Hostname: page-component-74656f9884-rdgjk Total loading time: 0 Render date: 2026-11-06T23:59:31.735Z Has data issue: false hasContentIssue false
  • English
  • Français

Two Kinds of Value Pluralism

Published online by Cambridge University Press:  26 January 2016

MILES TUCKER
MILES TUCKER*
Affiliation:
University of Massachusetts, Amherst mgtucker@philos.umass.edu
Rights & Permissions [Opens in a new window]

Abstract

I argue that there are two distinct views called ‘value pluralism’ in contemporary axiology, but that these positions have not been properly distinguished. The first kind of pluralism, weak pluralism, is the view philosophers have in mind when they say that there are many things that are valuable. It is also the kind of pluralism that philosophers like Moore, Brentano and Chisholm were interested in. The second kind of pluralism, strong pluralism, is the view philosophers have in mind when they say there are many values, or many kinds of value. It is also the kind of pluralism that philosophers like Stocker, Kekes and Nussbaum have advanced. I separate and elucidate these views, and show how the distinction between them affects the contemporary debate about value pluralism.

Information

Type
Research Article
Information
Utilitas , Volume 28 , Issue 3 , September 2016 , pp. 333 - 346
Check for updates
Copyright
Copyright © Cambridge University Press 2016 

Some philosophers say value pluralism is the view that there are many things that are valuable.Footnote 1 Others say value pluralism is the view that there are many values.Footnote 2 I think these two descriptions point to two different kinds of pluralism, both of which are present in contemporary value theory. But these views have not been properly stated, or distinguished.

In what follows, I present and explain these two concepts of value pluralism. I then defend an account of the distinction between them. I close by showing how this distinction affects contemporary arguments about pluralism.

WEAK PLURALISM

Let's start with the description of value pluralism that is better known.Footnote 3 According to this description, value pluralism is the view that there are many things that are intrinsically valuable. Conversely, value monism is the view that just one thing is intrinsically valuable. Thus many philosophers say that hedonism is a form of monism because it implies that ‘nothing is good but pleasure.’Footnote 4 But these philosophers would say that if we believe that pleasure, justice and beauty are intrinsically good, then we are value pluralists. Let's call this position weak intrinsic value pluralism.

There are many historical examples of such views: the axiologies of Moore, Brentano and Ross come to mind.Footnote 5 But we can find contemporary cases too: Thomas Hurka's perfectionist theory implies that knowledge, pleasure and virtue are intrinsically good.Footnote 6 And the theory Noah Lemos has advanced entails that pleasure, knowledge, beauty and the flourishing of non-sentient life are intrinsically valuable.Footnote 7

However, trouble arises when we think carefully about why these theories are supposed to be pluralistic. Philosophers often appeal to the standard description. They say that:

Plural Bearers: An axiology is a form of weak pluralism just in case it entails that more than one thing is intrinsically good.

But such accounts cannot succeed. One purpose of an axiology is to determine the intrinsic values of lives, outcomes and possible worlds. But there are many lives. Even a pessimist must admit that more than one is intrinsically good. The same problem arises with outcomes, and possible worlds. Surely, more than one has intrinsic value. But that means that every axiology is a form of value pluralism.Footnote 8

Plural Bearers therefore describes an uninteresting, vacuous view. Philosophers must be talking about something else when they talk about value pluralism.

Some have recognized this problem. They say that views like Moore's are not forms of pluralism because they imply that more than one thing is good. Rather, they are forms of pluralism because they imply that more than one kind of thing is good. Thus Zimmerman writes that ‘pluralism with respect to intrinsic value’ is the view that ‘there are irreducibly many different types of states that have intrinsic value.’Footnote 9 And Jonas Olson says that ‘pluralism about value is the view that more than one kind of thing are finally valuable.’Footnote 10 According to such views:

Plural Kinds: An axiology is a form of weak value pluralism just in case it entails that there is more than one kind of thing that is intrinsically good.Footnote 11

However, this revision is also unacceptable. As I've said, I believe that an axiology should determine the intrinsic values of lives, outcomes and possible worlds. I assume that every plausible theory will attribute intrinsic value to at least one entity in each of these categories. But lives, outcomes and possible worlds are different kinds of things. So this revised account still implies that every plausible axiology is a form of value pluralism.

I think we must conclude that neither the standard view nor its revision describes the position philosophers have in mind when they talk about value pluralism. We need a different account.

