Hostname: page-component-74656f9884-8r8nv Total loading time: 0 Render date: 2026-11-06T23:53:47.305Z Has data issue: false hasContentIssue false
  • English
  • Français

Some Difficult Intuitions for the Principle of Universality

Published online by Cambridge University Press:  12 November 2009

STEPHEN KERSHNAR
STEPHEN KERSHNAR*
Affiliation:
State University of New York at Fredoniastephen.kershnar@fredonia.edu
Rights & Permissions [Opens in a new window]

Abstract

The Principle of Universality asserts that a part retains its intrinsic value regardless of the whole in which it is a part or even whether it is part of a whole. The idea underlying this principle is that the intrinsic value of a thing supervenes on its intrinsic properties. Since the intrinsic properties remain unchanged so does the thing's intrinsic value. In this article, I argue that, properly understood, the Principle of Universality can handle seemingly troublesome intuitions about the relative intrinsic value of a vicious person having pain and his having pleasure. I specifically argue that the intuition that the former state is better is explained by the nature of the basic intrinsic-value states, which involve a person having a level of well-being and desert at a time. One implication of this is that given the nature of such basic intrinsic-value states, pleasure and pain are not value-bearing parts of virtuous and vicious attitudes.

Information

Type
Research Article
Information
Utilitas , Volume 21 , Issue 4 , December 2009 , pp. 478 - 488
Copyright
Copyright © Cambridge University Press 2009

I. THE PRINCIPLE OF UNIVERSALITY

Consider the Principle of Universality.

  1. (1) Principle of Universality. A part retains its intrinsic value regardless of the whole in which it is a part or even whether it is part of a whole.Footnote 1

The idea underlying this principle is that the intrinsic value of a thing supervenes on its intrinsic properties.Footnote 2 Since the intrinsic properties remain unchanged so does the thing's intrinsic value. The Principle of Universality and the underlying idea also entail that a bearer (a state of affairs) necessarily has its level of intrinsic value and that it would have this level were it to exist in isolation. In this article, I argue that, properly understood, the Principle of Universality can handle seemingly troublesome intuitions about pain and pleasure in a vicious person.

In this article, I assume that the bearer of intrinsic value is a state of affairs that obtains. States of affairs that do not obtain have intrinsic value in the counterfactual sense that were they to obtain, they would have this level of value.

Following Fred Feldman, I assume that there are basic intrinsic-value states, which are the core specification of those, and only those, facts that make a state of affairs have a particular level of intrinsic goodness.Footnote 3 For example, if intrinsic goodness supervenes on a person's having pleasure to degree n at a time t and pleasure and intrinsic value are related in a linear one–one ratio, then it is intrinsically good to +10 that Al has +10 units of pleasure at 12:30 a.m. on Saturday, November 4, 2006 regardless of the larger state of affairs in which that state is contained. The argument for a basic intrinsic-value state is that it prevents double counting with regard to goodness. For example, consider a possible world that contains the following.

Al's having +10 degree of pleasure at 12:30 a.m. on Saturday, November 4, 2006.

Someone's having +10 degree of pleasure at 12:30 p.m. on Saturday, November 4, 2006.

Al's having +10 degree of pleasure at some time.

If value didn't rest on, and only on, basic intrinsic value states then the value of this attitude would be overestimated. In fact, it would probably be infinite given the number of states of affairs (or, perhaps, distinct propositions) that are generated by the first state of affairs.

II. AN OBJECTION TO THE PRINCIPLE OF UNIVERSALITY

A. Assumptions

One reason to be concerned with the Principle of Universality is that it seems to get the wrong intuitions in several types of cases. In calculating the intrinsic value of a state of affairs, I begin with the following assumptions.

  1. (2) Principle of Non-Existing Objects. I assume that the object of a virtuous or vicious attitude can be a state of affairs that does not obtain.

For example, Smith can take pleasure in the thought of Jones's suffering even though the latter does not actually suffer.

  1. (3) Principle of the Value of Pleasure. Pleasure is an intrinsic good and its degree of intrinsic goodness lies in a one–one relation to the intensity of pleasure (or pain).

I will revisit this last assumption below.

  1. (4) Principle of Comparative Value of Virtuous Attitudes. The value of an attitude is always less than the value of the object at which it is directed.

Where the attitude and object differ in value in that one is positive and the other is negative, the attitude has a lesser absolute value than the object. Again, states of affairs that do not obtain have intrinsic value in the counterfactual sense that were they to obtain, they would have this level of value.

