According to the buck-passing account of value, as presented by T. M. Scanlon, ‘being good, or valuable, is not a property that itself provides a reason to respond to a thing in a certain way. Rather, to be good is to have other properties that constitute such reasons.’Footnote ¹ On this account, to say that something is good, or valuable, is to say that it possesses the ‘purely formal, higher-order property’ of having other, lower-order properties that provide reasons to respond to that thing in a certain positive way – perhaps by choosing, preferring, recommending, respecting, or admiring that thing. Consider, for example, a good resort, or a valuable scientific discovery. According to Scanlon's account, ‘the fact that a resort is pleasant is a reason to visit it, or to recommend it to a friend, and the fact that a discovery casts light on the causes of cancer is a reason to applaud it and to support further research of that kind’.Footnote ² On this account, the fact that the scientific discovery is valuable does not itself provide a reason to respond to it in certain ways; rather, the ‘buck is passed’ to those other, lower-order properties, such as the fact that the discovery casts light on the causes of cancer, that do provide such reasons.

Buck-passing is attractive to many because it presents a conceptual connection between concepts usually thought to be distinct: the axiological concept of good (or valuable) and the deontic concept of a reason. As some have noted, this conceptual connection allows for us to ‘demystify’ value by understanding value in terms of reasons.Footnote ³ For instance, Gerald Lang writes:

So, we can understand buck-passers as asserting the following biconditional, along with the claim that the concepts on the right-hand side reductively explain the concepts on the left-hand side: something is good (or valuable) if and only if it has properties (besides its being good) that provide a reason to respond to it in a certain positive way.

The Wrong Kind of Reason Problem (WKRP) is usually presented through examples which show that an object could have properties that provide a reason to respond to it in a certain positive way without its being good (or valuable).Footnote ⁵ Suppose an Evil Demon threatens to inflict a great deal of pain on us unless we admire him. The Evil Demon has a property (his having threatened us) which gives us a reason to respond in a certain positive way (by admiring him), but surely the Demon is not good or valuable.Footnote ⁶ The WKRP shows that buck-passing fails to state a sufficient condition for something's being good.

Intuitively, our reason here for admiring the Demon is a different kind of reason from our reason for admiring the scientific discovery. The challenge for defenders of buck-passing is to articulate clearly the distinction between these intuitively different kinds of reasons, and to formulate buck-passing to make it invulnerable to counterexamples in which the properties of an object provide ‘the wrong kind of reason’ for having a certain positive response towards it, as the Demon's threat does.

In his article ‘The Right Kind of Solution to the Wrong Kind of Reason Problem’, Gerald Lang presents a new formulation of buck-passing and argues – soundly, I think – that his formulation is not threatened by this Evil Demon example nor by any other similar counterexamples that have emerged in the philosophical literature on the WKRP. I'll here present (§I) Lang's promising version of buck-passing, and suggest a minor modification to improve it. I'll then argue (§II) that we should reject Lang's version of buck-passing (even with the minor modification) because it is not compatible with consequentialism. More precisely, I'll argue that if consequentialism (whether direct or indirect) is true, then Lang's formulation of buck-passing would rule out, by definition, certain plausible theories of the good.

I

After rejecting five potential versions of buck-passing, Lang settles on the following version:

This version of buck-passing requires that we employ a counterfactual test: we must ask, ‘If there were no benefits to us of having a certain positive attitude A towards X, would the properties of X still give us reason to have A towards X?’ If we answer in the affirmative, then, and only then, is X good.

The formulation is quite promising since it deals well with the examples we have considered, as well as with other examples that I won't have space to mention here. Consider the Evil Demon. If there were no benefits to us in admiring the Demon – if doing so would not help us avoid the great deal of pain – then the properties of the Demon would not give us reason to admire the Demon. And so (BPV6) does not entail that the Demon is good. Now consider the scientific discovery. If there were no benefits to us in admiring the scientific discovery, the properties of the discovery (specifically, that it sheds light on the causes of cancer) would still give us reason to admire the discovery. According to (BPV6), then, the scientific discovery is good.

One curious feature of (BPV6), however, is the use of the phrase ‘the benefits to us’ in the counterfactual clause, ‘. . . just as long as those properties of X that give us reason to have that attitude towards X would still be reason-giving in the absence of the benefits to us of having that attitude towards X’. Suppose the Evil Demon threatens that unless we admire him, he'll inflict a great deal of pain on some other people unknown to us. His threat gives us a (moral) reason to admire him, and this reason would persist in the absence of the benefits to us of admiring him (since we don't benefit, though others surely do, from our admiring him). And, so, (BPV6) here entails the absurd conclusion that this Evil Demon is good. In light of this worry, Lang could modify (BPV6), perhaps by simply replacing ‘benefits to us’ with ‘benefits’ in the counterfactual clause, giving us:

Since the Demon's threat here would not be reason-giving in the absence of the benefits of admiring him, it doesn't follow from (Modified BVP6) that the Demon is good.

