Autonomy and the second person within: A commentary on Stephen Darwall's the second-person standpoint

Ethics

Volumn 118, Issue 1, 2007, Pages 8-23

  Korsgaard, Christine M ^a  

^a NONE

Author keywords

[No Author keywords available]

Indexed keywords

EID: 38949166694     PISSN: 00141704     EISSN: None     Source Type: Journal    
DOI: 10.1086/522019     Document Type: Review

References (17)

1. I am tempted to say this is true if we talk to them at all, not as we might chatter to a pet, but with an expectation that they will understand. For given the normativity of meaning, talk is a kind of command - a command to think certain thoughts, although not necessarily to credit them. (For this comparison see Korsgaard, The Sources of Normativity [Cambridge: Cambridge University Press, 1996], 136-42.) Darwall resists the idea that his view shows that slavery involves "conceptual confusion or some sort of pragmatic contradiction" (SPS, 265). That's right: it is a normative felicity condition that is violated by addressing a command to a subject that one undertakes to treat as a mere object. But Darwall should not resist the idea that there is some sort of confusion or contradiction here. If treating a subject as a mere means or an object is the general form of immorality, then there is something especially perverse about both issuing commands to your slaves and lying, for in both of these cases it is the subject's subjectivity itself that is treated as a means.
2. References to Kant's ethical works are inserted into the text using the usual method of citing the volume and page number of the standard German edition of Kants gesammelte Schriften (published by Prussian Academy of Sciences in Berlin). The editions of Kant's ethical works I have used are those edited and translated by Mary Gregor for the Cambridge Texts in the History of Philosophy Series: Groundwork of the Metaphysics of Morals (Cambridge: Cambridge University Press, 1997) and Critique of Practical Reason (Cambridge: Cambridge University Press, 1997). For the Critique of Pure Reason I have used Norman Kemp Smith's edition (Boston: St. Martin's Press, Macmillan, 1929), citing the pages of the A and B editions as he does.
3. Both characterizations are on SPS, 11.
4. See Korsgaard, The Sources of Normativity, esp. 104-5 and 150-51.
5. I do not entirely agree with this. If we could give an otherwise adequate account of practical reason without taking on the commitments of substantive realism about reasons, we would not need to argue against substantive realism over and above doing that. This is because the only reason for accepting the commitments of substantive realism is the worry that we cannot give an adequate account of practical reason without them. I do agree that the Kantian must accept the burden of showing that her account yields a superior conception of agency, however.
6. This cannot strictly be correct, since theoretical reason in Darwall's sense will rule in beliefs like "X, Y, and Z cannot all be true." But perhaps these are not beliefs "about the world" in the sense Darwall has in mind.
7. Logic for Kant gives laws to thinking, and thinking can be either theoretical or practical, so there is nothing specifically " theoretical" about norms of coherence and consistency. And theoretical reason, according to Kant, also makes its own unconditional demands on the world. It demands that the world be a realm of causally interacting objects, and it also demands that the world conform to the teleological principles of speculative metaphysics: that it be a realm of free agents ordered to the good under the government of a creator who guarantees that ordering. But only the world of appearances meets the first demand, and the world does not, or cannot be known to, meet the teleological demand at all. Practical reason, Kant thinks, provides us with the materials to give positive content to the demands of speculative metaphysics. This fact provides what Kant calls, in the second Critique, a "credential" for the moral law (5:48), which he thinks takes the place of the deduction. Practical reason also succeeds in imposing its unconditional demand on one thing in the world, the will: that is the doctrine of the Fact of Reason, discussed in the text below. That, according to Kant, is the difference between practical and theoretical reason.
8. One may also have doubts about whether attitudes can be normatively guided. This is not to deny that attitudes may be, in Scanlon's sense, judgment sensitive (T. M. Scanlon, What We Owe to Each Other [Cambridge, MA: Harvard University Press], 20-22). But the kind of influence reason has on attitudes is clearly quite different from the kind it has on actions: one does not have an attitude "for a reason" or as a result of having taken something as a reason for the attitude. Of course we do say things like "you ought not to be afraid." But when we then go on to spell out the reasons why you ought not to be afraid, we do not expect you to draw the conclusion that you ought not to be afraid and as a result stop being afraid. Rather, we hope the reasons themselves will operate on your fear directly. Your agency is not involved in the same way as it is in acting for a reason.