Thomas Hurka provides several convincing reasons to accept this last principle.Footnote 4 First, consider our intuitions in the following case.

Virtuous Compassion. Alice has compassion for Betty's suffering.

In this case, it intuitively seems that this state of affairs is bad. The best explanation of this result is that the negative value of Betty's suffering outweighs the positive value of Alice's virtuous attitude. Hurka's idea is that our intuitions in this type of case are generalizable across different combinations of intrinsic love and hate for basic intrinsic goods (e.g. pleasure and knowledge) and basic intrinsic bads (e.g. pain and false belief).

Second, this principle is necessary to satisfy our intuitions about the diminishing value of higher-order attitudes. For example, a person's (intrinsic) love of intrinsically good things (e.g. pleasure and knowledge) intuitively seems more valuable than his love of his love of intrinsically good things. This in turn is intuitively more valuable than his love of his love of his love of intrinsically good things.

Third, the principle explains how the love of different objects should be distributed. Consider these cases.

Torturer: A torturer takes sadistic pleasure in giving pain to his victim. We can eliminate either the torturer's pleasure or the victim's pain, but not both.

Teacher: An uncaring teacher's methods do not work. We can change the methods or the teacher's attitude, but not both.

In both cases, we should care more about the object of the vicious attitudes (i.e., the victim's pain and the pupils’ false beliefs) than the vicious attitudes themselves.Footnote 5 The best explanation for this is that the value of an attitude is always less than its object.

It is worth noting that if virtuous and vicious attitudes directed at non-existent objects have a value other than zero, as Hurka claims, then principles (2) and (4) conflict.Footnote 6 This is because a vicious attitude toward an imagined pleasure or pain has a negative value that outweighs the value of its object. Hurka might respond by viewing the value of the object as counterfactual value.Footnote 7 That is, the object has the value that it would have were the state of affairs to obtain. Let us accept this response.

Hurka extends the principle so that the value of an attitude on the basis of its being virtuous or vicious is outweighed by its value on the basis of its being a pleasure or pain. For example, he asks us to imagine the following case.

Blissful Bad Guys: In this world, all persons lead blissful lives but wish suffering upon others.

On Hurka's account, this world has positive value even though it has plenty of vice and no virtue.Footnote 8 His explanation for this is that the value of the attitudes’ being a pleasure outweighs the disvalue of the same attitudes being a vice.Footnote 9 Note that this does not directly follow from (4) because the pleasure in this case is not the object of the inhabitants’ attitudes. If this is correct, then we end up with the following principle.

  1. (5) Principle of Comparative Value of Attitudinal Properties: The value of an attitude on the ground of its being a pleasure or pain always outweighs its virtue- or vice-grounded value.

Hence, when the object of an attitude has positive intrinsic value and the attitude has negative intrinsic value, the sum of the two is positive; when the converse is true, the sum is negative.

An additional underlying assumption here is that independently grounded values should be summed up rather than combined via some other mathematical function. The idea is that this assumption best fits with the notion that these values are actual and rest on independent facts.Footnote 10 Let us accept this assumption.

B. The Principle of Universality gets the wrong result in the context of vicious persons

Consider the following case.

Case 1: Mean-Spirited Pain. Smith is a vicious person. He takes −10 pain in the thought of Jones's pleasure.

Given these assumptions, the value of Smith's attitude is the following.

Consider also the following case.

Case 2: Malicious Pleasure. Smith is a vicious person. He takes +10 pleasure in the thought of Johnson's suffering.

The problem is that this valuation isn't correct. If we had to choose between an overall vicious person having mean-spirited pain or malicious pleasure, we would think the former is better since it gets him closer to what he deserves. This intuition is a widespread one.Footnote 11 However, if so, then at least one of the above assumptions is false. If assumptions (2), (3), and (5) are solid, then one explanation for the wrong result is that the Principle of Universality is false. That is, (1) is false.

One who rejects this implication might argue that only attitudes and actions are directly vicious (or virtuous) and that persons are vicious only on the basis of having a large number or ratio of vicious attitudes or having done many vicious acts. That is, a person is only indirectly virtuous or vicious.Footnote 12 The argument for the claim that thoughts and actions are the bearers of virtue and vice is that it intuitively seems that motives, thoughts, and individual actions are virtuous or vicious regardless of whether they reflect the general mindset of the person who has them.Footnote 13 The responder then argues that the intuitions that support the above objection to (1) are mistaken because they rely on the value of other thoughts and this is irrelevant in determining the value of the particular pain in Mean-Spirited Pain and the particular pleasure in Malicious Pleasure. He concludes that we should compare the two attitudes in isolation and when we do, Mean-Spirited Pain is not better than Malicious Pleasure.

