Consequentialists, be they maximizers or satisficers, can at least agree that the proper response to value is to promote it, however much they disagree about the proper degree of promotion. Kantians, on the other hand, believe that some values, at least, are to be honored rather than promoted. Roughly, one honors a value V when one does not act un-V-ly. A pacifist regards peace as a value to be honored and thus does not act violently (unpeacefully), while a just war theorist thinks peace is a value to be promoted and thus (subject to certain constraints) is willing to wage war if doing so will promote peace. (The realist, alas, does not regard peace as a genuine value at all, taking national interest to be the only value relevant to international relations.) Kantians, then, do not respond to the value personhood by promoting it – by trying to produce more persons – but by honoring or respecting it: by regarding it as constraining permissible ways of acting.
Christine Swanton and others have argued that promoting and honoring are not the only possible and plausible modes of response to value. Certain values, especially aesthetic ones, are to be appreciated; others, such as some values constitutive of friendship, are to be expressed. Indeed, Swanton has proposed a value-centered virtue ethic according to which ‘[a] virtue is a trait . . . whose possession tends to enable, facilitate, make natural, the possessor's promoting, expressing, honoring, and appreciating value’.Footnote 1 The viability of her virtue ethic depends in no small way on her being right about the plurality of responses to value.
Philip Pettit has argued against such value-response pluralism. He holds that expressing and appreciating are but species of honoring, and that honoring is reducible to a kind of idealized promotion: ‘[h]onoring a value . . . is acting in a way that would promote it in, roughly, a compliant world’.Footnote 2 Moreover, appreciation is not ‘a relevant alternative to promoting a value and . . . honoring that value’,Footnote 3 because it fails to satisfy a non-iteration constraint that any proper response to value must satisfy. Here is the heart of his argument:
Suppose that I countenance a value, V, and that I believe that the proper response to such a value – the response that marks out a corresponding action as right – is R-ing V. R-ing V is right, so I maintain. But if R-ing V is right, then R-ing V is a value. And so, if I am not to be launched on a regress of values, R-ing V had better not be a value that is itself distinct from the value, V. In particular, it must not be distinct in such a way that the project of R-ing R-ing V, if we can put it that way, is distinct from the original project of R-ing V. R must meet a certain non-iteration constraint. If R does not meet such a constraint, then a regress looms: V will be one value, R-ing V a second, distinct value, R-ing R-ing V a third, distinct value, and so on.
While it can certainly be described as a response to value in the ordinary usage of that phrase, appreciation does not meet this constraint and does not count as a response in the sense required for a theory of rightness. Appreciating a value is one thing. But appreciating appreciating a value is another. Appreciating a value is itself a value, in particular a value that is distinct from the value appreciated. And as a distinct value it can call itself to be appreciated. We may appreciate redwoods or we may appreciate the appreciation of those redwoods, whether by ourselves or by others.
There is a salient contrast between appreciation on the one hand and honoring and promotion on the other. For to promote the promotion of a value – or to promote the promotion of the promotion of a value – is not distinct from promoting the value itself. And similarly with honoring.Footnote 4
Pettit's argument from the non-iteration constraint against appreciation as a proper response to value, then, is quite simple:
P1 A proper response to value must satisfy the non-iteration constraint.
P2 Appreciation does not satisfy the non-iteration constraint.
∴ Appreciation is not a proper response to value.
Avoiding an infinite regress (or progress) is Pettit's reason for thinking P1 is true. An additional reason for thinking it true might find its inspiration in Donald Davidson's suggestion that intentional terms such as ‘deliberate’ and ‘incontinent’ be construed as sentence operators.Footnote 5 Following this suggestion, let us construe obligations as agent-indexed sentence operators and the various modes of value-response as value-indexed sentence operators. Thus a sentence such as ‘Andrea has an obligation to contribute to famine relief’ would be rendered
(1) □a Andrea contributes to famine relief,
where ‘□’ is interpreted deontically and the subscript ‘a’ refers to the agent, Andrea. Treating the obligation in this way allows us to avoid quantifying over it: we do not have to say that her obligation is a thing to which she is related by the relation of having, the same relation that holds between Andrea and her pet cocker spaniel. Sentences about value-promotion would be construed similarly, with ‘[Pv]’ as the value-indexed promotion operator and ‘[Hv]’ as the value-indexed honoring operator, etc. Thus ‘In contributing to famine relief, Andrea promoted human well-being’ could be rendered
(2) [Pw] Andrea contributed to famine relief
where ‘w’ refers to the value, human well-being. Note that in contributing to famine relief Andrea might also be honoring the promise she made to do so, which could be rendered
(3) [Hp] Andrea contributed to famine relief.
She might also be expressing the value empathy:
(4) [Ee] Andrea contributed to famine relief,
where ‘[Ev]’ is the value-indexed expression operator, and ‘e’ refers to the value, empathy. On this not implausible line of thinking, the non-iteration constraint embodies the requirement that a suitably developed logic of value-response be governed by the S4 axiom,
(S4) □□ p → □ p.
That is, operator iterations must in principle be eliminable: promoting the promotion of a value V must be reducible to the promotion of V, etc. Since iterated appreciation operators are not eliminable, the argument goes, appreciation is not a proper response to value.
