REASON, HUMANITY, AND THE MORAL LAW
__________________________________________________
1. You might think that, if you make a law, then that law binds you, because you made it. For, if you will the law, then how can you deny that it binds you, without contradicting your own will? But you might also think the opposite. You might think that, if you are the author of the law, then it cannot bind you. For how can it have authority over you when you have authority over it? How can it bind you when you, the lawmaker, can change it, at will, whenever you like?
Now, in that pair of arguments mutually contradictory conclusions are drawn from the self-same premise, the premise that you make the law. So at least one of the arguments is invalid. But, even if they are both invalid, they might still be unignorable, because they might have elements of truth in them. They might be healthy argument-embryos out of which sound arguments could develop. And, whatever else is true, each of the arguments is sufficiently persuasive that, mutually contradictory though they are, each was affirmed in Hobbes’s Leviathan, in much the form in which I just stated them.
There are, by my count, four arguments in Hobbes for the conclusion that the citizen is obliged to obey the law. Three of the arguments don’t matter here.1 The one that matters here has two premises, the first of which is that every act of the sovereign is an act of each citizen, since “every man gives their common representer authority from himself in particular; and owning all the actions the representer doth, in case they give him authority without stint,” and there is, of course, no stint in the authority Hobbesian people give their sovereign. Accordingly, “the subject is the author of every act the Sovereign doth.”2
Having thus possessed himself of the premise that I am the author of what the sovereign does, hence of each law he enacts, Hobbes now enters his second premise, which is that it is absurd for me to object to what I myself do. Accordingly, it is absurd for me to object to any law that I pass, and I must therefore, on pain of absurdity, obey the sovereign’s law.3 I must obey it because I made it.
Now you might think that, if I am subject to the law because I make it, not, albeit, directly, but through my representative, then that representative himself, the sovereign, is equally or even a fortiori subject to the law, because he makes it, and, indeed, makes it more directly than I do. But that is not Hobbes’s inference. Not only does Hobbes not infer, using the same reasoning that he used in the case of the citizen, which should, it seems, also apply to the sovereign, that the latter is subject to the law he makes; but Hobbes concludes, oppositely, that the sovereign is not subject to the law. And the reason that Hobbes gives for that conclusion is the very same one as the reason that he gives for concluding that the citizen is subject to the law, to wit that he, the sovereign, makes the law. Here is what Hobbes says:
[T]o those laws which the Soveraign himselfe, that is, which the Common-wealth maketh, he is not subject. For to be subject to Lawes, is to be subject to the Common-wealth, that is to the Sovereign Representative, that is to himself; which is not subjection, but freedom from the Lawes.
The Sovereaign of a Common-wealth, be it an assembly, or one man, is not subject to the civil laws. For having power to make, and repeal laws, he may when he pleaseth, free himself from that subjection, by repealing those laws that trouble him, and making of new; and consequently he was free before. For he is free, that can be free when he will: Nor is it possible for any person to be bound to himself; because he that can bind, can release; and therefore he that is bound to himself only, is not bound.4
This argument says (in the fuller version of it to be found in the second quoted passage, and very slightly reconstructed):
The sovereign makes the law. | |
So |
The sovereign can unmake the law. |
So |
The sovereign is not bound by the law that he makes. |
Hobbes claims that, when the citizen violates the law, he contradicts his own will: he, in the person of the sovereign, made the law, and therefore cannot without absurdity violate it. Yet it is precisely because the sovereign makes the law that he is not bound by it: according to Hobbes, it is conceptually impossible for him to violate it.
There is no inconsistency in the idea that two make the one law, for one makes as author what the other makes as representative of that author. But it cannot follow from “X makes the law” both that X is subject to it and that X is not subject to it. You cannot say both: because you make the law, you must obey it; and: because you make the law, it has no authority over you, so you need not obey it. The inference of the argument about the citizen requires the principle that, if I make the law, then I am bound by it. The inference of the argument about the sovereign requires the principle that, if I make the law, then I am not bound by it. At least one of those principles must be wrong.
Now, the truth of this whole matter is complicated, but the parts of it that concern us here seem to me to be this. I pass a law. Either the law says that everyone must act thus and so, or its scope is restricted to, say, everyone except me. If the latter is true, then I am clearly not obliged to obey the law: so the first point to make about the first argument is that the terms of the law need to be specified before the inference in that argument can be examined.
Suppose, then, that the law is indeed universal, or that it includes me within its scope by virtue of some other semantic or pragmatic feature of it. Then, if I had the authority to legislate it, it indeed binds me, as long as I do not repeal it. (It remains unclear, even then, that it binds me because, if I violate it, I contradict my will: so the kernel of truth in the first argument may be quite a small one.) The necessity to add that rider reflects the important element of truth in the argument about the sovereign, which is also incorrect in its unmodified form. The big mistake in that argument is the supposition that if I can repeal the law, then it fails to bind me even when I have not yet repealed it. Hobbes is wrong that, if you can free yourself at will, then you are already free, that “he is free, that can be free when he will.” But other important things do follow from my being able to free myself at will, for example, that I cannot complain about my unfreedom. And, more pertinently to our theme, although you may be bound by a law that you can change, the fact that you can change it diminishes the significance of the fact that you are bound by it. There’s not much “must” in a “must” that you can readily get rid of.