Even if the claim about virtue and vice depending on particular attitudes is correct, and I think it is, it is still a person who is the bearer of virtue and vice, albeit because of his thought content. That is, it is persons because of their thought content that have a particular level of virtue. The above examples capture the value of such a person with a particular desert at a time having a particular thought. And in such a case the above principles don't get the right result.

III. THE SOLUTION TO THIS PROBLEM

A. The Principle of Universality does not apply to attitudes

The principle of universality applies if the object of an attitude is part of the attitude. This allows it to retain its value regardless of the whole of which it is a part. That is, since the object (for example, Jones's being in pain) is a state of affairs that need not obtain, it doesn't have any value if it isn't part of the attitude. The problem is that the state of affairs that is the object of a virtuous or vicious attitude is not a part of the attitude.Footnote 14 It is not a paradigmatic part of a whole since in paradigm cases parts and wholes are concrete particulars. If we allow that one state of affairs is part of another, then the former is included in the latter.Footnote 15 In the context of attitudes, the state of affairs on which an attitude is focused is an essential feature of that attitude. This can be seen in that an attitude that focused on a different state of affairs would intuitively not seem to be the same attitude. For example, Smith's taking pleasure at the thought of Jones's suffering is a different thought than his taking pleasure in the thought of Johnson's suffering.Footnote 16 That is, attitudes are to be individuated in part by their objects.

Perhaps there is a principle analogous to the Principle of Universality that can be constructed to capture a relation similar to the Principle of Organic Unities. Here is such a principle.

  1. (6) Principle of Attitudinal Universality. An attitude retains its value regardless of the state of affairs at which it is directed.

My assumption here is that pleasure and pain have content. That is, they are always directed at states of affairs. The problem is that this principle doesn't seem to be true. The direction of an attitude (pleasure, pain, or indifference) and its content (the state of affairs on which it focuses) don't act separately to explain the value of Mean-Spirited Pain and Malicious Pleasure. This can be seen in that they generate the wrong results when we also assume that pleasure is an intrinsic good and pain an intrinsic bad. Hence, if we reject the Principle of Attitudinal Universality, then the Mean-Spirited Pain and Malicious Pleasure cases don't have a valuable part against which the organic-unity value conflicts.

B. The failure of the analogous principle is explained by the nature of the basic intrinsic-value state

The inapplicability of the Principle of Universality fits nicely with the notion of a basic intrinsic-value state. As mentioned above, basic intrinsic-value states allow us to identify the primary bearer of intrinsic value and avoid double counting in determining its level of intrinsic value. It also allows us to determine the intrinsic value of secondary bearers of value such as lives, possible worlds, and complex mental states. The basic intrinsic-value state is not a person with a level of well-being at a time, but rather a person with a level of well-being and desert at a time.Footnote 17 The idea here is that, by itself, we can't determine the amount of intrinsic value just from knowing that a person has a particular level of pleasure at a time. Nor can we determine the amount of intrinsic value just from knowing what a person deserves. Rather, we also need to know his level of well-being and desert. For example, the value of Ted having an ecstatic life is undetermined until we know whether Ted is virtuous or vicious. The basic intrinsic-value state in part explains why the above principles get the wrong results in the two key cases. This account of the basic intrinsic-value state also fits nicely with the absence of an attitudinal analogue to the Principle of Universality. The attitude, whether it is a pleasure or pain, does not provide a separate ground of value.

If this account is correct, then we can get the right results in the Mean-Spirited Pain and Malicious Pleasure cases because we do not attach independent values to the pain and pleasure. This account is independent of the Principle of Non-Existent Objects since it does not address what sort of things can be objects of virtuous and vicious attitudes. It is incompatible with the following principles: Principle of the Value of Pleasure, Principle of Comparative Value of Virtuous Attitudes, and Principle of Comparative Value of Attitudinal Properties because they misidentify the nature of the basic intrinsic-value state.

An objector might claim that my account does not provide an explanation for desert, for example, why it is intrinsically bad for a vicious person to flourish. My account is compatible with the claim that desert is a primitive relation and an explanation of desert in terms of a more general fittingness relation. Here I take no position on which is preferable.

