DARWALL'S ARGUMENT AGAINST CONSEQUENTIALISM

Darwall's argument begins with the claim that it belongs to the very idea of moral obligation that we can be held accountable for complying with our obligations: ‘It is conceptually impossible for one to be morally obligated to do something but not responsible for doing it, neither to the moral community, nor to God, not to anyone.’Footnote ⁵

It follows from this, according to Darwall, that moral obligations must be, or entail, second-personal reasons. If others can hold us accountable for complying with moral obligations, then they must have the authority to demand that we comply: ‘There can be no such thing as moral obligation and wrongdoing without the normative standing to demand and hold agents accountable for compliance.’Footnote ⁶ And authoritative demands provide us with reasons that are essentially second-personal – their validity depends on the presupposed authority and accountability relations.

To these points about accountability, authority and second-personal reasons, Darwall adds the claim that any outcome-regarding reason would be a reason of the wrong kind to support an authoritative demand. As he says:

According to Darwall, a reason of the right kind is ‘a fact about or feature of some object, appropriate consideration of which could provide someone's reason for a warranted attitude of that kind toward it.’Footnote ⁸ The basic idea is that different sorts of reasons justify different attitudes and actions: the fact that going to Mars would be thrilling might be a reason to desire going to Mars, but it is not a reason to believe that I will go to Mars. For Darwall, being the right kind of reason is explained by the appropriate role of reasons in our deliberations. A reason of the right kind ‘must be something on the basis of which someone could (and appropriately would) come to hold the attitude as a conclusion of a process of considering (and deliberating about) whether to do so’.Footnote ⁹

Now it might seem that, contra Darwall, outcome-regarding reasons are exactly the right kind of reason to support a demand that another do something. If I demand that you get off my foot, and you ask why, it seems perfectly reasonable to support my demand by saying, ‘Because it hurts!’ And that reply points to a state of the world to support my demand, and not simply to my authority. However, Darwall's argument can accommodate this point. The crucial claim is that outcome-regarding reasons alone cannot support such a demand. Even if your pain is a bad thing, and even if I have (decisive) reason to get off your foot because of the badness of your pain, that alone is insufficient to show that you have the authority to make any demand of me, or that I am accountable to you. After all, there are many ways that I might act irrationally – e.g. making errors in logical reasoning, or acting imprudently. And while I should act differently in those situations, and while you might recognize my failing, that alone does not give you the authority to demand that I act differently. Thus, as Darwall says, there is ‘an important difference between the idea of an authoritative or binding norm in the familiar sense of a valid ought that entails genuine normative reasons, on the one hand, and that of an authoritative (second-personal) claim or demand on the other’.Footnote ¹⁰ And no matter how desirable it might be that I do certain things, in terms of outcome-regarding reasons, that cannot establish that you (or anyone) have the authority to hold me accountable to do those things. For just as the category of the desirable is distinct from the category of the credible, so too the desirable is different from the responsible.Footnote ¹¹ And just as the desirability of going to Mars does not make it credible to believe that I will go to Mars, so too the fact that it would be desirable for me to do X – or even desirable for you to hold me accountable to do X – is a reason of the wrong kind for actually holding me accountable to do X.Footnote ¹² Darwall refers to this as ‘Strawson's Point’: ‘Desirability is a reason of the wrong kind to warrant the attitudes and actions in which holding someone responsible consists in their own terms.’Footnote ¹³ The right sort of reason for holding me accountable includes your authority to do so.

Darwall combines Strawson's Point with the conceptual connections between obligation, accountability and second-personal reasons, and he concludes that the ultimate grounds of our moral obligations cannot be (merely) outcome-regarding reasons. The grounds must include the authority of those who can hold us accountable. Since consequentialist accounts of obligation include only outcome-regarding reasons in the ultimate grounds of obligation, consequentialist accounts are inadequate.

It is easiest to see how Darwall's argument is intended to apply to act-consequentialism. Act-consequentialism explains an action's being permissible or obligatory in terms of the overall value that the action brings about. But the fact that an action will bring about the most overall value is an outcome-regarding reason, and hence on its own cannot show that I am accountable to perform that action. It is a reason of the wrong kind to do that. While outcome-regarding reasons might show that it would be desirable, or rational, for me to perform some action, they cannot establish that anyone has the authority to demand that I perform it. Thus the types of grounds available to act-consequentialism prevent it from capturing the conceptual tie between moral obligation and accountability.