2. In Christine Korsgaard’s ethics, the subject of the law is also its author: and that is the ground of the subject’s obligation—that it is the author of the law that obliges it. That sounds like Hobbes’s first argument. So we should ask a question inspired by Hobbes’s second argument, the one about the sovereign: how can the subject be responsible to a law that it makes and can therefore unmake? As we know, Korsgaard’s answer relates to the circumstance that the subject has a practical identity.
Now Korsgaard’s ethics descends from Kant, but it contrasts in important ways with Kant’s ethics. Korsgaard’s subject is unequivocally the author of the law that binds it, for its law is the law of its practical identity, and the subject itself “constructs” that identity. But in Kant the position is more equivocal. We can say that the Kantian subject both is and is not the author of the law that binds it. There is an important duality with respect to the source of the law in the following characteristic text from the Grundlegung. In previous moral philosophy, Kant says,
Man was seen to be bound to laws by his duty, but it was not seen that he is subject to his own, but still universal, legislation, and that he is bound to act only in accordance with his own will, which is, however, designed by nature to be a will giving universal law. For if one thought of him as only subject to a law … this necessarily implied some interest as a stimulus or compulsion to obedience because the law did not arise from his own will. Rather his will had to be constrained by something else to act in a certain way. This might be his own interest or that of another, but in either case the imperative always had to be conditional, and could not at all serve as a moral command. The moral principle I will call the principle of autonomy of the will in contrast to all other principles which I accordingly count under heteronomy.5
Kant thought that if the moral law came just from my own will, then it would have no claim on me, rather as the law of the sovereign has none over the sovereign in Hobbes. If, on the other hand, the law was just externally imposed, and did not come from my own will, then it would be heteronomous slavery for me to obey it, and the challenging argument that Hobbes uses about the citizen, that he must obey the law because it is his own law, he must obey it on pain of inconsistency, would not be available. So the passage I’ve exhibited, while stating that man is subject to his own law, and bound to act only in accordance with his own will, is quick to add that man’s legislation, though his own, is still universal, emanating from a will “designed by nature to be a will giving universal law.” And that makes Kant’s person different from Hobbes’s sovereign. Kant’s person indeed makes the law, but he cannot unmake it, for he is designed by nature to make it as he does, and what he is designed to make has the inherent authority of reason as such. So Hobbes’s sovereignty argument does not apply, and Kant can stay with the citizen argument. He can give the citizen argument for obedience and rebut the sovereignty argument for freedom from law by pointing out that this law is not one that the agent can unmake. In the moral realm, we are, Kant says, “subjects … not sovereigns.”6
So the Hobbesian reflection about the sovereign and the law sheds light on Kant’s insistence that the imperative of morals must not come from human nature, nor even from human reason, should there be any respect in which human reason differs from reason as such. For it is reason as such that is sovereign over us, and that gives determinacy, stability, and authority to a law that would otherwise lack all that: “the ground of obligation must not be sought in the nature of man … but a priori solely in the concepts of pure reason.”7 “For with what right could we bring into unlimited respect something that might be valid only under contingent human conditions?”8 And
whatever is derived from the particular natural situation of man as such, or from certain feelings and propensities, or even from a particular tendency of human reason which might not hold necessarily for the will of every rational being (if such a tendency is possible), can give a maxim valid for us but not a law … This is so far the case that the sublimity and intrinsic worth of the command is the better shown in a duty the fewer subjective causes there are for it and the more there are against it.9
Since Kant was certain that to root the moral law exclusively in human nature was to derogate from its authority, Korsgaard is not entirely right when she says that
Kant, like Hume and Williams, thinks that morality is grounded in human nature, and that moral properties are projections of human dispositions.10
For Korsgaard, morality is grounded in human nature,11 and that difference between her and Kant is consequential here, for Kant has a ready answer to Hobbes’s argument about the sovereign, whereas Korsgaard may have no answer to it, because she has abandoned the element of Kant that transcends merely human nature. She appears to agree with the insistence she attributes to Pufendorf and Hobbes, that the only possible source of obligation, not of its being good or sensible or beneficial or desirable that you do something, but of your having to do it, is that you are ordered to do it by a lawgiver. No “ought” without law and no law without a lawgiver. Korsgaard affirms all that,12 but she adds that the only person “in a position” to give that law is the self-commanding self. Accordingly, to secure the binding force of law, Korsgaard has to have a way of answering Hobbes’s second argument, and she does not have Kant’s way.
If Hobbes’s position can be rendered consistent, if he can bind the citizen to obey while nevertheless freeing the sovereign from all duty of obedience, then legislating qua sovereign must for some reason not be self-binding whereas legislating through a sovereign representative is. But Korsgaard’s persons are autonomous self-legislators: no delegation or representation occurs here. Accordingly, it is hard to see how anything becomes a law for them that they must obey. So Korsgaard can maintain the authority of the law over its subject legislator neither in Kant’s way nor in the just hypothesized revisionist Hobbesian way.