C. Implications

There are several interesting consequences that flow from this result. First, in calculating intrinsic value, we should not simply add or subtract the value of the well-being or degree of virtue from the value of desert-adjusted well-being. Nor should we add or subtract the value of virtue or vice from the deserved. This suggests that several articles that use such a model misconstrue the way in which pleasure and pain function.Footnote 18

Second, if a level of well-being is part of the basic intrinsic-value state, then the degree to which desert is satisfied (e.g. a person gets what he deserved) is not by itself a ground of intrinsic value. This suggests that graphs which appear to treat desert-fit as an independent ground of value are mistaken.Footnote 19

Third, the Principle of Universality is true. That is, the parts of states of affairs retain their value independent of the whole of which they are parts. So, for example, the first state of affairs retains its value even when it becomes part of the second.

Smith's deserving and receiving +10 units of well-being at 12:30 a.m. on November 5, 2006.

Smith's deserving and receiving +10 units of well-being at 12:30 a.m. on November 5, 2006 and Jones deserving and receiving +20 units of well-being at 12:30 a.m. on November 5, 2006.

However, this principle doesn't do the work normally assigned to it. For example, it has been hypothesized that it is the misalignment of an intrinsic good (pleasure) directed at an intrinsic bad (another person's suffering pain or having false belief) that explains why the complex state consisting of the pleasure and its content is a net intrinsic bad.Footnote 20 The Principle of Universality has also been hypothesized to offer a unified explanation of why attitudes of different orders (e.g. loving evil and loving the loving of evil) and mismatches between desert and well-being (e.g. positive desert and pain) are intrinsically bad.Footnote 21 However, if my analysis is correct, these hypothesized relations are false because the separate components aren't intrinsically good or bad.

Fourth, the attempts to graph intrinsic value are plausible and the shape of these graphs is explained by a single intrinsic-value-making property, namely desert-adjusted well-being.Footnote 22 Furthermore, as can be seen in the graphs, the mathematical relation between intrinsic value, desert, and well-being is complex.

Fifth, whether the Principle of Universality is true is independent of whether the Principle of Organic Unities is true. This principle asserts that the intrinsic value of a whole is not necessarily equal to the sum of the intrinsic value of the parts. If the intrinsic value of a whole is equal to the sum of the intrinsically valuable parts, then the Principle of Organic Unities is false and yet at the same time the parts of a whole retain their intrinsic value regardless of the whole in which they are embedded.

Some theorists refer to the value of such organic-unity-like effects as the result of prima facie or virtual intrinsic value.Footnote 23 However, on the above account, a level of well-being in isolation or desert in isolation does not have prima facie or virtual intrinsic value. Consider Ross's analogy: just as an act of promise-keeping is something that is actually right if there are no other morally relevant characteristics, pleasure is actually good if there are no other morally relevant states. Regardless of whether the former is true, the latter is not true since pleasure alone can't constitute a basic intrinsic-value state. That is, since it lacks a minimal set of morally relevant states, it does not constitute something that would be good in the absence of an overriding or undermining state.

IV. CONCLUSION

The Principle of Universality is correct and consistent with the correct valuation of the Mean-Spirited Pain and Malicious Pleasure cases. The latter are explained by the nature of the basic intrinsic-value states, which involve a person having a level of well-being and desert at a time. Given the nature of such states, pleasure and pain are not value-bearing parts of virtuous and vicious attitudes. Thus, the Principle of Universality is not relevant in determining the value of individual attitudes.Footnote 24

References

1 Moore, G. E., Principia Ethica (Cambridge, 1903), 27–30Google Scholar; Ross, W. D., The Right and the Good (Oxford, 1930), pp. 136–8Google Scholar; Lemos, N., Intrinsic Value: Concept and Warrant (New York, 1994), pp. 40–7CrossRefGoogle Scholar; Bradley, B., ‘Is Intrinsic Value Conditional?’, Philosophical Studies 107 (2002), pp. 23–44CrossRefGoogle Scholar.