Good-making properties

A complete axiology should be able not only to identify the things that are intrinsically good but also to explain why those things are good.Footnote 12 To do so, the theory must select a set of good-making properties. Intuitively, these are the properties that provide the ultimate explanation of the goodness of things.Footnote 13 More rigorously, we may say that to be a good-making property is to be a member of the minimal set S such that for anything that is intrinsically good either (i) that thing is intrinsically good because it instantiates one of the properties in S or (ii) that thing is intrinsically good because it has a part that instantiates one of the properties in S.Footnote 14

Let me give an example. Imagine that a hedonist makes a list of all the things that he thinks are intrinsically good. Imagine we ask him about each item on the list: why did you list this thing? It seems that he would say in each case either (i) it goes on the list because it is an episode of pleasure or (ii) it goes on the list because it is a complex good, like a life, or outcome, and it contains episodes of pleasure. When he gets to the end of the list, he will have explained the goodness of each thing in terms of just one property, being an episode of pleasure. This is the only property he needs to explain the goodness of things. It is therefore the only property he takes to be good making.

I believe that the question of value pluralism is not ‘how many good things are there?’ nor ‘how many kinds of good things are there?’ Those questions are not substantive. Rather, the question is ‘how many good-making properties are there?’ If we say there is just one, we are monists. If we say there is more than one, we are pluralists.Footnote 15 That is:

Plural Good Makers: An axiology is a form of weak intrinsic value pluralism just in case it entails that there is more than one good-making property.Footnote 16

This view gets our test cases right. We have seen that it provides the right verdict about hedonism. And it provides the right judgements about classic examples, like Moore's axiology: it is a form of pluralism, as it should be. Some things would be on Moore's list because they are instances of pleasure, some because they are instances of beauty, some because they are instances of certain kinds of relationships. And we can say something similar about other examples of weak value pluralism, such as Brentano's theory, and Hurka's view.

STRONG PLURALISM

There is another view called ‘value pluralism’. This is the view that philosophers advance when they say there are irreducibly many values, or kinds of value. And it is the kind of pluralism Berlin, Kekes, Stocker and Nussbaum have defended.Footnote 17 I'll call it strong intrinsic value pluralism.

Strong pluralism is not the view that there are many things that are intrinsically good. Nor is it the view that there are many reasons why things are intrinsically good. Rather, it is the view that there are many kinds of intrinsic goodness.

Strong value pluralists say they find inspiration in Aristotle. Stocker and Kekes point to Book I of the Nicomachean Ethics:Footnote 18

Let us separate, then, things good in themselves from things useful, and consider whether the former are called good by reference to a single Idea. . .[If this is true] the account of the good will have to appear as something identical in them all as that of whiteness is identical in snow and in white lead. But of honour, wisdom, and pleasure, just in respect of their goodness, the accounts are distinct and diverse. The good, therefore, is not some common element answering to one Idea.Footnote 19

Strong pluralists interpret this passage as follows. Imagine two very different intrinsic goods. Perhaps one is an instance of pleasure and the other is an instance of wisdom. We say both are intrinsically good. But they are good in different ways; there is a kind of goodness the one has that the other does not.

Strong pluralists therefore claim that there are irreducibly many kinds of intrinsic goodness.Footnote 20 Each kind of intrinsic goodness is importantly different from the rest and has its own unique force. ‘Intrinsic goodness’ is therefore an umbrella term; it picks out a family of different types of ethical value. Thus we may say that:

Irreducible Values: An axiology is a form of strong value pluralism just in case it entails that there are at least two irreducible kinds of intrinsic value.

But what does it mean for a kind of intrinsic value to be irreducible? Mason suggests that a kind of value is irreducible just in case it is unanalysable – i.e. that it cannot be broken into more fundamental concepts.Footnote 21 Thus, she says, a person will be a strong pluralist if she believes in many unanalysable kinds of intrinsic value; she will be a strong monist if she believes in just one.

But such accounts are unacceptable; they confuse metaethics with axiology. Consider philosophers like Brentano and Zimmerman: they accept only one kind of intrinsic goodness but hope to analyse this kind of goodness in terms of what it is intrinsically fitting to favour, or love.Footnote 22 All such philosophers will fail to be either monists or pluralists on Mason's account, as their conceptions of intrinsic goodness are amenable to analysis. The same will be true of those who hope to analyse intrinsic goodness in terms of some natural property or relation.

We should, then, restrict Mason's view; we should say that a kind of goodness is irreducible only if it cannot be analysed in terms of some other kind of goodness. This is the kind of irreducibility that is critical to axiology. After all, if the monist's one kind of goodness were constructed out of others, he would be a pluralist: his one kind of goodness would contain many.