II

So far we have considered a couple of good objects (a good resort and a valuable scientific discovery) and one bad object (an Evil Demon) that have properties that provide us with reasons to respond in various ways: the good resort has a property that provides us with a reason to visit it and recommend it, the valuable scientific discovery has a property that provides us with a reason to applaud it and support further research of that kind, and the Evil Demon has a property that provides us with a reason to admire it. In just these three examples, there is a fairly wide variety of responses that, it is claimed, we have reason to have.

These claims about how we should respond to the objects of value are, like the claims about which objects are valuable, ordinary ethical claims. As such, they may be informed by ethical theory. An ethical theory may tell us, first, what is valuable – it may provide a so-called “theory of the good” – and may tell us, second, how we should respond to what is valuable. Consequentialism is usually understood to be an answer to the second query, one which tells us that agents, whether individual or institutional, should respond by promoting the general realization of what is of value. Non-consequentialists, however, think that one should respond by honoring what is of value (perhaps by having the object of value figure prominently in one's own life, or in one's own actions) even when honoring what is of value fails to promote its general realization. Let's take an example borrowed from Philip Pettit. Suppose our theory of the good specifies that what matters above all in life is the enjoyment of personal loyalties. As Pettit notes, we are now left with a question of how we should respond to what is of value:

Consequentialists answer that the proper response to objects of value is to promote their general realization; one should honor values only insofar as doing so constitutes, or is necessary for, promoting them.Footnote ⁹ Here, one should permit oneself devotion to developing the bonds of kith and kin only insofar as doing so is part of the more general prospect of promoting the enjoyment of personal loyalties.

Consequentialists, however, are also sensitive to the fact that the most effective way to promote what is of value may not be the direct approach of calculating, for each individual option, whether that option better promotes what is of value than any other option. In other words, the attitude of aiming to promote what is of value may be an attitude that we have good reason not to have; we may actually do better at promoting what is of value by having other attitudes. For instance, we can easily imagine how someone who is an ‘incessant loyalty calculator’ – that is, someone who calculates, for each individual option, whether that option promotes the general realization of the enjoyment of personal loyalties better than any other option – would fail to inspire the loyalty of others. (After all, if you seriously consider such questions as ‘Would my breaking my loyalties to Anna allow me to develop loyalties with Bob and Carol, and thereby more effectively promote the general realization of the enjoyment of personal loyalties?’ and you are the least bit transparent, you would likely fail to inspire others to be loyal to you. And your political and journalistic efforts would then be unconvincing and ineffective.) It might be that you best promote loyalties by honoring the bonds of kith and kin, even when it appears to you that by breaking those bonds you could better promote the general realization of the enjoyment of personal loyalties.

In light of such concerns, most consequentialists are indirect consequentialists. They hold that although we ought to promote the general realization of value, it may be the case that we don't have reason to aim to promote the general realization of value. Rather, we may have reason to have those other attitudes the having of which more effectively promotes the general realization of value. As indirect consequentialists often put the point, consequentialism should be understood as presenting a criterion of right action, but not a ‘theory of deliberation’ or ‘decision procedure’. (Direct consequentialists, in contrast, insist on proceeding directly; they deny that we may have reason to have attitudes other than aiming to promote the general realization of value.)

Much more could be said here. We could go on to provide a precise account of the notions of promoting and honoring, and we could also note the variety of ways in which a consequentialist could be an indirect consequentialist. But these details, which are available elsewhere,Footnote ¹⁰ do not matter for the argument that follows. So, we can safely work with the rough sketch of consequentialism presented here. In the remainder of this section, I'll argue that Lang's version of buck-passing is not compatible with consequentialism, whether direct or indirect.