2 This taxonomy has recently come under fire from theorists who think that things that are valuable for their own sake can rest on extrinsic properties. See, e.g., Kagan, S., ‘Rethinking Intrinsic Value’, The Journal of Ethics 2 (1998), pp. 277–97CrossRefGoogle Scholar; Rabinowicz, W. and Ronnow-Rasmussen, T., ‘A Distinction in Value: Intrinsic and For Its Own Sake’, Proceedings of the Aristotelian Society 100 (1999), pp. 33–49CrossRefGoogle Scholar; and Korsgaard, C., Creating the Kingdom of Ends (Cambridge, 1996)CrossRefGoogle Scholar. I find the examples used in support of this claim to be unconvincing and so will assume this criticism fails. In addition, the notion that the intrinsic value of a part supervenes on an extrinsic property fits best with the notion that the intrinsic value in question is prima facie good, where prima facie is given an epistemic interpretation. For the notion that it is a prima facie good and that prima facie goods are tendencies rather than an actual species of rightness or value, see Ross, The Right and the Good, pp. 28–9, 136–8.

3 Feldman, F., ‘Basic Intrinsic Value’, Philosophical Studies 99 (2000), pp. 319–46CrossRefGoogle Scholar; Pleasure and the Good Life: Concerning the Nature, Varieties, and Plausibility of Hedonism (Oxford, 2004), pp. 172–82.

4 Hurka, T., ‘Two Kinds of Organic Unity’, The Journal of Ethics 2 (1998), pp. 299–320CrossRefGoogle Scholar; Virtue, Vice, and Value (New York, 2001), ch. 5.

5 These intuitions might break down where the pain doesn't exist. Intuitively, the positive value of compassion at the thought of someone's pain, where the pain doesn't exist, seems intrinsically good. I owe this point to Thad Metz.

6 Stephen Kershnar argues that Hurka's theory doesn't permit such a move (S. Kershnar, ‘Hurka's Theory of Virtue’, Philosophia, forthcoming). He argues that it breaks away from the recursive structure of Hurka's theory in that it thus allows virtue or vice to rest on objects that have no value. Since the recursive theory is what unifies his account of virtue and vice, Kershnar claims that this is a real problem for his theory. For the purposes of this article, I shall ignore this criticism.

7 Kershnar, ‘Hurka's Theory of Virtue’.

8 Hurka, ‘How Great a Good is Virtue?’, p. 190.

9 Contra Hurka, it is not clear that there are no cases when an attitude doesn't have a greater absolute intrinsic value than its object. One such case might occur when there is an incorrect attitude with great intensity that is directed toward an object of minor value.

10 Other plausible functions might involve averaging the values or treating them as contributing in a diminishing manner to the total value of the world. The former has a problem in explaining whether the averaging should take place over persons, times, or something else. For the different ways of averaging, see T. Hurka, ‘Average Utilitarianisms’, Analysis 42 (1982), pp. 65–119. For an example of the latter approach, see T. Hurka, ‘Value and Population Size’, Ethics 93 (1983), pp. 496–507. It is hard to see how the latter approach is plausible given that an earlier person makes the world a better place than a molecule-for-molecule identical person to whom the diminishing effect applies.

11 Feldman, F., ‘Adjusting Utility for Justice: A Consequentialist Reply to the Objection from Justice’, Philosophy and Phenomenological Research 55 (1995), pp. 567–85CrossRefGoogle Scholar; Hurka, T., ‘The Common Structure of Virtue and Desert’, Ethics 112 (2001), pp. 6–31CrossRefGoogle Scholar; Sher, G., Desert (Princeton, 1987), ch. 5Google Scholar; Nozick, R., Philosophical Investigations (Cambridge, Mass.: 1981), pp. 366–80Google Scholar; Feit, N. and Kershnar, S., ‘Explaining the Geometry of Desert’, Public Affairs Quarterly 18 (2004), pp. 286–8Google Scholar.

12 A thing has a derivative property only if it has it because of its relation to another thing (L. R. Baker, Persons and Bodies: A Constitution View (Cambridge, 2000), ch. 2). The idea here is that a person's overall character is constituted by his mental states and the relations between them.

13 Hurka, T., ‘Virtuous Act, Virtuous Dispositions’, Analysis 66 (2006), pp. 69–76CrossRefGoogle Scholar.

14 Here I assume that if attitudes have propositions as their objects, then propositions are a type of state of affairs.

15 The idea that one state of affairs can be part of another must be true if the Principle of Organic Unities (the intrinsic value of a whole is not necessarily equal to the sum of the intrinsic value of its parts) is true and if states of affairs are the bearers of value (N. Lemos,’Organic Unities’, The Journal of Ethics 2 (1998), pp. 321–37, at p. 322). The notion of inclusion can be found in the work of Alvin Plantinga (see Plantinga, A., The Nature of Necessity (Oxford, 1974), pp. 44–5Google Scholar).