More surprisingly, Darwall claims his argument also tells against indirect consequentialism (e.g. rule- or motive-consequentialism). Darwall recognizes that indirect consequentialism can register the distinction between reasons for desirability and reasons for accountability. Moral obligations might be understood as actions in accordance with second-personal practices that establish relations of accountability – i.e. practices that specify normative standing, enable the making of claims, holding accountable, blaming, etc. However, Darwall insists that the problem faced by act-consequentialism simply recurs at the level of the justification of the practices in indirect consequentialism. For the indirect consequentialist claims that the practices that define moral obligation are themselves justified entirely in terms of the good outcomes that they bring about. Such outcome-regarding considerations, however, are the wrong kind of reasons for holding people accountable, and hence the wrong grounds for moral obligation. As Darwall says:

Darwall's anti-consequentialism argument is part of his case for claiming that second-personal authority goes ‘all the way down’. That is, second-person authority cannot be grounded in more basic, outcome-regarding considerations. The second-person standpoint is normative bedrock. Having said that, however, Darwall allows that one might have a consequentialist theory of right that is grounded in second-personal claims to welfare or happiness.Footnote ¹⁵ On such a view, the principle(s) of right action might aim at maximizing average expected utility, but the justification of the principle(s) would rest on the second-personal authority of members of the moral community. In this way, second-personal authority still has the last word, rather than outcome-regarding considerations.

For my purposes, what matters is not this sort of limited consequentialism, but a consequentialism that goes all the way down. Such ultimate consequentialism is the target of Darwall's anti-consequentialist argument, and it is what interests me here.

A CONSEQUENTIALIST REPLY

One might reject Darwall's claim that reasons of desirability are always the wrong kind of reason for holding someone accountable. J. J. C. Smart, for example, considers a situation in which preventing a disaster – e.g. a riot by a dangerous mob – requires punishing an innocent man. And Smart judges that if the threat of harm was great enough, it would be correct to punish the man.Footnote ¹⁶ Thus perhaps there are times when it is correct to hold someone accountable, and even to punish him, simply on the basis of outcome-regarding reasons.

But this case does not actually undermine Darwall's right kind of reasons point. For blaming the innocent man can be understood in two different ways. One might claim that the outcome-regarding reasons are reasons for actually regarding the innocent man as blameworthy. That, however, is an implausible claim. For if I know that the man is innocent, it is hard to see how I could even bring myself to view him as accountable for something that he has not done, short of some massive act of self-deception on my part. Alternatively, one might claim that in our attitude towards the man, we do not regard him as truly responsible or as meriting punishment. Rather in punishing him to prevent a riot, we are in effect pretending to regard him as responsible because of the desirability of doing so. But interpreted this way, the example does not show that reasons of desirability are reasons of the right kind to blame someone – only reasons to act as if someone is blameworthy.

Indirect consequentialism, however, need not reject Darwall's claim about the right kind of reasons for holding others accountable. Indirect consequentialists can explain moral obligation in terms of conformity to accountability-structuring practices, and within such practices our reasons for holding people responsible need not be reasons of desirability. For example, within the practice of promise-keeping, the promisee has standing to demand performance, and the promisor has an authority-regarding reason to perform the action because he promised. Moreover, when we hold others accountable in accordance with second-personal practices, we are not merely acting as if they are accountable. Of course the standing to hold accountable, and the specific norms governing accountability, are established by the practice. But this does not mean that we are merely pretending to hold accountable, any more than that we are pretending when a runner is thrown out in a game of baseball. The runner could not be ‘out’, and the umpire would not have the authority to declare him so, without the existence of the practice of baseball. But given the practice and the facts of this case, this runner really is out. And the same thing can be said for accountability and second-personal authority, under the practices that establish obligations.

Indeed, the indirect consequentialist need not appeal only to particular practices such as promise-keeping, but can instead think of the second-person standpoint as a kind of uber-practice, or even a kind of ‘conscience’, that we adopt.Footnote ¹⁷ Embracing the second-person standpoint, however, is itself ultimately justified by the way that conforming to the standpoint promotes the greatest overall good, impartially understood.

Darwall, however, insists that a consequentialist's appeal to practices only pushes the wrong kind of reason problem back further. He seems to have something like the following in mind. When you hold me accountable to do X, I might protest: ‘What right do you have to demand compliance from me? You’re not the boss of me!’ Merely pointing out that I should do X is an insufficient answer, because the rationality or desirability of my X-ing does not show that you have authority to hold me accountable to act in the most rational or desirable way. My protest can only be answered adequately by a reply that includes your authority to hold me accountable. Now, if there is some authority-structuring practice within which you have authority, then you can appeal to your authority. If I have promised to feed your dog, then my reason to do so is not simply the desirability of your dog being fed. Rather, as the promisee, you have standing to demand that I comply. However, Darwall seems to think this answer allows a further protest, such as: ‘What right do you have to hold me accountable for complying with this authority-structuring practice? You’re (still) not the boss of me!’ And Darwall believes that unless there is an answer to this further question, the initial act of holding accountable will be unjustified. Moreover, the answer to this further question cannot appeal to merely outcome-regarding reasons, since those would be reasons of the wrong kind to ground the authority to hold others accountable for complying with any practices.