Thus we may say that a person is a strong monist just in case he believes in just one kind of intrinsic goodness, and this kind of intrinsic goodness cannot be analysed in terms of other kinds of goodness. Conversely, a person is a strong pluralist just in case he believes in more than one kind of intrinsic goodness, and these kinds of goodness cannot be analysed in terms of each other, or in terms of some ‘super-value’.

Incommensurability and pluralism

However, strong pluralists do not claim only that there are irreducibly many kinds of value. These different kinds of value are also supposed to be incommensurable.

To say that two quantities are incommensurable is to say that, in principle, they cannot be correctly measured on a common scale.Footnote 23 For example, heat and intelligence are incommensurable: there is no way to compare some amount of heat with some amount of intelligence.Footnote 24 The issue is not epistemic: we could know everything about heat and intelligence but we would still not know how to compare these quantities.

Similarly, to say that values are incommensurable is to say that, in principle, they cannot be measured on a common scale. Again, the issue is not epistemic: even omniscient agents could not compare the values of incommensurable goods.

Why do strong pluralists accept value incommensurability? Kekes writes:

The reasons why pluralists suppose that values are incommensurable are, first, that it does not seem to them that there is a highest value. . .to which all other values could always be reasonably subordinated and with reference to which all other values could be authoritatively ranked. Second, they are also dubious about there being some medium. . .in terms of which all the different values could be expressed, quantified, and compared.Footnote 25

Galston, another prominent pluralist, suggests that incommensurability is a sufficient condition for pluralism:

I distinguish value pluralism from various forms of nonpluralist accounts of morality. A theory is nonpluralist, I say, if it either (a) reduces goods to a single measure of value or (b) creates a comprehensive hierarchy or ordering among goods. (Theories that do (a) are usually called monistic.) A moral theory is pluralistic if it does neither (a) nor (b).Footnote 26

And Stocker goes further, suggesting that to say that values are plural just is to say that they are incommensurable.Footnote 27

Thus, according to many strong pluralists, value incommensurability is closely tied to pluralism. And for some, incommensurability seems to be a necessary and sufficient condition for their view.

Begin with the necessity claim. I believe the strong pluralist reasons as follows: suppose there are many kinds of intrinsic goodness and that these many kinds of goodness cannot be analysed in terms of each other. Then how could there be some scale that measures these different kinds of value? Such a scale would have to measure at least two distinct qualities. But this is impossible.Footnote 28 Indeed, if any comparison between these kinds of intrinsic goodness were possible then there would have to be some kind of value to compare them in terms of, some super-value. And this is exactly what the strong pluralist denies.

Now in the other direction: assume value pluralism is false. Then there is only one kind of intrinsic goodness and, for any thing that is intrinsically good, it must possess some determinate amount of this kind of goodness. But two amounts of the same quality can be compared. Thus if strong value pluralism is false, so is incommensurability. If we take the contrapositive, we get our result.Footnote 29

We might then conclude that incommensurability and strong value pluralism are simply different sides of the same coin. We would claim:

Incommensurable Values: An axiology is a form of strong value pluralism just in case it entails that there are at least two kinds of intrinsic value that are incommensurable.

An objection: incommensurability in a monist system

Still, we should be careful before endorsing such views. Some might object to the sufficiency claim: must intrinsic goodness come in determinate amounts? Consider again views like Brentano's. Brentano believes in just one kind of intrinsic goodness. But he suggests that while both ‘acts of insight’ and ‘high-minded love’ possess the same kind of intrinsic goodness, there is ‘no criterion [of judgment] available to us’ and thus there may be ‘no way of comparing the intrinsic value of acts of insight. . .with acts of high-minded love’.Footnote 30

Of course, Brentano's point might be merely about what is possible in practice: he might be claiming only that, while there are facts about how to compare the values of insight and love, we rarely, if ever, have epistemic access to these facts and so cannot hope to compare these goods.Footnote 31

But suppose that Brentano's claim is not merely epistemic: suppose instead that he believes that the values of insight and love cannot be compared in principle – that while both goods possess intrinsic value, they do not possess any determinate amount of value. We might think this impossible, but we should not be too hasty. Consider classical cases of vagueness: it may be vague, or indeterminate, whether a person is bald or hirsute, or whether it is day or twilight. Similarly, we might argue, it could be vague or indeterminate whether insight is better than love.Footnote 32

If this is possible, then goods need not possess determinate amounts of goodness. This will undermine our argument for the sufficiency claim. But, more worryingly, it also shows that the sufficiency claim is false: if it is indeterminate whether insight is better than love, then we cannot compare these goods – it is neither true nor false that one is better than another. Thus, pace Incommensurable Values, incommensurability may arise within a monistic system.