9. Darwall argues that this idea - that reason must not accept direction from outside - also applies equally to theoretical and practical reason, and so does not explain the distinctive sense in which practical reason is autonomous. As I have already suggested (see n. 7), I think Darwall is looking for the difference in the wrong place. Darwall will want to argue that if we insist on taking negative and positive freedom in a stronger sense than the very weak sense he proposes here, it will seem mysterious that our beliefs have to conform to the evidence of our senses-because that will now seem to violate the principle that reason must not be directed from outside. I think that part of the difficulty is that Kant thinks the achievement of understanding, not the truth of belief, is the enterprise of theoretical reason. And part of the difficulty is that although Kant would agree that we are trying to conceptualize the world that is "there anyway" (as Darwall likes to say), for him that does not mean that our conception is supposed to "match" the world that is there anyway. Of course our beliefs must be true-they are responsible to the world that is there anyway-and this is part of the difference between theoretical and practical reason. But the difference is rather subtle. We can reject an apparent "perception" that does not fit our best scientific understanding of the world, just as we can reject a desire that is ruled out by the Categorical Imperative as not providing any reason to act. But in the first case, we seem to need an explanation of why the mental state in question seemed to be a perception, or why the perception was misleading, whereas we do not seem to need anything further in order to dismiss the outlaw desire, at least for practical purposes.
10. According to another way to read the story, this first example already shows that the man is free in a sense - he can resist his inclination when prudence demands it. Prudence is one sort of command of reason, and we can follow it, so we must be free. That Kant did hold an idea like this at one point we know, from some remarks he makes in the Critique of Pure Reason, in the Canon of Pure Reason. There Kant said that we can know what he calls "practical freedom" through experience, because "the human will is not determined ... by that ... which ... immediately affects the senses; we have the power to overcome the impressions on our faculty of sensuous desire, by calling up representations of what, in a more indirect manner, is useful or injurious" (A802/B830). One might take this example as illustrating that point - the man displays practical freedom in being moved by a representation of the injurious - gratifying his lust will be injurious if he is hanged for it. I do not think this is the right way to read the passage in the second Critique, since Kant is clear in the second Critique that we learn about our freedom from morality. But it would be possible to argue that Kant intends to be using the pair of examples to be arguing from practical freedom to transcendental freedom.
11. Kant is using the German version of "ought implies can," not the English version. In the English version, the fact that I cannot A is used to rule out the claim that I ought to A. In the German version, the fact that I ought to A is used to rule in the claim that I can.
12. The last two paragraphs are lifted from my introduction to Kant's Groundwork of the Metaphysics of Morals in the Cambridge Texts in the History of Philosophy edition (Cambridge: Cambridge University Press, 1997).
13. The Rossian intuitionist is an awkward example for my point, since he accepts a bunch of different norms, each of which is conditional upon the implications of the others in any given case. But presumably the Rossian intuitionist thinks his deliberative conclusion expresses an unconditional requirement, and that is enough for purposes of the argument: he thinks it is a categorical imperative.
14. A parallel move is made later, by John Rawls, when he proposes that any proposed principles of justice must be tested by constructing the original position in which it would be chosen, and seeing whether its assumptions are plausible; see A Theory of Justice, 2nd ed. (Cambridge, MA: Harvard University Press, 1999), 102-5, 130-68.
15. The formulation here, which suggests that second-personal reasons are real because we can validate them by viewing them from a more objective standpoint, recalls Thomas Nagel's views about the role of assuming more objective standpoints in The View from Nowhere (New York: Oxford University Press, 1986).
16. I think this is no accident. In fact, I think we can usefully see Darwall's book in general as presenting us with a variant of Nagel's argument in his first book, The Possibility of Altruism (Princeton, NJ: Princeton University Press, 1970). Nagel argued that we can capture the essence of moral reasons if we think about how a certain familiar form of moral argument is supposed to work. When we ask "how would you like it if someone did that to you?" Nagel proposed, we invite the agent to view himself and his victim as interchangeably situated "someones," and to conclude that doing that to any "someone" is objectionable. In a similar way, Darwall's argument invites us to view ourselves and others as interchangeably situated "yous" and to conclude that doing certain things to anyone with whom we stand in a "I-thou" relationship is objectionable. Although Darwall doesn't make this comparison explicitly, he does draw a contrast between his own view and Nagel's along these lines at SPS, 102.
17. Kant's argument that anyone who accepts an unconditional requirement is in effect accepting the Categorical Imperative amounts to an argument that it is psychologically impossible to respond to an unconditional demand without responding to the Categorical Imperative.