Suppose I ask: why should I obey myself? Who am I, anyway, to issue a command to me?13 Kant can answer that question. He can say that, although you legislate the law, the content of the law comes from reason, not from anything special about you, or your reason, or even human reason, but from reason as such. And, when that is so, then, perhaps, reflective endorsement of the law is inescapable. But if the content of the law reflects my nature, my engagements and commitments, then could I not change its content? Trafficking at the human level as she does, Korsgaard must say that my practical identity, with which the law is bound up, is inescapable, but one may doubt both that there is a special connection between morality and practical identity and that practical identity is inescapable. One may therefore doubt that Korsgaard can achieve her goal, which, so I read it, is to keep the “must” that Kant put into morality while humanizing morality’s source.14
I have said that, for Korsgaard, morality is rooted in human nature. But at one place Korsgaard says the somewhat different thing, that “value is grounded in rational nature—in particular in the structure of reflective consciousness—and it is projected onto the world.”15 But rational nature, if it means the nature of reason, is different from human nature, at least for Kant, and Korsgaard returns us to the fully human when she adds the gloss referring to the structure of reflective consciousness, for, as she will surely not deny, all manner of all-too-human peculiarities can gain strength in reflective consciousness. Kant can say that you must be moral on pain of irrationality. Korsgaard cannot say that.
3. What Korsgaard says, instead, is that you must be moral on pain of sacrificing your practical identity, which is to say, who you are from a practical point of view. You act morally because you could not live with yourself, “it could be. … worse than death,” if you did not.16
But I find it very difficult to put together the motif of practical identity with the emphasis on law that Korsgaard takes from Hobbes and Pufendorf and, especially, Kant. If morality is to do with law, then the liaison between morality and practical identity is questionable, since the commitments that form my practical identity need not be to things that have the universality characteristic of law. Practical identity is a matter of loyalty and identification, and whereas there is indeed such a thing as loyalty to general principles, there also exists loyalty to family, to group, to another individual; and no credible characterization of what practical identity is, in general terms, would yield a general priority for principled over particularistic identifications. Being Jewish plays a role in my practical identity, and so does being a Fellow of All Souls. But neither of those features signifies an attachment for me because I believe some principle that says: cleave to the ethnic group to which you belong, or to the College that was sufficiently gracious to receive you. As Bernard Williams famously said, if I save my wife not just because she is my wife, but because I believe that husbands in general have special obligations to their wives, then I act on “one thought too many.”17
My sacrifice for a person need not come from a general belief about right and wrong, but from solidarity with that person, and not because of characteristics which she and I have and which are such that, where characteristics of that kind obtain, solidarity is always required. I might find it hard to live with myself if I gave nothing to Oxfam, which is for me a matter of principle, but I would find it harder still to live with myself if I gave to Oxfam instead of paying for the operation that my mother needs. It does not distinguish my moral from my other commitments18 that if I resile on my principled ones I prejudice my practical identity.19
Korsgaard writes:
The reflective structure of human consciousness requires that you identify yourself with some law or principle which will govern your choices. It requires you to be a law to yourself. And that is the source of normativity. So the argument shows just what Kant said that it did: that our autonomy is the source of obligation.20
The reflective structure of human consciousness may require, as Korsgaard says, following Harry Frankfurt, that, on pain of reducing myself to the condition of a wanton, I endorse the first-order impulses on which I act, that, as we say, I identify myself with them. But it does not follow, and it is not true, that the structure of my consciousness requires that I identify myself with some law or principle. I do not do that when I identify myself with the impulse to save my own drowning child. What the reflective structure requires, if anything, is not that I be a law to myself, but that I be in command of myself. And sometimes the commands that I issue will be singular, not universal. If, as Korsgaard says, “the necessity of acting in the light of reflection makes us authorities over ourselves,” then we exercise that authority not only in making laws but also in issuing singular edicts that mean as much to us as general principles do.21
Using Richard Hare’s terms, we can say that Korsgaard’s solution is imperativist or prescriptive, but not universally prescriptive. And whether or not the moral must be law-like if it is prescriptive, Korsgaard says that it is law-like, yet it is just not true that every claim on me that survives reflection is, or, presupposes, a law.
Korsgaard remarks that, if she calls out my name, and I do not stop, then I am rebelling against her.22 She then asks, “But why should you have to rebel against me?” Why should my failure to stop count as rebellion? Her answer is that “[i]t is because I am a law to you. By calling out your name, I have obligated you. I have given you a reason to stop.” Well, suppose we accept that, just by calling out my name, you’ve given me a reason to stop. I think that could be regarded as extravagant (maybe you’ve [also] given me a reason to speed up), but suppose we accept it. Then, even so, what you have given me is no law but just an order, a singular order lacking the universality of law.