Could it be vague or indeterminate whether insight is better than love? Most claim that vagueness is merely linguistic.Footnote 33 The superevaluationist model is pervasive and attractive; Lewis writes:

The only intelligible account of vagueness locates it in our thought and language. The reason it's vague where the outback begins is not that there's this thing, the outback, with imprecise borders; rather there are many things, with different borders, and nobody has been fool enough to try to enforce a choice of one of them as the official referent of the word ‘outback’. Vagueness is semantic indecision.Footnote 34

According to Lewis's proposal, we have not decided, for example, what we mean by ‘bald’; it is ‘hyper-ambiguous’ – we might mean having fewer than 1,000 hairs, fewer than 999, and so on. We have no reason to be specific, to demand an ‘official referent’. This hyper-ambiguity engenders indeterminacy: claims about baldness are neither true nor false when they hold for only some candidate meanings (or ‘precisifications’) of ‘bald’. Similarly, then, the claim that insight is intrinsically better than love could be indeterminate if it holds only according to some precisifications of ‘intrinsically better than’.

But this is not plausible. While we have no reason to pick a particular candidate for the meaning of ‘bald’ or ‘twilight’, we must demand a single referent for ‘intrinsically better than’ – it plays a critical role in axiology, practical reasoning and deontology; hyper-ambiguity here would shake the foundations of these fields. Further, while it is easy to specify the different candidate meanings for ‘bald’ or ‘twilight’ this is not true in our case; what are the relevant candidates for ‘intrinsically better than’?

So the orthodox linguistic accounts of vagueness will not allow us to claim that it is vague or indeterminate whether insight is better than love. The alternative is metaphysical – we may say that it is indeterminate whether insight is better than love because the relation intrinsically better than is itself vague. But what would it mean for a universal like better than to be vague, or indeterminate? Williamson suggests that a relation might be vague if the relata can stand in the relation to a degree.Footnote 35 This is not to say that the relata entirely instantiate some determinate of the relation – like being intrinsically better by degree n – but that the relata may only instantiate the determinable intrinsically better than relation to some degree.

I am inclined, however, to side with the orthodoxy: I think this kind of vagueness is unintelligible.Footnote 36 Russell writes: ‘nothing is more or less than what it is, or to a certain extent possessed of the properties which it possesses.’Footnote 37 The same, I believe, holds of relations.Footnote 38 I therefore accept Incommensurable Values, and our argument in favour of it: I claim that if a thing is intrinsically good, it must possess some determinate amount of goodness.

But of course some philosophers are willing to accept this kind of ontological vagueness.Footnote 39 I cannot hope to dissuade those who endorse such views here. For such readers, I suggest that we depart from the claims of Stocker and other like-minded pluralists and reject the sufficiency condition: we will say only that if someone is a strong pluralist, then she must believe in value incommensurability. This will not affect our first criterion: we can still say that a person is a strong pluralist just in case she believes there are irreducibly many kinds of intrinsic goodness. And in conjunction with the necessity claim of Incommensurable Values, this will be enough: we will have a rich understanding of the strong pluralist view, and we will be able to explain why it has been so closely connected with value incommensurability.

DRAWING THE DISTINCTION

I have presented two different kinds of value pluralism. One view says that there are at least two properties that can make a thing intrinsically good; the other says that there are at least two irreducible kinds of intrinsic goodness. It is critical to keep these theories apart; they have radically different entailments.

For example, as we have said, strong pluralism necessitates value incommensurability. This requires a non-standard model of rational choice: if this kind of pluralism is true then we can no longer weigh or even compare the values of many of our ends. Of course sometimes our goals may be compatible, or they may possess only one kind of value. But problems will necessarily reappear: our ends will sometimes be inconsistent, and their values incommensurable. In such cases, it is unclear how we could act rationally.

