable. How can the option of respecting someone’s right to free speech be compared with an option that doubles everyone’s pleasure? Insofar as the value of a right to free speech is irreducibly distinct from the value of pleasure, it seems that there is no common basis on which to make a comparison between their bearers. The argument goes, if options for choice bearing plural values are incomparable, rational choice between them is precluded.

There is good reason, however, to think that the plurality of values does not entail the incomparability of options that bear them. Take any two putatively ultimate values such as the value of the right to free speech and the value of pleasure. One can always imagine some option that bears the one value in a notable way that can be compared with another option that bears the other value in a nominal way. So, for instance, an option that involves violating everyone’s right to free speech is worse than an option that involves reducing one person’s pleasure by a small amount. Given any two putatively irreducibly distinct values, there will always be some comparison between a notable bearer of the one value with a nominal bearer of the other value. The existence of ‘nominalnotable’ comparisons demonstrates that if alternatives are incomparable, it is not the plurality of values per se that entails their incomparability (Chang 1997). Assuming that rational choice depends on the comparability of alternatives, the plurality of values, then, does not preclude rational choice in this way.

Although value pluralism does not entail incomparability, perhaps the issue of value pluralism is important because if value monism is true, complete comparability follows. If there is ultimately only one value, evaluative differences between items must al- ways reduce to differences in amount or quality of the supervalue, and quantities or qualities of the same thing can always be compared. On some views, if alternatives are comparable, the existence of a rational choice is assured. Thus, monism seems to guarantee that there is a rational choice in every choice situation, no matter how intractable the conflict may seem.

The inference from monism to comparability, however, is mistaken. It is not clear that different qualities of a single value can be compared. How can one compare the luxurious, wallowing pleasure of lying in the sun and the intense, sharp pleasure of quenching a fierce thirst? Making the case for comparability is arguably just as hard under the assumption of monism as it is under the assumption of pluralism. Moreover, even assuming that the supervalue does not admit of different qualities, it is a mistake to assume that all quantities of a single supervalue are comparable. Pleasure can be greater in amount without the value of that pleasure being thereby greater. The five-course feast that ends with petit fours involves a larger quantity of pleasure than the five-course feast without, for the petit fours add a quantity of pleasure to the pleasure already had. But the additional quantity of pleasure may make the meal less valuable with respect to pleasure—the petit fours induce pleasure ‘overkill.’ Since a larger quantity of pleasure is not always better with respect to pleasure, it seems possible in principle that different quantities of a single value are incomparable. Assuming that there are such genuine cases of monistic conflict, the issue of value pluralism cuts across the question of whether bearers of value are comparable or incomparable.

4. ‘Nonreductive’ Value Pluralism

In general, the question of how many ultimate values there are is irrelevant to the question of how conflicts between them are to be resolved. What matters in conflict resolution is how the values relate to one another in the context of choice, not how they stand to one another metaphysically. As already suggested, a monist might happily allow conflicts between value qualities, while insisting that, as a metaphysical matter, there is ultimately only one value.

Given that reductive value pluralism does not have the implications for practical reason that many value pluralists seem to think it has, it is natural to think that there is some other notion of value pluralism besides the metaphysical one. There is, however, no single specific idea that would account for the concerns of all value pluralists who think the ‘plurality of values’ has implications for rational choice. At best, the cluster of views can be generally characterized as holding that there is some or other difference among values—something apart from mere metaphysical numerical difference—that has important implications for rational choice. The loose collection of views about the ways in which differences between values can have practical implications might be called ‘nonreductive’ value pluralism.

Nonreductive value pluralism comes in many different varieties, but there are three leading kinds (for general discussion see Griffin 1997). First is the view that some values are incompatible, that they cannot be pursued (instantiated, etc.) together. The incompatibility of values raises the possibility of unavoidable loss of value and, more alarmingly, the specter of moral dilemmas in which the unavoidable loss of value is expressed by a moral requirement.