Now whether or not morality is, as Korsgaard thinks, a matter of law, it is false that whatever I do for fear of compromising my practical identity counts as moral, and also false that whatever counts as moral is done for fear of compromising identity. It is a huge exaggeration to say, as Korsgaard does, that “an obligation always takes the form of a reaction against a threat of a loss of identity.”23 I could remain me both in the evident banal sense and in every pertinent nonbanal sense if I gave nothing to help the distant dying who oppress my conscience. I just wouldn’t feel very good about myself. And I might even say, in morose reflection, “How typical of me, to be so bloody selfish.” And I might lose my grip on myself if I suddenly found myself being very philanthropic. So this is not Korsgaard’s point that my identity is solid enough to withstand a measure of wrongdoing that contradicts it.24 This is the different point that plenty of what I do that I regard as wrong does not challenge my identity at all.
4. Korsgaard provides two (entirely compatible) characterizations of the problem of normativity, one general and unexceptionable, and the other more specific and of a sort which makes the problem so difficult that it seems impossible to solve. With the hard version of the problem in hand, she finds it easy to dismiss rival solutions to it. I shall argue that, if we press the problem in its harder form against her own solution, then it too fails. But I shall also hypothesize that the resources of Korsgaard’s solution might be used to produce an interesting candidate for solving the normative problem if we characterize that problem differently, but not altogether differently, from the way she does.
Introducing the problem, Korsgaard says that what “we want to understand” is “the normative dimension,” which is that “ethical standards … make claims on us: they command, oblige, recommend, or guide. Or at least, when we invoke them, we make claims on one another.”25 The question is “Why should I be moral?” “We are asking what justifies the claims that morality makes on us. This is what I am calling ‘the normative question.’ ”26
Korsgaard lists three conditions which the answer to the normative question must meet. It must “succeed in addressing” someone who is in “the first-person position of the agent who demands a justification of the claims which morality makes upon him.” Consequently, and this is the second condition, a successful normative theory must meet the condition of “transparency”: when I know what justifies my acting as required, I must “believe that [my] actions are justified and make sense.” Third, “the answer must appeal, in a deep way, to our sense of who we are, to our sense of our identity … [Moral claims] … must issue in a deep way from our sense of who we are.”27
I shall concentrate, in Section 5, on the first condition, that the answer to the normative question must address the agent who asks it, for, as I shall argue, Korsgaard presents that agent as asking that question in so intransigent a spirit that I doubt that such an agent could be satisfied by any theory, Korsgaard’s included. Here I remark that Korsgaard’s third condition of adequacy on an answer is inappropriate in its assigned role. It is question-begging to say in advance that the answer must appeal to the agent’s sense of her own identity, even if that should indeed turn out to be a feature of the right answer.
5. Korsgaard’s answer to the normative question is that the reason why ethical standards make claims on me is that they represent commands that I give to myself, either in virtue of my practical identity or in exercise of my practical identity: I am not sure which of those is the right way to put her claim. On the first interpretation, the cost of violating ethical standards is loss of the practical identity that I would otherwise still have had; on the second, the cost is failure to have a practical identity, where I might never have had one anyway. I shall suppose that the first interpretation is correct—it fits more of what Korsgaard says.
Now, as I said, the further specification that Korsgaard attaches to the normative problem, the specification that fells the candidate solutions to it which are rival to her own, makes the problem so hard that, so I believe, her own solution too is seen to fall if, as she did not, we forthrightly confront it with her tough specification of the problem.
Return to the general characterization of the problem. The problem is to answer the question “Why should I be moral?” But consider two very different discursive contexts in which that question can occur. The first is the context of protest. “Why should I be moral? If I behave morally here, I wreck my career, I lose friends, I become poor …” The second is the context of self-justification. “Why should I be moral? Why should I act morally, like a decent human being? I’ll tell you why I should act morally. Because I could not live with myself if I did not.” Now, Korsgaard has to fashion an answer which meets the question in its first, protestant, guise. But I doubt whether anything can be guaranteed to persuade that questioner, and I am certain that Korsgaard can do no better at persuading him than the rivals she criticizes do. Yet her answer does fit what the person figured above says when he addresses the question in its second, and milder, guise.
At various points in lecture 1 Korsgaard taxes moral realism in particular with incapacity to answer the normative question in its protestant form. We are told, first, that
when the normative question is raised, these are the exact points that are in contention—whether there is really anything I must do, and if so whether it is this. So it is a little hard to see how realism can help,28
since all that realism can say is: well, it’s in the nature of things that this is what you must do. But, we have to ask, when so radical a stance of doubt is struck, how Korsgaard’s own answer can be expected to help. Again:
If someone finds that the bare fact that something is his duty does not move him to action, and asks what possible motive he has for doing it, it does not help to tell him that the fact that it is his duty just is the motive. That fact isn’t motivating him just now, and therein lies his problem. In a similar way, if someone falls into doubt about whether obligations really exist, it doesn’t help to say “ah, but indeed they do, they are real.” Just now he doesn’t see it, and therein lies his problem.29
But when he is in such a state, a state in which he does not feel the force of reason or obligation, that can be because in such a state, and, indeed, if Korsgaard is right, that must be because in such a state he does not feel the force of, does not see what is involved in, his practical identity; and, echoing Korsgaard, we can say: therein lies his problem. He asks: why should I continue to dedicate myself like this? And then there is no point saying to him: because that is what you are committed to. Korsgaard says that “the normative question arises when our confidence [‘that we really do have obligations’] has been shaken whether by philosophy or by the exigencies of life,”30 and that someone’s confident affirmation of the reality of obligation will then do nothing for us. But one thing which life’s exigencies can shake is a person’s practical identity, and, when that happens, then Korsgaard’s answer will not help. Something shatters my sense of being and obligation in the world, consequently my confidence that obligation is real. It is then useless to tell me that it lies in my practical identity to be thus obliged. When I doubt that “obligations really exist,” or do not recognize that moral “actions” are “worth undertaking,”31 I am setting aside any relevant practical identity that the philosopher might have invoked.