Strong pluralists argue that such problems need not defeat their theory; Isaiah Berlin is here foundational. He writes: ‘Where ultimate values are irreconcilable, clear solutions cannot, in principle, be found. To decide rationally in such situations is to decide in the light of general ideals, the over-all pattern of life pursued by a group or society.’Footnote 40 Galston and Kekes make similar comments – although they appeal not to the life patterns of groups but to ‘shared human experience’ or the ‘traditions and conceptions’ we regard as acceptable.Footnote 41 According to all such views, it is these shared standards that will, somehow, help us rank our ends. But there is nothing normative in such theories: they appeal only to the standards we in fact have.Footnote 42 Further they ask these standards to do the impossible: to reasonably compare our ends without comparing their values.

Strong pluralism is also inconsistent with many standard accounts of right action. Any view that enjoins us to maximize the good, like Moore's ideal utilitarianism, is incoherent if strong pluralism is true. This is because it is not possible to maximize incommensurable values. Thus nearly all consequentialist views will be eliminated. A similar argument will eliminate moderate views, like Ross's.

Strong pluralism is therefore a revolutionary – and costly – position. But weak pluralism has none of these costs: it is consistent with a standard account of rational choice, and with all traditional accounts of right action.

These two views are, then, very different. But they have nonetheless been confused – and this confusion has influenced the contemporary dialectic about pluralism. For example, in The Morality of Pluralism, Kekes hopes to show that strong pluralism is superior to its competitors.Footnote 43 He begins by focusing on hedonistic theories. He is moved by Millian concerns: he thinks pleasures of the same intensity and duration may differ in value, due to their quality. But Kekes rejects Mill's theory too.Footnote 44 He moves to preferentism but cannot accept this either: persons can have ‘perverse, trivial, foolish, and self-destructive preferences’ and the satisfaction of such preferences is not intrinsically good.Footnote 45 So, Kekes says, preferentism fails as well.

But from the failure of these two theories, Kekes concludes that we have good reason to accept ‘the incompatibility and incommensurability of conflicting values’ – that is, good reason to accept strong pluralism.Footnote 46 However we can see now that such arguments cannot succeed. Classical hedonism and preferentism are not the only alternatives to strong pluralism. These views are monistic in both senses: they postulate one kind of intrinsic goodness and one property that gives a thing that kind of goodness. But we need not endorse such a theory to reject strong pluralism.Footnote 47 We could accept a view like Moore's – a view that is a form of weak pluralism but not strong pluralism. Further, as we have discussed, views like Moore's can be combined with a thesis of practical (or epistemic) incommensurability. According to such theses, while there are always facts about how the values of goods compare, we do not have reliable epistemic access to these facts. This will allow us to simulate much of the pluralist view, without committing ourselves to the heavyweight kind of incommensurability that the strong pluralist accepts. We may therefore help ourselves to the rich view of the moral life that the pluralist desires, without many of the costs.

Kekes, like most strong pluralists, never addresses such views. Further, those pluralists who do discuss views like Moore's seem to misunderstand them. Stocker writes that he was ‘convinced by G. E. Moore. . .of the plurality and incommensurability of moral considerations’.Footnote 48 But Moore never accepts the kind of pluralism Stocker is interested in. In fact, he explicitly rejects it.Footnote 49 And rightly so: the kind of pluralism Stocker has in mind would rule out Moore's own signature view about moral obligation.Footnote 50

There are other places where the distinction drawn here may be useful. But I hope this is sufficient to show just how important it is to distinguish these two kinds of views.Footnote 51

References

1 See e.g. Chisholm, Roderick, Brentano and Intrinsic Value (New York, 1986), p. 4 Google Scholar; Lemos, Noah, Intrinsic Value: Concept and Warrant (New York, 1994), p. 99 CrossRefGoogle Scholar; Elinor Mason, ‘Value Pluralism’, Stanford Encyclopedia of Philosophy, <http://plato.stanford.edu/archives/sum2015/entries/value-pluralism/> (2015), p. 6.

2 See e.g. Stocker, Michael, Plural and Conflicting Values (New York, 1990), pp. 167–8Google Scholar; Kekes, John, The Morality of Pluralism (Princeton, 1993), p. 17 Google Scholar; Nussbaum, Martha, The Fragility of Goodness: Luck and Ethics in Greek Tragedy and Philosophy (New York, 2001), p. xxix CrossRefGoogle Scholar.

3 My interest in this article is pluralism about value. But even within value theory there are many kinds of pluralism. I want to discuss pluralism about intrinsic value, rather than pluralism about extrinsic value, prudential value, legal value, or some other kind of value. Some philosophers believe that the concept of intrinsic value should be replaced with some other concept, such as final value. I take no stance on this issue here: the term ‘final value’ may be freely substituted for the term ‘intrinsic value’.

