¹ Inclusive fitness holds that all organisms act to maximize not only their individual fitness, but the fitness of their entire genetic line as well. Thus parents will normally take into account the welfare of their children to reflect the 50 percent of the genetic endowment that they have in common. The level of overlap with grandchildren is 25 percent, and the level between siblings is 50 percent (half of the genes each offspring receives from one parent will be held by the other offspring). See generally Hamilton, W.D., “The Genetical Evolution of Social Behavior,” Journal of Theoretical Biology, vol. 7 (1964), p. 1.CrossRefGoogle Scholar Given inclusive fitness, organisms then act to maximize not their individual welfare, but their overall genetic endowment over time.

² I have examined this theme in a related paper, “The Utilitarian Foundations of Natural Law,” Harvard Journal of Law and Public Policy (1989), p. 713.

³ See Hayek, F.A., The Road To Serfdom 74 (1944)Google Scholar; the more ambitious program is set out in National Broadcasting Co. v United States, 319 U.S. 190 (1943) (per Frankfurter J.). The source of the conflict is this: Hayek uses the example of the highway to explain how it is possible to develop an institution of common property without having to know the reasons and purposes for which people want to use the road. Lane assignments, speed limits, and other traffic rules can be set well without that knowledge. Frankfurter rejected this model of rules in determining the Federal Communication Commission's (FCC) administrative allocation of broadcast licenses. A Hayekian mode of analysis would only ask whether the license trespassed (that is, interfered) with broadcasters using nearby frequencies. But Frankfurter held that the FCC could not only determine these relatively simple matters, but do far more:

But the [Federal Communications] Act does not restrict the Commission merely to supervision of the traffic. It puts upon the Commission the burden of determining the composition of that traffic. The facilities of radio are not large enough to accommodate all who wish to use them. Methods must be divided for choosing from among the many who wish to apply. And since Congress could not do this, it committed the task to the Commission. [p. 215]

Frankfurter's analysis completely overlooks the possibility that, with property rights established, the frequencies could be allotted by bid. See generally, for a devastating criticism of Frankfurter's approach, Coase, R.H., “The Federal Communications Commission,” Journal of Law and Economics, vol. 2 (1959), p. 1.CrossRefGoogle Scholar On the conflict between Hayek and Frankfurter, my sympathies lie with Hayek. See Epstein, R.A., “Beyond the Rule of Law: Civic Virtue and Constitutional Structure,” George Washington Law Review, vol. 56 (1987), p. 149.Google Scholar

⁴ See, e.g., Gilligan, Carol, In a Different Voice: Psychological Theory and Women's Development (Cambridge: Harvard University Press, 1982)Google Scholar as part of a vast literature on the subject.

⁵ See Barash, David P., Sociobiology and Behavior, pp. 181–82 (1977).Google Scholar

⁶ There is here a limiting condition that applies in some cases. Thus, with some kinds of birds there is very little sexual dimorphism between the male and female, while in others the differences are enormous. The cases of convergence between the sexes are those in which the efforts of the male are necessary to ensure the survival of the newborn chicks after birth, which itself is dependent upon the difficulty of acquiring food. The cases in which the female is able to acquire sufficient food for the offspring by herself are those in which the sexual dimorphisms are the greatest. Note that the sexual symmetry in birds is possible in part because once the egg is laid, the male is able to help with warming and feeding as much as the female. There seems to be no situation in the case of birds in which females will adopt higher risk strategies than males. In mammals that condition is never reached, given pregnancy and nursing. See generally, Beehler, Bruce, “The Birds of Paradise,” Scientific American (December, 1989), pp. 116–23.Google Scholar

⁷ See Cohen, Lloyd, “Marriage, Divorce, and Quasi Rents; or, ‘I Gave Him The Best Years of My Life’,” Journal of Legal Studies, vol. 16 (1987), p. 267.CrossRefGoogle Scholar

⁸ See West, Robin, “Jurisprudence and Gender,” University of Chicago Law Review, vol. 55 (1988), p. 1.CrossRefGoogle Scholar

⁹ See, e.g., Markoff, John, “Computing in America: It's a Masculine Mystique,” New York Times, February 13, 1989, p. 7.Google Scholar “Men tend to be more adventurous and riskier,” according to Sydney Springer, a female programmer at Sun Microsystems. It might also be added that they doubtless make more mistakes as well. The key to assessing risk therefore lies in the payoffs, which are apt to be enormous on the high side. The issue is also picked up in Hennig, Margaret & Jardim, Anne, The Managerial Woman (1976), pp. 47–50Google Scholar, 107ff, which notes that women are less comfortable with risk than men, because they tend to treat it only as negative and never as positive. Their explanations for the phenomenon have nothing to do with the biological determinants but stress socialization as an independent variable, which they then analyze in Freudian terms.