In expressing skepticism about whether Korsgaard’s—or anyone else’s—theory could address and convert the radically disaffected, I am not committing myself to skepticism about moral obligation. What I am skeptical of is the requirement that an answer to the normative question, in its general specification, has to sound good when addressed to the radically disaffected. If we scale down the difficulty of the question, we can, I think, find illumination in Korsgaard’s answer to it.
“The normative question,” says Korsgaard, “is a first person question that arises for the moral agent who must actually do what morality says … You … ask the philosopher: Must I really do this? Why must I do it? And his answer is his answer to the normative question.”32 But, to repeat my objection, if his answer is that it belongs to my practical identity to do it, then why am I asking the question in the alienated style on which Korsgaard insists?
If, on the other hand, we turn the thing around, we get something better. Suppose, again, that I am the moral agent, but this time not an alienated one, and I am faced by the skeptic who knows it will cost me to go on the march and who asks me why I bother. Then I can say a great deal that is persuasive about my practical identity. If I say, in radical disaffection: “I do not know why I should march,” then it is fatuous for you to reply: “Because your conscience compels you to.” But if you ask why I am going to do it, at substantial sacrifice of self-interest, it is not at all fatuous for me to reply, “because my conscience compels me to.” “Hier steh ich, ich kann nicht anders” makes sense. “Dort stehst du, du kannst nicht anders” is manifestly false for the case of extreme disaffection which Korsgaard insists a moral theory must address.
If Korsgaard’s defense of morality does not meet her own standard, which is that it should be capable of convincing the disaffected, then that could be because hers is the wrong standard, or hers is the wrong defense, or both. For my part, I am more clear that the standard is wrong than that the defense is. I do not think that we can show the intransigent why they should be moral. But I think that I can show the sincere inquirer why I must be moral. I have to be moral because, indeed, I could not otherwise live with myself, because I would find my life shabby if I were not moral. I can show that morality is a rational way, without being able to show that it is the (only) rational way.33
That morality is an option within rationality rather than a requirement of rationality necessitates the indicated first-person approach, in which the defender of morality is the moral agent herself. In the defense I sketch, the defender speaks in the first person, in Korsgaard’s in the second person, to me as a sincere but disaffected inquirer. So I am not against the proposal that the issue be framed in I-thou terms, but I think that the roles of speaker and audience need to be reversed.
Korsgaard calls her solution “the appeal to autonomy,” and in one place she describes it as follows:
[T]he source of the normativity of moral claims must be found in the agent’s own will, in particular in the fact that the laws of morality are the laws of the agent’s own will and that its claims are ones she is prepared to make on herself. The capacity for self-conscious reflection about our actions confers on us a kind of authority over ourselves, and it is this authority which gives normativity to moral claims.34
I have asked some questions about our supposed authority over ourselves in Section 2 above.35 Right now I want to register that the rhetoric of the foregoing passage is more suited to how I would explain why I bother to be moral than to what someone else could say if I intransigently insist on being told why I must be moral.
Your practical identity is given by the
[d]escription[s] under which you value yourself … description[s] under which you find your life to be worth living and your actions to be worth undertaking … these identities give rise to reasons and obligations. Your reasons express your identity, your nature: your obligations spring from what that identity forbids.36
I think all that is powerful stuff for me, the moral agent, to say to my interrogator, but it is entirely impotent when addressed to someone who, being disaffected, ex hypothesi finds no actions to be worth undertaking, or, more pertinently and more plausibly, no moral ones. It is powerful to say, “I couldn’t live with myself if I did that,”37 but off the mark to say, “You couldn’t live with yourself if you did that,” to someone who is evidently managing to do so.38 The intransigent person who insists on a justification for being moral is close to saying: “As far as my deep identity goes, I feel no force in morality’s claims.” To that little can be said, so that, if we set Korsgaard’s answer to the normative question against her own too demanding description of that question, then her answer to it does not work.
6. I have objected to Korsgaard’s claim that “[a]n obligation always takes the form of a reaction against a threat of loss of identity.” It is an overstatement, whatever may be the truth that it overstates.39 Not all obligations are like that. But, even if they were, it would remain true that, as I have also complained, not everything that is like that is an obligation of the sort for which, we can suppose, Korsgaard wants to supply foundations.