4 Moore, G. E., Principia Ethica, rev. edn, ed. Baldwin, T. (New York, 1993), p. 111 Google Scholar.

5 See Moore, Principia; Brentano, Franz, The Origin of our Knowledge of Right and Wrong, trans. Chisholm, R. and Schneewind, E. (London, 1969)Google Scholar; Ross, W. D. The Right and the Good, ed. Stratton-Lake, P. (New York, 2009)Google Scholar.

6 Hurka, Thomas, ‘Monism, Pluralism and Rational Regret’, Ethics 106 (1996), pp. 555–75CrossRefGoogle Scholar.

7 Lemos, Intrinsic Value, chs. 5–6.

8 Fred Feldman makes a similar point; see his ‘Basic Intrinsic Value’, Philosophical Studies 99 (2000), pp. 319–46.

9 Zimmerman, Michael, The Nature of Intrinsic Value (Lanham, 2001), p. 173 Google Scholar.

10 Olson, Jonas, ‘Intrinsicalism and Conditionalism about Final Value’, Ethical Theory and Moral Practice 7 (2004), pp. 31–52 CrossRefGoogle Scholar, at 49. Olson's view is about final value, but we can translate; see n. 3. See also Darwall, Stephen, ‘How Should Ethics Relate to (the Rest of) Philosophy?’, Southern Journal of Philosophy 41 (2003), pp. 1–20 CrossRefGoogle Scholar, at 4; Lemos, Intrinsic Value, pp. 67, 99.

11 Of course, to say that a kind of thing is intrinsically good is not to attribute intrinsic value to the kind itself, but rather to some member(s) of that kind.

12 There is an important parallel here between axiology and the normative ethics of behaviour. As Ross reminds us, we do not want to know merely which actions are morally right – we also want to know why. Thus a theory of right action that gives only necessary and sufficient conditions for the moral rightness of actions is necessarily incomplete. Such theories need to be supplemented with some claim about what makes right actions right – some account of the right-making features of actions. I believe that something similar is true in axiology.

13 We may wish to make an exception for mixed goods, i.e. those things that are intrinsically good but have parts that are intrinsically bad. We may wish to say that such things are not good because of their good-making properties, but because their good-making properties defeat their bad-making properties. Thus, on the account I prefer, hedonism does not entail that containing more pleasure than pain is a good-making property. Rather hedonism entails that containing more pleasure than pain is the property something has when its good-making properties outweigh its bad-making properties.

14 It may be necessary to insist that no disjunctive or otherwise gerrymandered value properties appear in the set, depending on how we understand the because of or in virtue of relation. I do not believe that this restriction is ad hoc. The good-making properties of a thing should provide the ultimate explanation of its goodness. But the explanation of a thing's goodness cannot end with a disjunctive property – a thing instantiates a disjunctive property only because it instantiates one or more of its disjuncts. For this reason, I do not believe that disjunctive properties can be good making.

15 Suppose a philosopher puts some things on his list because they are episodes of pleasure containing 10 hedons, others because they are episodes of pleasure containing 11 hedons, and so forth. Is such a person a value pluralist? No. While our philosopher cites many good-making properties, they are all degrees of the generic property being an episode of pleasure (containing n hedons). When I speak of good-making properties above, I mean to speak of these generic properties. We can say then that our philosopher's theory is a form of monism because it entails that there is just one generic good-making property, such that all the specific, degreed good-making properties cited are degrees of this generic property.

16 Hurka provides a similar, though much less detailed, proposal; see his ‘Monism’. My account is also similar to the elegant solution offered by Feldman; see his ‘Basic Intrinsic Value’. Indeed, in most cases Feldman's account provides the same judgements as the view I describe. But Feldman's theory is, I think, more complex. It is also apparently incompatible with a number of increasingly popular theories about intrinsic value, such as particularism about intrinsic value (i.e. the view that only concrete particulars bear intrinsic value). Further, it delivers counter-intuitive verdicts when combined with the thesis of organic unities. I therefore believe that the simpler view I state here is preferable for our purposes.

17 See Kekes, Morality; Stocker, Plural; Nussbaum, Fragility; Berlin, Isaiah, ‘Two Concepts of Liberty’, Liberty, ed. Hardy, H. (New York, 2002), pp. 166–217 CrossRefGoogle Scholar. Other philosophers endorse similar views: see e.g. Galston, William, The Practice of Liberal Pluralism (New York, 2005)Google Scholar; Taylor, Charles, ‘The Diversity of Goods’, Utilitarianism and Beyond, ed. Sen, Amartya and Williams, Bernard (New York, 1982), pp. 129–44CrossRefGoogle Scholar; Williams, Bernard, Moral Luck (New York, 1981)CrossRefGoogle Scholar; Hardy, Henry, ‘Taking Value Pluralism Seriously’, The One and The Many: Reading Isaiah Berlin, ed. Hardy, H. and Crowder, G. (Amherst, 2007), pp. 279–92Google Scholar.

