¹ According to many, equal opportunity requires in addition that persons be given equal chances to develop their qualifications – that they enjoy equally favorable developmental conditions and access to training. See, for example, Williams, Bernard, “The Idea of Equality,” in Problems of the Self (Cambridge: Cambridge University Press, 1973)CrossRefGoogle Scholar; Rawls, John, A Theory of Justice (Cambridge, MA: Harvard University Press, 1971)Google Scholar, ch. 2; and Fishkin, James, Justice, Equal Opportunity, and the Family (New Haven, CT: Yale University Press, 1983)Google Scholar, ch. 2. In the present paper, I will not be concerned with this further claim.

² The idea of increasing people's control over their lives pervades discussions of equal opportunity. Its influence can be discerned both in the familiar view that providing opportunities consists precisely in removing obstacles to the achievement of goals (see, for instance, Fullinwider, Robert, The Reverse Discrimination Controversy (Totowa, NJ: Rowman and Littlefield, 1980)Google Scholar, ch. 7) and in the claim, cited in footnote 1, that the obstacles to be removed include not only legal barriers but also the lack of preparation and training. Yet despite the pervasiveness of the idea of control, its justificatory role is not often brought into the open. In the published defenses of equal opportunity that I have read, the most straightforward statement of this that I have found is William Galston's remark that “equal opportunity may be defended on the grounds that it is conducive to personal satisfaction. Within the limits of competence, individuals are permitted to choose their lives' central activity….” (Galston, William A., “Equality of Opportunity and Liberal Theory,” Lukash, Frank S., ed., Justice and Equality Here and Now [Ithica, NY: Cornell University Press, 1986], p. 100Google Scholar). Ironically, a more forceful statement is presented by critic John Schaar, who correctly notes that a good part of the appeal of equal opportunity is that it purportedly “(does not) set artificial limits on the individual. On the contrary, it so arranges social conditions that each individual can go as high as his natural abilities will permit” (Schaar, John, “Equality of Opportunity, and Beyond,” Pennock, J. Roland and Chapman, John, eds., Equality; Nomos IX (New York: Atherton Press, 1967), p. 233Google Scholar).

³ Fishkin, Justice, Equal Opportunity and the Family, p. 24.

⁴ For discussion of the implications of these facts, see Schaar, “Equality of Opportunity, and Beyond”; Rawls, A Theory of Justice, sees. 12, 17, 48, and passim; and my “Effort, Ability, and Personal Desert,” Philosophy and Public Affairs vol. 8, no. 4 (Summer 1979), pp. 361–76.

⁵ This point is made by Schaar in “Equality of Opportunity, and Beyond.”

⁶ Of course, this aim will only be worth pursuing if the desert of the best qualified has real normative force – if there is good reason for an applicant to get the job he deserves. In my book Desert (Princeton, NJ: Princeton University Press, 1987), I argue that such desert claims do have normative force and that they draw this force from an obligation to treat all persons as agents rather than mere recipients of largesse or passive links in causal chains. Roughly, the argument is that in hiring those who actually will perform best, we focus on the difference the applicants' own actions can make in the world. By thus taking seriously their ability to influence events through their own intentional acts, we take them – the originators of these acts – seriously as well. If correct, this argument will obviously deepen the case for saying that the real basis of an applicant's desert is his potential future performance itself.

⁷ For elaboration of this argument, see my “Groups and Justice,” Ethics vol. 87, no. 2 (January 1977), pp. 174–81.

⁸ Thurow, Lester, “A Theory of Groups and Economic Distribution,” Philosophy and Public Affairs vol. 9, no. 1 (Fall 1979), p. 30.Google Scholar

⁹ ibid., p. 31.

¹⁰ If we assume that our use of group characteristics to predict performance is what confers moral standing on groups, it may follow that the only groups that can have moral standing are those whose defining characteristics support predictions. This in turn may seem to imply that an arbitrary collection of individuals, such as the group of all those not chosen, lacks moral standing, and thus that its disproportionate exclusion is not objectionable. But in the current context, what defines this group is precisely its members' lack of whatever characteristics license the selection of members of its complement. If the possession of these characteristics is predictively significant, then the fact that an applicant lacks them while others possess them surely is also. Thus, on the suggested account, the group of those not chosen can have moral standing.

¹¹ Thurow, “A Theory of Groups and Economic Distribution,” p. 30.

¹² ibid., pp. 29–30.

¹³ Here I assume that the strength of one's obligation to produce good consequences depends on how good the consequences are compared to those of alternative acts. Although this assumption is controversial, I believe that it underlies, and is needed to explain, our judgments about how much weight to assign to consequentialist obligations when these conflict with nonconsequentialist obligations.

¹⁴ See, for example, Nagel, Thomas, “Equal Treatment and Compensatory Discrimination,” Philosophy and Public Affairs vol. 2, no. 4 (Summer 1973), pp. 348–63.Google Scholar

¹⁵ Upon members of which groups is “underrepresentation” in professions actually likely to inflict psychic damage? Although blacks and women are the two groups most often mentioned, I am skeptical about the parallel between them. Besides occupying comparatively few positions of security or authority, blacks often must endure miserable housing, do without amenities that others take for granted, and depend in humiliating ways on the decisions of bureaucrats. By contrast, women observing other women find no such gross failure to procure a share in life's goods. Given this difference, I suspect the message conveyed by the “underrepresentation” of blacks is far more undermining than any message conveyed by comparable “underrepresentation” of women. I suspect, too, that when predictive methods do contribute to a group's “underrepresentation,” their harmfulness will vary with such factors as the visibility of those who occupy the contested positions, the prestige that attaches to those positions, and the shamefulness of not possessing the abilities that constitute qualifications for them. Since the last two factors are largely absent in the case of firefighters, it seems unlikely that strength tests for aspiring firefighters are harmful. But given this paper's theoretical purposes, these (admittedly controversial) issues may be set aside. We can agree that contributing to a psychically damaging employment pattern tells against a method of predicting performance without agreeing which methods do in fact have that effect.