I think Pettit is correct that – in many cases, anyway – appreciation violates the non-iteration constraint. Here is one such case. Suppose that Edmund Wilson appreciates the beauty of Sophocles' Philoctetes and expresses this appreciation in an essay called ‘The Wound and the Bow’.Footnote 6 And now suppose that in addition to appreciating the beauty of the Philoctetes, I also appreciate Wilson's appreciation of the Philoctetes: I admire the way he has read and appreciated the play – his insight, his subtlety of expression, etc. It might be argued that in appreciating Wilson's appreciation of the Philoctetes I am really just appreciating the Philoctetes (such often is the reason for reading literary criticism), so in this instance appreciation would obey the non-iteration constraint. I do not think this I correct, but let us grant it for the sake of argument. Let us suppose, then, that a friend who has not read the Philoctetes beholds me reading Wilson. She surmises from my expression that I am appreciating what I am reading. She values my appreciating worthy objects, a value she responds to by appreciating my appreciation of Wilson's appreciation. But since she has not read the Philoctetes she can hardly be said to be really appreciating the Philoctetes in appreciating my appreciation of Wilson's appreciation of the Philoctetes.
Here is a clear case of appreciation violating the non-iteration constraint. But why exactly is this supposed to be a problem? Absent some compelling case for the non-iteration constraint, it seems wrong-headed to deny that appreciation is a proper response to value because its logic resists assimilation to the S4 principle. The ineliminability of the appreciation-operator seems a strength rather than a weakness, for it enables us to distinguish between appreciations that have as their objects values such as beauty and meta-appreciations that have as their objects appreciations.
Perhaps a compelling case can be made for the non-iteration constraint. But even so, this would not be good news for Pettit, for by his own lights he should reject the non-iteration constraint, since promotion violates it, as I will now show. If promotion violates the non-iteration constraint then Pettit faces a dilemma: he must either deny that promotion is a proper, ‘basic response’ to value, or he must give up the non-iteration constraint. His consequentialism rules out the former, but if he gives up the non-iteration constraint then his argument against appreciation as a proper mode of value-response fails.
Pettit claims that ‘to promote the promotion of a value – or to promote the promotion of the promotion of a value – is not distinct from promoting the value itself’.Footnote 7 To see that this is false, consider the value of wealth – in particular, shareholder wealth. Dinesh, the CEO of a publicly traded company, believes there is one appropriate response to the value of shareholder wealth: promotion. Indeed, Dinesh is a maximizer: within the bounds of the law, he will do whatever will maximize the price of the stock and thus shareholder wealth. So even though keeping open the factory in Rustbelt will promote wealth to a satisfactory degree – better than any industry competitor – Dinesh decides to relocate the factory in a country where the daily wage is significantly less than the hourly wage in Rustbelt. Consider now Anne, a philosopher who taught Dinesh business ethics. A zealous advocate of the views of Milton Friedman (some might even call her a Friedmaniac), she seeks to impress on her students that as managers their only obligation is to promote shareholder wealth. Anne is not herself a manager, nor is she on any board of directors. Her project is not to promote wealth, but to promote the promotion of wealth.
Pettit holds that Anne's project, promoting the promotion of wealth, is not distinct from Dinesh's, promoting wealth. There are two good reasons for thinking the projects are distinct. First, they aim at different ends, and a plausible criterion for individuating goal-directed entities such as projects is the goal aimed at. Dinesh aims at increasing his company's stock price; Anne aims at convincing her students of her Friedmanite view. Second, the projects have different success conditions. Anne's project succeeds if her students come away committed to the promotion of shareholder wealth – if they come away confirmed Friedmanites. Dinesh's project, on the other hand, succeeds if the wealth of his shareholders has been maximized. Recession, natural disaster or just plain bad management may prevent Dinesh from promoting wealth, but the failure of his project does not detract from the success of Anne's, which succeeds if Dinesh and his classmates embrace Anne's Friedmanite views. Since the projects have different ends and different success conditions, the projects are distinct.
Imagine, further, that Ian comes into a substantial inheritance and decides to endow a chair in business ethics at the school from which he was expelled in his misspent youth. Anne is appointed to the chair, which enables her to promote her Friedmanite views about the promotion of wealth to an even wider audience. So in endowing the chair, Ian is promoting Anne's project – he is promoting the promotion of the promotion of shareholder wealth. But let us suppose further that Ian roundly rejects Anne's Friedmanite views, though he believes that the best way to undermine them is to have them promulgated by Anne, whom he regards as pedagogically ineffective. He hopes that her students will be turned off by her promotion of the promotion of wealth, and embrace the view he secretly holds, that the interests of other stakeholders, such as the employees and the community, must be accorded significant weight. So in promoting the promotion of the promotion of wealth, Ian in fact hopes ultimately that shareholder wealth will not be promoted. His project succeeds if Anne's and Dinesh's fail. It seems plain that Ian's project (promoting the promotion of the promotion of a value) is distinct from Dinesh's project (promoting a value) and from Anne's (promoting the promotion of a value). The projects aim at different objects and have different success conditions.
Nor should we think that the value in question, shareholder wealth, is somehow anomalous. Consider a value nearer and dearer to a consequentialist's heart – happiness. Though he clearly wishes to promote happiness, Mill's project in Utilitarianism is better described as promoting the promotion of happiness: he tries to convince his readers to accept the promotion of happiness as their basic normative principle. As before, Mill's readers might come away convinced utilitarians, but, through weakness of will, poor probabilistic reasoning skills, etc., they might fail to promote happiness. So Mill's project, the promotion of the promotion of happiness, can succeed, even though his readers' projects, the promotion of happiness, fail. And imagine a Nietzschean who promotes the promotion of the promotion of happiness by speaking enthusiastically of Mill's project to groups of counter-suggestible students in the hope that they will reject the promotion of happiness as their basic normative principle.
If the foregoing is correct and these various projects are distinct, then promotion itself fails to satisfy the non-iteration constraint. But then by P1 promotion is not a proper response to value, which is surely a conclusion that is anathema to consequentialists – but not only to consequentialists, since most non-consequentialists believe that promotion is a proper response to value. The remaining alternative is to reject the non-iteration constraint; but then of course Pettit's argument against appreciation as a genuine response to value fails.Footnote 8