18 See Kekes, Morality, p. 38 and Stocker, Plural, p. 168. Nussbaum's pluralism is also clearly inspired by Aristotle; see Fragility, pp. xxix, 294.

19 Aristotle, Nicomachean Ethics, trans. Ross, W. D., ed. Broadie, S. and Rowe, C. (New York, 2002), pp. 99–100 Google Scholar.

20 See Stocker, Plural, pp. 169, 184–94; Nussbaum, Fragility, p. xxix.

21 Mason, ‘Value Pluralism’, pp. 9–11.

22 Brentano, Origin, p. 18; Zimmerman, Nature, pp. 84–97.

23 I here follow Kekes's Morality, Kelly's ‘Impossibility’ and Klocksiem, Justin’s ‘Moorean Pluralism as a Solution to the Incommensurability Problem’, Philosophical Studies 153 (2011), pp. 335–50CrossRefGoogle Scholar: I assume that value incommensurability entails, or is equivalent with, value incomparability. To say that the values of two things are incommensurable is to say that their values cannot be correctly represented on a common scale. To say that the values of two things are incomparable is to say that these things do not stand in any axiological relation to each other (e.g. better than, worse than, etc.). I believe that incommensurability and incomparability cannot come apart. But a defence of this position would take us too far afield. However, if we do believe that these concepts can come apart, then I believe we should link strong pluralism to incomparability, as these commenters, and others, have done. See Mason's ‘Value Pluralism’ for a review of the use of the term in connection with strong pluralism.

24 I borrow this example from Chris Kelly. See his ‘The Impossibility of Incommensurable Values’, Philosophical Studies 137 (2008), pp. 369–82.

25 Kekes, Morality, p. 21.

26 Galston, Practice, p. 11.

27 Stocker writes: ‘I agree that if values are plural, they must be incommensurable, since I understand “plural values” to mean pretty much the same as “incommensurable values”.’ See his ‘Abstract and Concrete Value’, Incommensurability, Incomparability, and Practical Reason, ed. R. Chang (Cambridge, 1997), pp. 196–214, at 203. Hardy writes similarly that, ‘pluralism means . . . that ultimate human values are irreducibly many; that they cannot be translated into a single super-value; and that they are sometimes (or often) incommensurable’. See his ‘Taking’, p. 285.

28 Some might object. They might agree that it is impossible to create a scale that measures more than one quality but claim that we can always merely sum the quantities in question. For example, suppose there are two irreducibly distinct kinds of intrinsic goodness, G1 and G2. Surely, for anything that is intrinsically good, we can calculate the amount of G1 + G2 it has. However a scale of G1 + G2 is not a value scale. It is simply a scale of G1 + G2. Likewise, we could construct a scale that measures the sum of a person's temperature in Fahrenheit and their I.Q. But such a scale would not measure their temperature, or intelligence. Of course, one could insist that our G1 + G2 scale measures some new kind of value. But this will not help us to compare G1 and G2 unless both kinds of value can be correctly analysed as mere components of G1 + G2. And this is not possible: it would conflict with our assumption that strong value pluralism is true; G1 and G2 would not be fundamental kinds of goodness, but simply many components of one ‘super-value’.

29 Kelly presents a similar argument in ‘Impossibility’.

30 See Brentano, Origin, pp. 29–30, 33.

31 This is the reading of Brentano I prefer. See Origin, p. 30, especially the end of sect. 32.

32 I am grateful to an anonymous referee for advancing this powerful objection.

33 The other popular approach is epistemic: according to such views, in cases of vagueness it is not our language or the world that is vague or indeterminate – it is rather that we are, in a sense, necessarily ignorant. See Williamson, Timothy, Vagueness (London, 1994)Google Scholar for a powerful explication and defence of such views; Roy Sorensen provides a brief summary in his ‘Vagueness’, The Stanford Encyclopedia of Philosophy, <http://plato.stanford.edu/archives/win2013/entries/vagueness/> (2013). This epistemic approach to vagueness roughly corresponds to the practical, epistemic reading of Brentano given above. Such views are therefore inapplicable to the critic of Incommensurable Values – incommensurability, as we have understood it, is not epistemic.