Consider an idealized Mafioso: I call him “idealized” because an expert has told me that real Mafiosi don’t have the heroic attitude that my Mafioso displays. This Mafioso does not believe in doing unto others as you would have them do unto you: in relieving suffering just because it is suffering, in keeping promises because they are promises, in telling the truth because it is the truth, and so on. Instead, he lives by a code of strength and honor that matters as much to him as some of the principles I said he disbelieves in matter to most of us. And when he has to do some hideous thing that goes against his inclinations, and he is tempted to fly, he steels himself and we can say of him as much as of us, with the same exaggeration or lack of it, that he steels himself on pain of risking a loss of identity.
What the Mafioso takes to be his obligations can be made to fit Korsgaardian formulae about loss of identity as much as what most of us would regard as genuine obligations can be made to fit those formulae. So it looks as though what she has investigated is the experience or phenomenology of obligation, not its ground or authenticating source. Autonomy, she says, “is the source of obligation, and in particular of our ability to obligate ourselves,”40 but the Mafia man has that as much as anyone does, this capacity to transcend impulses through reflection and endorse or reject them.
Korsgaard realizes that she might be interpreted as I interpret her when I press the Mafioso example. Accordingly, she emphasizes that “the bare fact of reflective endorsement … is [not] enough to make an action right.”41 It cannot be enough, she says, because, while there always is at least a minimal reflective endorsement of action, not all action is right. The argument has merit as far as it goes, but it is unreassuring, since it is consistent with the view that more than minimal reflective endorsement does always make an action right. It is unreassuring that the reason given for denying that reflective endorsement always makes an action right does not confront the reason we have for fearing that it might, reasons like this one: that the Mafioso is entirely capable of (more than minimal) reflective endorsement.
At 3.4.4 Korsgaard seems to grant the present insistence, that the apparatus of reflective endorsement and practical identity is content-neutral: she thinks that it gains its different contents from the different social worlds that self-identifying subjects inhabit. But then we do not have what was demanded in the original characterization of the problem of normativity, which was an answer to the question why I must do the specifically moral thing. Unless, again, we turn the question around, and you are asking me why I undertake the labor of morality, as such. If I do undertake it, I can explain why. If I don’t, Korsgaard supplies nothing sure to work that you can say to me, for morality might not be part of the practical identity that my social world has nourished. Or, worse, my social world might indeed be a morally constituted one, but the nourishment might have failed to take in my case.
7. An attempt to derive specifically moral obligation is prosecuted in Korsgaard’s lecture 3: see, in particular, sections 3.4.7–3.4.10 and the important summary at 3.6.1. I shall here articulate the argument which I believe to be embodied in the cited sections, and which I find multiply questionable. I shall then pose some of the questions that I have in mind.
Here, then, is what I take to be Korsgaard’s argument:
1. |
Since we are reflective beings, we must act for reasons. |
But |
2. If we did not have a normative conception of our identities, we could have no reasons for actions. |
So |
3. We must have a normative conception of our identities (and our factual need for a normative identity is part of our normative identity).42 |
So |
4. We must endorse ourselves as valuable. |
So |
5. We must treat (all) human beings as valuable. |
So |
6. We find human beings to be valuable. |
So |
7. Human beings are valuable. |
So |
8. Moral obligation is established: it is founded in the nature of human agency. |
The above argument can be decomposed into four subarguments, on each of which I now invite focus: (1) from 1 to 3, (2) from 3 to 5, (3) from 5 to 7, and (4) from 7 to 8.
(1) The passage from 1 (which I shall not question) to 3 rests on 2, but I do not see that 2 is true, except in the trivial sense that, if I treat something as a reason, then it follows that I regard myself as, identify myself as, the sort of person who is treating that item, here and now, as a reason. I do not see that I must consult an independent conception of my identity to determine whether a possible spring of action is to be endorsed or not, nor even that such endorsement must issue in such a conception, other than in the indicated trivial sense. When I am thirsty, and, at a reflective level, I do not reject my desire to drink, I have, or I think that I have, a reason for taking water, but not one that reflects, or commits me to, a (relevantly) normative conception of my identity. Merely acting on reasons carries no such commitment.
(2) The inference from 3 to 5 depends on the idea that, being, as we are, inescapably reflective, we must employ the normative conception of our identities (that we therefore necessarily have) to “endorse or reject”43 the impulses which present themselves to us as possible springs of action. But the very fact (supposing that it is one) that I must endorse and reject shows that I do not endorse a human impulse just because it is a human impulse. Human impulses are not, therefore, of value just because they are human. So, consistently with the structure of reflective consciousness, I can pass harsh judgment on my own, or on another’s panoply of desires and bents, the more so if that other is disposed to endorse them. And if my endorsement of a given impulse means that I regard my humanity as pro tanto of positive value, then, by the same token, my rejection of another impulse must mean that I regard my humanity as pro tanto of negative value. No reason emerges for the conclusion that I must treat human beings, as such, as valuable, or for the requirement, which some might think a Kantian morality embodies, that I must treat them as equal in their value.
(3) The inference from 5 to 7 might be thought to illustrate the fallacy of equivocation, for it seems to depend on an ambiguity in the expression “to find,”44 which is sometimes a success-verb, where what is found to be thus and so must be so, and sometimes not. There is a sense of “find” in which 6 follows from 5, and another one in which 7 follows from 6: but Korsgaard needs one sense, on pain of equivocation. Yet this comment of mine may reflect boneheadedness on my part about the character of transcendental arguments, for this is supposed to be one: maybe, in a transcendental argument, “find” in its (normally) weaker sense is good enough to derive such a conclusion as 7. Accordingly, being uncertain whether there is any objection worth raising here, I pass on.
