Rethinking intrinsic value

Journal of Ethics

Volumn 2, Issue 4, 1998, Pages 277-297

  Kagan, Shelly ^a  

^a YALE UNIVERSITY   (United States)

Author keywords

Instrumental value; Intrinsic properties; Intrinsic value

Indexed keywords

EID: 0001069854     PISSN: 13824554     EISSN: 15728609     Source Type: Journal    
DOI: 10.1023/A:1009782403793     Document Type: Article

References (13)

1. The contrast between intrinsic and relational properties is indeed only a rough one, since (as Gary Rosenkrantz pointed out to me) some relational properties are actually intrinsic (for example, certain relations between an object and its parts). For simplicity, however, I'll continue to refer to "relational properties"; this can be read as shorthand for "nonintrinsic relational properties." Obviously, giving an adequate characterization of the distinction between intrinsic and nonintrinsic properties would be a difficult matter; I won't attempt that here.
2. There are, of course, still other questions as well. For example, does everyone have reason to promote every object that has value as an end, or is such value, rather, "agent- relative," so that for any given valuable object only particular individuals have reason to promote it? But I will leave these important questions aside.
3. Failure to see this point seems to me the most significant error in Christine Kors- gaard's otherwise commendable "Two Distinctions in Goodness" [reprinted in her Creating the Kingdom of Ends (Cambridge: Cambridge University Press, 1996), pp. 249-274]. Korsgaard makes many points similar to those I make here.
4. Though the case of a creature that values itself may provide an exception to this generalization: isn't it an intrinsic property of the creature that it values itself?
5. To mention just one aspect in need to further specification, a view like this will presumably have to appeal to some notion of a minimal relevant level of usefulness. After all, even "useless" skills have some instrumental value.
6. That is, but for its instrumental value, the pen would have no intrinsic value at all. Note, however, that to say this is not to claim that the pen's being instrumentally valuable is the only feature relevant to its having intrinsic value. One could still insist, plausibly, that other properties help ground the pen's intrinsic value as well. Thus, instrumental value may here be necessary for intrinsic value, without being sufficient.
7. Let me note, if only in passing, that similar locutions may be necessary for other types of value as well. I take it, for example, though I won't argue the point here, that symbolic value is not necessarily a form of instrumental value. And typically, no doubt, having symbolic value does not itself ground an object's having intrinsic value. But it does seem possible to me that for at least some symbols the symbolic value does itself provide (at least part of) the basis of the object's intrinsic value. So we may need to distinguish between "mere" symbolic value, on the one hand, and "intrinsically valuable symbolic value," on the other.
8. There is no threat of infinite regress here. If you and I help each other, for example, then each of our lives has intrinsic value, and each of our lives has instrumental value - and each life has its intrinsic value by virtue of its having instrumental value.
9. This is Korsgaard's proposal, in "Two Distinctions in Goodness."
10. Consider, for example, logical goodness (the goodness that an argument has when it is a logically good - i.e., valid - argument). This is presumably a kind of value that depends solely upon the intrinsic properties of the objects that have it (that is, arguments); yet few would take it to be an instance of value as an end. (I owe this example to Fred Feldman, "Hyperventilating about Intrinsic Goodness," in this issue of The Journal of Ethics.)
11. See Shelly Kagan, "The Additive Fallacy" [Ethics 99 (1988), pp. 5-31], for a related discussion of a similar point from another part of ethics.
12. This point is made with regard to the particular case of pleasure in Fred Feldman, "On the Intrinsic Value of Pleasures" [Ethics 107 (1997), pp. 448-166]. I presume that he also sees that the move can be generalized, although he doesn't say so explicitly there.
13. But perhaps I misunderstand the dominant philosophical tradition on this matter. After all, some philosophers firmly within that tradition do note the relevance to intrinsic value of such apparently nonintrinsic properties as knowing, as opposed to merely believing. (I owe this objection to Ben Bradley.) Note, however, that precisely in such cases friends of the tradition typically feel the need to start talking about the intrinsic value of (facts about) complex wholes (consisting of, for example, the knower and the object known). This allows them to insist that what is actually relevant to intrinsic value is simply an intrinsic property of the whole (roughly, that one part, the knower, stands in the right relation to another part, the object known). Yet it is often difficult to see what motivates the turn to wholes in this way, except the very belief that the only base level properties that are relevant to intrinsic value are intrinsic properties.