Second is the view that some values are so different that they—and therefore their bearers—are incomparable. Suppose you must choose between a career as a clarinetist and one as a lawyer (Raz 1986). If the values of those careers are incomparable, and if their incomparability precludes rational choice between them, it seems that whichever career one chooses, the choice cannot be rational. Thus, this kind of nonreductive value pluralism has important implications for practical reasons. It suggests that there is perhaps a wide swathe of choices that are beyond the scope of practical reason.

A third view is that some values are so different that their bearers cannot be measured by the same cardinal scale of value (cf. Sunstein 1994, Anderson 1993, 1997). Suppose you must choose between guns and butter. Since there is no cardinal scale along which the value of guns and butter can be measured, the alternatives are incommensurable. Many pluralists have thought that rational choice between incommensurable options is impossible. But it is clear that the lack of cardinal comparability does not preclude rational choice. One alternative might be better than another though there is no cardinal scale according to which it is better. Lexically superior options, for instance, cannot be measured on the same cardinal scale as the options they are lexically superior to, but they are nevertheless better. Moreover, the failure of precise cardinal comparability need not entail incomparability. Items may be imprecisely comparable (Parfit 1984, Griffin 1986). Although incommensurable options do not preclude rational choice, the fact that they are incommensurable entails that any justification for choosing one over another cannot appeal to the idea that there is some amount of one that provides the equivalent measure of some amount of the other.

Bibliography:

1. Anderson E 1993 Value in Ethics and Economics. Harvard University Press, Cambridge, MA
2. Anderson E 1997 Practical reason and incommensurate goods. In: Chang R (ed.) Incommensurability, Incompatability and Practical Reason. Harvard University Press, Cambridge, MA
3. Berlin I 1950 Concepts and Categories. Oxford University Press, Oxford, UK
4. Berlin I 1959 The Crooked Timber of Humanity. Random House, New York
5. Broome J 1991 Weighing Goods: Equality, Uncertainty and Time. B. Blackwell, Cambridge, MA
6. Chang R 1997 Introduction. In: Chang R (ed.) Incommensurability, Incomparability, and Practical Reason. Harvard University Press, Cambridge, MA
7. Griffin J 1986 Well-Being: Its Meaning, Measurement and Moral Importance. Clarendon Press, Oxford, UK
8. Griffin J 1997 Incommensurability: What’s the problem? In: Chang R (ed.) Incommensurability, Incomparability, and Practical Reason. Harvard University Press, Cambridge, MA
9. Hurka T 1996 Monism, pluralism, and rational regret. Ethics 106: 555–75
10. Kekes J 1993 The Morality of Pluralism. Princeton University Press, Princeton, NJ
11. Nagel T 1979 Mortal Questions. Cambridge University Press, Cambridge, UK
12. Nussbaum M C 1986 The Fragility of Goodness. Cambridge University Press, Cambridge, UK
13. Parfit D 1984 Reasons and Persons. Oxford University Press, Oxford, UK
14. Raz J 1986 The Morality of Freedom. Oxford University Press, Oxford, UK
15. Richardson HS 1994 Practical Reasoning About Final Ends. Cambridge University Press, Cambridge, UK
16. Sen A 1981 Plural utility. Proceedings of the Aristotelian Society 81: 193–215
17. Stocker M 1990 Plural and Conflicting Values. Oxford University Press, Oxford, UK
18. Stocker M 1997 Abstract and concrete value: Plurality, conflict and maximization. In: Chang R (ed.) Incommensurability, Incompatability and Practical Reason. Harvard University Press, Cambridge, MA
19. Sunstein C 1994 Incommensurability and valuation in law. Michigan Law Review 92: 779–851
20. Taylor C 1982 The diversity of goods. In: Sen A, Williams B (eds.) Utilitarianism and Beyond. Cambridge University Press, Cambridge, UK
21. Wiggins D 1978 Weakness of will, commensurability and the objects of deliberation and desire. Proceedings of the Aristotelian Society 79: 251–77