(4) My final comment concerns the passage from 7 to 8. My difficulty with it is that it appears to me that the Mafioso can accept 7, in any sense in which what precedes it shows that it is true (I rely here on points made in comment (2) above), yet reject 8. For the Mafioso can honor human beings the springs of whose actions are congruent with his own practical identity. So whatever endorsement of humanity as such comes out of this argument, it seems to me not to distinguish the Mafioso ethic from morality, and therefore not to move us beyond the mere phenomenology of obligation to providing a foundation for specifically moral obligation.
The problem lies in our freedom at the level of endorsement, the old problem with which these remarks began: that the sovereign can change the law. To hammer that home a bit more, I want to look at Harry Frankfurt’s concept of free personhood.
8. The debt that Korsgaard acknowledges to Frankfurt45 is instructive in connection with my related claims that she has offered an option for the first person rather than a constraint that the second person must accept, and that what she has enabled the first person to provide is a defense of any set of commitments and not of specifically moral ones.
For Frankfurt, I am free when my will conforms to a higher-order volition, when, that is, I act on a first-order volition that I wish to act on, when the spring of my action is one that I want to be moved by.
We should pause to modify this formula. We should add a restriction, a further condition for such conformity to betoken freedom, which has to do with the direction of the conformity. That is, the direction of conformity must be that my lower will conforms to my higher one, for, if it goes the other way, if my higher adjusts to my lower, then we have not freedom but second-order adaptive preference formation.46 That category covers the addict who has come to endorse his own pursuit of drugs: he now likes desiring drugs, and he likes acting on that desire. (I do not say that a willing addict’s second-order volition could not be determinative: I am just using as an example the more plausible case in which it is not.)
But there is no restriction either in Frankfurt’s presentation or in fact on what the content of second-order volition can be, or, better, for this weaker claim will suffice here, no restriction sufficiently restrictive to yield moral obligation.47 Thus, to return to my example, the ideal Mafioso is entirely capable of Frankfurt freedom: he can prescribe the Mafia ethic to himself. Yet, to repeat my qualified defense of Korsgaard, I can defend my ethic even to him. I can explain why I strive not to succumb to some of my first-order desires, including some that move him.
Reference to Frankfurt also reinforces the point made earlier, that Korsgaard’s legislator is too like the Hobbesian sovereign, as opposed to the Hobbesian citizen, to serve as the sort of model she requires. The Kantian reflective endorsement is inescapable, but Frankfurt’s person, like Hobbes’s sovereign, is at liberty to reassess his commitments.
So, I return to the thought that something transcending human will must figure in morality if it is to have an apodictic character. Kant was right that, if morality is merely human, then it is optional, as far as rationality is concerned. But it does not follow that morality cannot be merely human, since Kant may have been wrong to think that morality could not be optional. What does follow is that Korsgaard’s goal is unachievable, because she wants to keep the “must” that Kant put into morality while nevertheless humanizing morality’s source.
1 They are (i) the consequentialist argument that the state of nature is intolerable, (ii) the argument of hypothetical consent: any rational person would agree to submit to government, and (iii) the argument of actual consent: all citizens in fact agree to submit to government.
2 Hobbes, Leviathan, pp. 221, 265, and cf. ibid., p. 276.
3 Stepwise, the argument runs as follows:
What the sovereign does, I do. | |
But | The sovereign makes the law. |
So | I make the law. |
Now, | It is absurd to object to what I myself do. |
So | I cannot object to the law. |
So | I must obey the law. |
In a different version of the argument, which Hobbes also gives, the further premise is not that I cannot object to what I myself do but that I cannot “injure” (that is, do an injustice to) myself. The argument then runs as follows:
What the sovereign does, I do. | |
But, | A man cannot injure himself. So |
So | The sovereign does not injure me. |
So | I cannot object to what the sovereign requires of me. |
So | I must obey the sovereign. |
4 Hobbes, Leviathan, pp. 367, 313, and see Hobbes, De Cive, 12.4, 11.14.
5 Kant, Foundations of the Metaphysics of Morals, pp. 49–50 (emphases added, except on “autonomy” and “heteronomy”).
6 Kant, Critique of Practical Reason, p. 85: “We are indeed legislative members of a moral realm which is possible through freedom and which is presented to us as an object of respect by practical reason; yet we are at the same time subjects in it, not sovereigns, and to mistake our inferior positions as creatures and to deny, from self-conceit, respect to the holy law is, in spirit, a defection from it even if its letter be fulfilled.”