According to (Modified BPV6), X is good only if X has properties that give us reason to have a certain attitude towards X, and those properties of X that give us reason to have that attitude towards X would still be reason-giving in the absence of the benefits of having that attitude towards X. But here's the problem: for the indirect consequentialist, we have reason to have certain attitudes towards objects of value precisely because of the benefits of having those attitudes – specifically, precisely because our having those attitudes best promotes the general realization of value. We could illustrate this objection with Pettit's example of personal loyalties. According to one possible theory of the good, what matters above all in life is the enjoyment of personal loyalties.Footnote ¹¹ (Nothing really hinges on this particular claim; one could also make the same argument with liberty, happiness, or some other value or combination of values in place of personal loyalties.) An indirect consequentialist might say that I have a reason to honor loyalty, but I have a reason to have this attitude only because having this attitude provides certain benefits – specifically, having it best promotes the enjoyment of personal loyalties.Footnote ¹² If there were no such benefits, I would not have any reason to have the attitude. And since my reason to have this attitude would not survive in the absence of the benefits of having it, it would follow, implausibly, from (Modified BPV6) that enjoyment of personal loyalties is not good.Footnote ¹³

It's worth noting that (Modified BPV6) would also yield implausible results for direct consequentialists. According to direct consequentialism, we have reason to aim to promote value. One fairly obvious benefit of aiming to promote value is that doing so greatly increases the probability that one will in fact promote value. But let's run the counterfactual test and ask, ‘If there were no benefits to my aiming to promote value – that is, if my having this aim did not increase the probability that I will in fact promote value, or yield any other benefits – then would I have any reason to aim to promote value?’ It seems clear that we should answer ‘No.’ (Imagine a case in which you are certain that some aim of yours would be thwarted and there are no other benefits to your having that aim. What reason would there be to have that aim?) And so the counterfactual test in (Modified BPV6) would also yield implausible results for the direct consequentialist.

Perhaps we could provide an explanation for why Lang falls into error here. Lang is concerned to formulate buck-passing so as to make it immune to the Evil Demon counterexample. In the Evil Demon counterexample, the attitude that we have reason to have (because of the Demon's threat) is admiration – an attitude that, typically, we don't have reason to have because of the benefits expected from having it. Our reasons to admire something (like a scientific discovery) seem to persist in the absence of the benefits to be had from admiring it. But this is not so with other attitudes, including those attitudes central to consequentialism. Our reasons to aim to promote something of value do not persist in the absence of the benefits to be had from so aiming. And, for the indirect consequentialist, our reasons to have certain other attitudes (such as honoring personal loyalty) do not persist in the absence of the benefits to be had from having those attitudes. Perhaps Lang's mistake is that he focuses on reasons to admire at the expense of other attitudes we may, according to some ethical theory, have reason to have.

It is important to remember that buck-passing is intended as a formal or abstract account of value, one which should not rule out substantive ethical theories by definition. There may indeed be good reasons to reject consequentialism (or, at least, Pettit's particular conception of it), but such reasons are surely not to be provided by a reductive explanation of the concept good. Consequentialism may be false, but it is not false by definition. It's also worth noting that Scanlon, in his original presentation of the buck-passing view, noted the compatibility of buck-passing and consequentialism:

Scanlon presents a number of arguments against consequentialism, but he thinks a virtue of his account of buck-passing is its compatibility with consequentialism. But it seems that Lang's formulation of buck-passing cannot claim this same virtue.Footnote ¹⁵

One final note: though I think, for the reasons advanced in §I above, that (Modified BPV6) is a better formulation of buck-passing than (BPV6), we should note that the objection presented here would also work against (BPV6). Suppose our theory of the good specifies something as good which is also beneficial to us. For example, suppose we think that developing one's own talents is good. According to direct consequentialism, we have a reason to aim to promote the development of our own talents. But, again, were there no benefits to us in having this aim, then we wouldn't have a reason to have this aim. And so it would follow, implausibly, from (BPV6) that it's not the case that developing one's own talents is good. (And, as before, an indirect consequentialist might hold that we have reason to have some other attitude towards the development of our own talents, and that we have reason to have that attitude only because of the benefits to us of having that attitude – namely, having that attitude best promotes the development of our own talents. It would again follow, implausibly, from (BPV6) that it's not the case that developing one's own talents is good.)

III

The original WKRP objection to Scanlon's account of buck-passing challenges whether Scanlon's account really states a sufficient condition for something's being good. The Evil Demon is clearly not good, but yet he has properties that provide us with a reason to admire him. But the above objection to Lang's account of buck-passing challenges whether Lang really states a necessary condition for something's being good. We begin by asserting, by hypothesis, a certain theory of the good. We then note that the condition stated on the right-hand-side of the biconditional in (Modified BPV6) and in (BPV6) may not be met if consequentialism, whether direct or indirect, is true.

In light of this objection, we should reject Lang's formulation of the buck-passing account of value. His formulation does indeed succeed insofar as it excludes what ought to be excluded: his account does not allow for the reasons provided by the Evil Demon's threat to make it the case that the Demon is good. But his account fails in that it excludes too much: his account also, when combined with (direct or indirect) consequentialism, renders certain plausible theories of the good false.Footnote ¹⁶