This is the heart of Darwall's argument against indirect consequentialism: Strawson's Point about the wrong kind of reason simply reappears at the level of the justification of our accountability-structuring practices. Now, indirect consequentialists need not deny that there is a further question about the justification of our practices. The key issue is what form the question must take, and what kinds of reasons are appropriate to answer it. Darwall supposes, it seems, that the justificatory question concerns our authority to hold others to the practices: by what right are we held accountable to comply with the practices, and by what right do we hold others accountable? And if we assume that the question takes that shape, then indeed consequentialist reasons about states of affairs appear to be reasons of the wrong sort. Even if a certain practice leads to the best outcomes, what authority do we (or anyone) have to hold others accountable to comply with practices that promote the best outcomes? The desirability of following the practice might justify holding others accountable if we already had some general authority to hold others accountable for acting in ways that promote the best outcomes. Without that authority, however, merely outcome-regarding considerations are reasons of the wrong kind to explain anyone's authority to hold another accountable to do anything, including complying with desirable practices. Hence Darwall says that the most rule-consequentialism could establish is the rationality of the desire for an accountability-structuring practice, or the desire to hold people accountable to such a practice, but not actually holding people to such a practice.Footnote ¹⁸

However, the indirect consequentialist should insist that the justificatory question is not by what right we can hold others accountable to rules and practice, but rather if we are right to do so – i.e. if we are rationally justified in so doing. And the answer to that question can be a consequentialist answer about the desirability of having certain practices. As Darwall recognizes, the desirability of an outcome is often a reason of the right kind for doing something. And there is no reason to think that it is a reason of the wrong kind to establish an accountability-structuring practice. If someone asks by what right we establish the practice, the consequentialist can answer that we do not have any prior right or authority to do so. But neither do we need any such right. For we need not pose the question of justification in terms of a right to do anything, only in terms of what it is right to do. And once posed that way, the appearance of a wrong kind of reasons problem disappears.

FURTHERING THE DEBATE

As Darwall conceives of moral reasoning, the justification of our moral practices – and not only justification within those practices – must be a justification to other persons. This means that it is more than recognizing reasons and sharing them with others. If you ask why I decided to vacation at one spot rather than another, then I might explain my reasons for doing so. But I am not thereby giving a justification to you, or justifying myself to you.Footnote ¹⁹ For I don't suppose that you have any authority to demand a justification from me: ‘To justify oneself to someone is to give her a kind of second-personal authority. It is to regard and treat her as having a standing to claim justification from one.’Footnote ²⁰

Because Darwall conceives of ultimate justification this way, he claims that indirect consequentialism offers the wrong sort of reason for the justification of our practices. On their own, outcome-regarding reasons can't capture the second-person authority that is implicit in justifying our practices to others. However, so far as I can tell, Darwall never offers an independent argument to show that justifying our practices must be justification to others, or that it must presuppose a right by which we establish authority-structuring practices. Thus, in claiming that indirect consequentialism merely postpones the difficulty from Strawson's Point, Darwall seems to be assuming what needs to be proved. His anti-consequentialism argument depends upon, rather than establishes, a picture of justification that the consequentialist rejects.

Is there any way to adjudicate this dispute about the justification of our practices, and hence to determine if Darwall's argument has real force against indirect consequentialism? On behalf of Darwall, we might argue that unless we presuppose second-personal authority at the level of the justification of a practice, then we lose authority to hold anyone accountable to keep the substantive obligations that the practice defines – e.g. to keep promises, not to steal or murder, etc.Footnote ²¹ Imagine someone who rejects all second-personal practices. If we lack authority to establish practices, then (the argument goes) we lack the standing to blame such a person. For blame, Darwall argues, is a form of holding others accountable, and hence blaming others implies that we have some authority to hold accountable.Footnote ²² In this respect, blame must be distinguished from simply criticizing someone, as when we point out a logical mistake.Footnote ²³ But surely (the argument continues) we should blame such a person who rejects the second-personal authority of others and violates his obligations. Thus we must take ourselves to have authority to establish accountability-structuring practices, and hence authority prior to those practices.