7 Kant, Foundations of the Metaphysics of Morals, p. 5.
8 Ibid., p. 24.
9 Ibid., p. 42. cf. pp. 28, 65, and Kant, Critique of Practical Reason, p. 19.
10 Korsgaard, The Sources of Normativity, p. 91.
11 See, ibid., e.g., pp. 131–32.
12 See her conclusion, ibid., pp. 164–65. Korsgaard’s claim that the solution to the problem of normativity must be imperativist puzzles me. She brings out, brilliantly, the difference between doing logic because of its merits as a subject and doing it because it is a required course (pp. 25–27, pp. 105–7), but why isn’t it good enough if our reason for being moral is as good as the one a person has for doing logic where that isn’t a required course? This question relates to the point that morality is a choice within rationality, not a requirement of it: see Section 5 below.
13 Korsgaard rejects Pufendorfian voluntarism, remarking that “the very notion of a legitimate authority is already a normative one and cannot be used to answer the normative question” (p. 29). In her own answer to that question, I am the legislating will, so I must possess legitimate authority for my legislation to be valid. Does it follow, in a defeating way, that I cannot cite my own legitimate authority any more than Pufendorf can cite God’s, by way of answer to the normative question? I think that Korsgaard’s answer would be that I must have authority for me, in virtue of some or all of the argumentation that I discuss in Section 7 below.
14 For further comment on Korsgaard’s project, so conceived, see the final paragraph of the present reply.
15 Korsgaard, The Sources of Normativity, p. 116.
16 Ibid., p. 18.
17 Williams, “Persons, Character, and Morality,” p. 18.
18 And—see below—it is not even true of all of my moral ones, any more than it is true of all of my nonmoral ones.
19 For a sensitive defense of the claim that one can act unselfishly for the sake of a collective to which one belongs, and other than for reasons of principle, see Oldenquist, “Loyalties.” For an illuminating application of the point to Marx on proletarian solidarity, see the section on “Morality” in Miller, Analyzing Marx, especially pp. 63–76, and see too, in the same connection, Whelan, “Marx and Revolutionary Virtue,” pp. 64–65.
20 Korsgaard, The Sources of Normativity, p. 174.
21 The text continues: “And in so far as we have authority over ourselves, we can make laws for ourselves and those laws will be normative” (p. 165). Yes, we can make laws in exercise of our authority, if, indeed, we have it. But, as I’ve just protested, we can do other relevant things too, in exercise of that (supposed) authority.
22 Ibid., p. 140.
23 Ibid., p. 102.
24 Ibid., pp. 103, 158–60.
25 Ibid., p. 8.
26 Ibid., pp. 9–10.
27 Ibid., p. 16–18. I cannot here forbear from the comment which restates, in a different way, points made in Section 3, that who we are is not what we are. Who I am is a matter of my specified situation. And that takes us miles away not only from Kant, for whom only what we are enters the moral, but from the specifically moral, on any account of morality.
28 Ibid., p. 34.
29 Ibid., p. 38.
30 Ibid., p. 40.
31 Ibid., p. 38 (quoted more fully in text to note 29 above) and p. 102 (quoted more fully in text to note 36 below).
32 Ibid., p. 16, and see pp. 85–86.
33 I said earlier that not all instances of failure to be moral compromise my practical identity. But to not be moral at all would wreck my practical identity, and that of all my fellow nonsociopaths. (I am conscious that this qualified rehabilitation of Korsgaard may achieve nothing more than a return to the Williams position that she wanted her own to supersede.)
34 Korsgaard, The Sources of Normativity, pp. 19–20.
35 Recall the Hobbesian conundrum with which I began, the problem of whether I have the authority to legislate over myself. If you say to me: but look, it is your law, your practical identity, then I might say: yes, but who am I to impose such a law on me? But when I say, “Hier steh ich,” then it is odd for you to say: but who are you to issue such a command to yourself?
36 Korsgaard, The Sources of Normativity, pp. 101.
37 Ibid.
38 “A human being is an animal whose nature it is to construct a practical identity which is normative for her. She is a law to herself. When some way of acting is a threat to her practical identity and reflection reveals that fact, the person finds that she must reject that way of acting, and act in another way. In that case she is obligated” (ibid., p. 150).
But you can’t get me to construct a practical identity that will matter to me. And, if I do have one, then there is my answer to you.
39 See above for the protest against it and note 33 for the element of truth in it.
40 Korsgaard, The Sources of Normativity, p. 91.
41 Ibid., p. 161.
42 See, in particular, ibid., pp. 125–26.
43 Ibid., 120.
44 See ibid, pp. 123–25.
45 Ibid., p. 99n.
46 Second-order adaptive preference formation because you adapt, here, not your first-order desire to the course of action that’s available, but your second-order desire to the first-order desire that’s available, or even unshakable.
47 Frankfurt himself supports the stronger claim:
In speaking of the evaluation of his own desires and motives as being characteristic of a person, I do not mean to suggest that a person’s second-order volitions necessarily manifest a moral stance on his part toward his first-order desires. It may not be from the point of view of morality that the person evaluates his first-order desires. Moreover, a person may be capricious and irresponsible in forming his second-order volitions and give no serious consideration to what is at stake. Second-order volitions express evaluations only in the sense that they are preferences. There is no essential restriction on the kind of basis, if any, upon which they are formed. (Frankfurt, “Freedom of the Will and the Concept of a Person,” p. 19, fn. 6)