34 See Lewis, David, On the Plurality of Worlds (Oxford, 1986), p. 212 Google Scholar.

35 See Williamson, Vagueness, p. 251. To be clear, Williamson does not believe in vagueness of this kind – he is only trying to make sense of it.

36 Michael Dummett for example, writes: ‘the notion that things might actually be vague, as well as being vaguely described, is not properly intelligible.’ See his ‘Wang's Paradox’, Synthese 30 (1975), pp. 301–24. This view is widely held – see e.g. Lewis, David, ‘Many, but Almost One’, Papers in Metaphysics and Epistemology (New York, 1999), pp. 164–82CrossRefGoogle Scholar; Sainsbury, Mark, ‘Why the World cannot be Vague’, Southern Journal of Philosophy 33 (1994), pp. 63–82 CrossRefGoogle Scholar.

37 Russell ‘Vagueness’, Australasian Journal of Philosophy 1 (1923), pp. 84–92, at 85CrossRefGoogle ScholarPubMed.

38 This worry seems especially acute in our case. Suppose that it is indeterminate whether insight is better than love because the better than relation is vague. To what degree, then, do these two goods stand in the better than relation? .5? .6? How could we ever answer such questions, even in principle?

39 See e.g. Colyvan, Mark, ‘Russell on Vagueness’, Principia 5 (2001), pp. 87–98 Google Scholar; Barnes, Elizabeth and Williams, J. R. G., ‘A Theory of Metaphysical Indeterminacy’, Oxford Studies in Metaphysics 6 (2011), pp. 103–48CrossRefGoogle Scholar.

40 Berlin, Isaiah, ‘Introduction’, Four Essays on Liberty (Oxford, 1969): p. ix–lxiii, at lGoogle Scholar.

41 See Galton, Practice, p. 15; Kekes, Morality, p. 76.

42 Kekes nearly admit this; he says that such rankings will be ‘relative but not arbitrary’. They are not arbitrary because these conceptions are open to rational criticism ‘at least on one ground; namely, on how they compare with respect to the realization of primary [i.e. intrinsic] values’ (Morality, p. 78). But such criticism would itself be groundless unless we can compare the values of the things these conceptions rank.

43 See Kekes, Morality, pp. 67–74. Kekes does consider other forms of ‘monism’ before coming to this conclusion, but these views are drastically different from the kind of views that philosophers like Moore and Lemos have advanced.

44 Kekes, Morality, pp. 67–8.

45 Kekes, Morality, p. 69.

46 Kekes, Morality, p. 73.

47 Of course, I do not mean to assent to the idea that by merely giving reason to reject classical hedonism and preferentism, Kekes has thereby shown that no theory that is monistic in both senses may succeed. Novel forms of hedonism and preferentism have been developed, and these theories are much more difficult to defeat. See e.g. the hedonistic views developed in Feldman, Fred, Pleasure and the Good Life (New York, 2004)CrossRefGoogle Scholar.

48 Stocker, Plural, p. 3.

49 Moore, G. E., ‘A Reply to my Critics’, Philosophy of G. E. Moore, 3rd edn, ed. Schlipp, P. A. (La Salle, 1968), pp. 533–677 Google Scholar, at 583.

50 It is also critical to state these views carefully; doing so may help dissolve contemporary disputes. The axiologies in Feldman's Pleasure are a good example. While Feldman's views are clearly not forms of strong pluralism, he claims that they are not forms of weak pluralism either. But Mason alleges in her ‘Value Pluralism’ that ‘Feldman's view is not a monist one.’ And Serena Olsaretti provides a similar argument; see her ‘The Limits of Hedonism: Feldman on the Value of Attitudinal Pleasure’, Philosophical Studies 136 (2007), pp. 439–50. If we apply our account of weak value pluralism we may hope to obtain a clear verdict. The same is true of historical debates: at least part of Moore's famous attack on qualified hedonism is his claim that Mill's theory is, in fact, a kind of pluralism and therefore cannot be a pure form of hedonism. Our theory can help us decide if we want to agree with Moore, or reject his complaint.

51 Many thanks to Eden Lin, Chris Heathwood, and audiences at New Mexico State University and the University of Massachusetts, Amherst for their thoughtful comments on earlier drafts of this article. I am especially grateful to Fred Feldman, Jean-Paul Vessel, Lisa Tucker, and an anonymous referee from Utilitas for all their help on this project.