16 I mean to avoid the internalist/externalist debate over thought content and will assume that Smith knows that Jones and Johnson are different persons and correctly refers to them.

17 W. D. Ross adopted this view and Erik Carlson and Michael Zimmerman have mentioned it. (Ross, The Right and the Good, pp. 136–7; Carlson, E., ‘Organic Unities, Non-Trade-Off, and the Additivity of Intrinsic Value’, The Journal of Ethics 5 (2001), p. 341CrossRefGoogle Scholar; Zimmerman, M., ‘Virtual Intrinsic Value and the Principle of Organic Unities’, Philosophy and Phenomenological Research 59 (1999), p. 662.CrossRefGoogle Scholar) Ross adopts it only if he is viewed as asserting that prima facie goods are a species of actual good. In the context of rightness, he is not willing to make this claim (Ross, The Right and the Good, ch. 2). It is not clear if Carlson or Zimmerman adopts it. Also, I am assuming here that desert is non-comparative. That is, desert depends on properties of individuals rather than relational properties between individuals.

18 Thomas Hurka also counts pleasure and knowledge as intrinsically good and pain and false belief as intrinsically bad and as independent of the value of the larger virtuous or vicious (and thus deserved and undeserved) facts about a person (‘Two Kinds of Organic Unity’, esp. pp. 317–20). Ben Bradley determines the intrinsic value of the world by adding up the basic intrinsic values of the states of affairs that are true there and counts aesthetic contemplation as a good independent of desert (Bradley, B., ‘Is Intrinsic Value Conditional?’, Philosophical Studies 107 (2002), pp. 23–44, at pp. 34–8CrossRefGoogle Scholar). Note that Bradley is doing this for illustration purposes and does not seem to commit to a particular view as to the nature of basic intrinsic-value states. Hurka, Noah Lemos, and Roderick Chisholm treat virtue and vice as having intrinsic value independent of the level of well-being of the virtuous or vicious person (Hurka, T., ‘How Great a Good is Virtue?’, The Journal of Philosophy 95 (1998), pp. 181–203CrossRefGoogle Scholar; Hurka, ‘Two Kinds of Organic Unity’, pp. 299–329; Hurka, Virtue, Value, and Vice, chs. 1–5; Lemos, Intrinsic Value, pp. 73–7; Chisholm, R., Brentano and Intrinsic Value (Cambridge, 1986), pp. 63–6Google Scholar).

19 Kagan, S., ‘Equality and Desert’, What Do We Deserve?, ed. Pojman, L. and McLeod, O. (New York, 1999), pp. 298–314Google Scholar; Hurka, T., ‘The Common Structure of Virtue and Desert’, Ethics 112 (2001), pp. 6–31CrossRefGoogle Scholar. Note my assertion that Hurka treats desert-fit as an independent ground of value rests on the notion that Hurka's graph explicitly labels the Y-Axis as desert-value (Hurka, ‘The Common Structure of Virtue and Desert’. pp. 18–19). In Kagan's work, his graph contains a range of features (e.g. bell motion, curved desert, and desert curves with equal peaks) that make sense only if there is a separate category of desert-value. For an extended argument in support of this, see Feldman, ‘Return to Twin Peaks: On the Intrinsic Moral Significance of Equality’, Desert and Justice, ed. S. Olsaretti (Oxford, 2003), pp. 145–68, esp. 162–7.

20 Lemos, ‘Organic Unities’, pp. 333–5.

21 Hurka, ‘The Common Structure of Virtue and Desert’.

22 Feldman, ‘Adjusting Utility for Justice’; Pleasure and the Good Life: Concerning the Nature, Varieties, and Plausibility of Hedonism (Oxford, 2004).

23 Ross, The Right and the Good, p. 138; Zimmerman, ‘Virtual Intrinsic Value and the Principe of Organic Unities’. Zimmerman's defense of virtual intrinsic goods is tied to an analysis of intrinsic value in terms of ethically required attitudes. I think this fails since I don't think the two co-vary. In particular, I don't think it wrong to take pleasure in the thought of Jones's suffering if doing so doesn't wrong anyone. Even if the two co-vary, the explanatory relation would go in the opposite direction.

24 I am grateful for the extremely helpful comments and criticisms of Neil Feit, Thad Metz, and George Schedler.