The problem with this argument is that it fails to reckon adequately with the distinction that is crucial to indirect consequentialism, between: (a) how we regard ourselves and others within a practice, and (b) reflection upon a practice and its justification. If someone rejects all second-personal standing, that does not undermine our authority to hold him accountable for substantive second-personal norms. For that authority is established by the second-personal practice, and it is not part of that practice to suppose that someone who rejects the practice is thereby exempt from the demands that the practice grounds. In holding accountable to second-personal obligations, we operate within an accountability-structuring practice, and that practice includes holding accountable even those who claim to reject second-personal norms. Of course, we might imagine an authority-structuring practice that exempted anyone who claimed to reject second-personal authority. But presumably such a practice would be much less beneficial than a practice that did not exempt such people, and hence it would be rejected on consequentialist grounds.

What, then, if someone protests, ‘You have no authority to hold me accountable to these second-personal norms’? If this is interpreted to mean authority to hold accountable to substantive norms – about promising, honesty, murder, etc. – then the protest is simply false. For our holding accountable is a move we make within the relevant practice, and within the practice we do have such authority. On the other hand, we might be reflecting upon the practice as a whole and asking about its ultimate justification. And at this level of reflection, the indirect consequentialist can give the reply outlined earlier: we need not think of ourselves as having any right, or authority, to establish the practice, simply as being right to do so. To insist on second-personal authority at this level begs the question against the indirect consequentialist.

A slightly different argument on Darwall's behalf is more promising. This argument also appeals to blame. But whereas the first argument appeals to blame for violations of substantive norms, and to the claim that authority within a practice requires authority to establish the practice, the second argument appeals to blame within the scene of ultimate justification itself. Suppose that we are inquiring about the ultimate justification of our moral principles and practices. And suppose that I insist that I need be concerned only with my own welfare, and not the welfare of anyone else. Moreover, I insist that I don't have to justify myself to anyone. Now, the indirect consequentialist can surely say that I am in error about what I have reason to care about and what I should do. But can the indirect consequentialist blame me for what I do within the scene of ultimate justification? It seems not, if we accept Darwall's (plausible) claim that blame is a second-personal concept that implies second-personal authority to hold accountable. For the indirect consequentialist holds that at the level of ultimate justification, second-personal authority does not obtain, and the most basic considerations are all outcome-regarding. Thus it appears that indirect consequentialists cannot blame anyone for what he does, or does not do, at the level of ultimate reflection. The consequentialist can maintain that my refusal to care about the welfare of others is irrational. But the indirect consequentialist cannot hold that I am failing to satisfy an authoritative second-personal demand, or that I am blameworthy.

In contrast, Darwall's own view allows us to say that I am to blame, not only for failing to care about the welfare of others, but also – and more fundamentally – for failing to justify myself to other members of the moral community. If second-personal authority goes all the way down, then as we begin the process of seeking justifications for our moral principles or practices, and before we put forward any particular answers to our questions about how to live, we already stand in a normative relation that requires us to give a justification to one another. Thus, at the most basic level, others have the authority to hold me accountable, and hence I can be blamed (and not only criticized) for what I do within the scene of ultimate justification. As Darwall says,

Thus we have the following argument against indirect consequentialism: if indirect consequentialism is true, then we cannot blame one another at the level of ultimate justification. But surely we can blame at this level. So indirect consequentialism is false (and second-personal authority must go all the way down).Footnote ²⁵ The first premise in this argument depends on ideas that Darwall develops in SPS, including the second-personal nature of blame and Strawson's Point. However, unlike the anti-consequentialism argument in SPS, this argument does not beg the question against consequentialism over the nature of justification.

How successful, then, is this argument? Indirect consequentialists can, of course, simply reject the second premise. This might seem like an unappealing option. But two points should be kept in mind. First, like the first argument in this section, this second argument does not show that if we lack second-personal authority to establish a practice, then we also lack authority within the practice. So even if the indirect consequentialist cannot account for blame within the scene of ultimate justification, this does not undermine our substantive practices of blaming and accountability. The conceptual tie between obligation and accountability is preserved within the practice, even if the level of ultimate reflection is not a scene that is itself governed by second-personal norms. Second, the indirect consequentialist can recognize a kind of moral failure within the scene of ultimate justification. We might be tempted to suppose that for Darwall our most basic moral reasoning is ‘moralized’ activity (= governed by moral norms), whereas for the consequentialist it is a non-moral or ‘purely theoretical’ exercise.Footnote ²⁶ But that is not quite right. For the consequentialist can say that failure at this level to recognize the importance of others’ welfare is a kind of moral failure. Even if second-personal authority does not go all the way down, the equal value of well-being or happiness might. And proper reasoning at the level of ultimate justification will recognize that value.

For both of these reasons, it is unclear how much cost there is in denying a place for blame within the scene of ultimate justification. I conclude, then, that while this extension of Darwall's original argument might force indirect consequentialists to accept a counterintuitive result, it is far from decisive as an argument against consequentialism.