¹⁰ The point is emphasized in this volume by Alexander Rosenberg, “Human Nature and Ethics: What Can Biology Tell Us?” His example is that fish, who are unable to distinguish between predators and other fish, flee at the sight of any fish. The costs of this practice include the unnecessary expenditure of energy. But these costs are relatively low. A fish that remains when it should flee is committing a fatal error. Given the inability to develop perfect discrimination between friend and foe, the corner solution – always flee – dominates any intermediate response, notwithstanding its admitted error cost.

¹¹ See, e.g., Fuchs, Victor R., Women's Quest for Economic Equality (Cambridge: Harvard University Press, 1988)Google Scholar, ch. 3, noting that the differences in job patterns are significantly greater between sexes than they are between races. Fuchs rejects “prejudice and exploitation” as an explanation for those differences, and notes the role that biology might play (including differences in “upper body strength,” etc.).

¹² See, e.g., EEOC v. Sears, Roebuck & Co., 839 F.2d 302 (7th Cir. 1988) where Judge Wood, in an excellent opinion, rejected the EEOC's effort to prove that Sears had discriminated in its hiring of its commissioned sales force by showing that a somewhat greater percentage of men had taken these jobs than women. The key witness for Sears for the proposition that some difference in “interest” between men and women might explain why women gravitate toward noncommissioned sales was Professor Rosalind Rosenberg who has in turn been savaged (no lesser word will do) by other feminists. See, e.g., Williams, Joan C., “Deconstructing Gender,” Michigan Law Review, vol. 87 (1989), p. 797.CrossRefGoogle Scholar Rosenberg has been defended by Haskell, Thomas & Levinson, Sanford, “Academic Freedom and Expert Witnessing: Historians and the Sears Case,” Texas Law Review, vol. 66, p. 1629Google Scholar, to which see the reply of Kessler-Harris, Alice, “Academic Freedom and Expert Witnessing: A Response to Haskell and Levinson,” Texas Law Review, vol. 67 (1988), p. 429Google Scholar, and the rejoinder thereto, Haskell, Thomas & Levinson, Sanford, “On Academic Freedom and Hypothetical Pools: A Reply to Alice Kessler-Harris,” Texas Law Review, vol. 67 (1989), p. 1591.Google Scholar The major objection to Kessler-Harris's work is her analysis of sex differences either as nonexistent or as all-or-nothing phenomena. Thus Kessler-Harris writes of Rosenberg's argument: “her argument contended that family and work orientations of women would make them unlikely candidates for commission sales jobs. This position ignored the history and perspectives of the working poor. Women are a diverse group, but research about wage earners demonstrates that women, like men, work primarily for income” [p. 436]. But the relevant question is not whether women are “unlikely candidates for commission sales jobs.” Rather, it is the differential inquiry: whether they are less likely candidates, in part because of preferences about risk. That both men and women work “primarily for income” is again beside the point. The question is: is the preferred level of risk for an income stream identical for men and women? Everything turns on marginal analysis. Small differences in employee taste will lead to disparate patterns of employment without any employer discrimination.

¹³ See, for the difference in approach, Epstein, R.A., “The Utilitarian Foundations of Natural Law,” Harvard Journal of Law & Public Policy, vol. 12 (1989), p. 713Google Scholar, and the criticism thereof in Mack, Eric, “A Costly Road to Natural Law,” p. 753Google Scholar, and my reply, “Postscript: Subjective Utilitarianism,” p. 769.

¹⁴ The utilitarian theories do not always lie in the “objective camp.” See below for a discussion of utilitarian theories that rest upon a more subjective view of what value is and that stresses how difficult value may be to compare across persons. See generally Coleman, Jules, “Efficiency, Utility and Wealth Maximization,” Hofstra Law Review, vol. 8 (1980), p. 509Google Scholar, for a discussion of the connections among the various strands of social theory.

¹⁵ See section V.

¹⁶ See, for further elaboration, Epstein, R.A., “The Utilitarian Foundations of Natural Law,” pp. 713Google Scholar, 745–49, esp. n. 60.

¹⁷ Thus assume a social loss function that measures each person's dissatisfaction by (p_i — x)² wherep p_i is the individual preference and x the collective choice. The squared function is used to represent what seems to be the case, that the further one moves from the median the more vigorous the dissent. Without the squared function, there would be a great deal more freedom in choosing the optimal social choice. With the squared function, the objective should be to get close to the mean, which ordinary voting (dominated on single issues by the median) does not do particularly well when distributions are heavily skewed. The point is seen in the simplest of examples: A has a preference for −2 and B for +2. If the social loss function were linear, any place along the line between −2 and +2 would be as good as any other. But once the squared function is introduced, the only place to minimize the loss is at 0, which is right between them. Thus at either extreme the total loss is 16, at either +1 or −1 it is 10 (3² + 1²), and at 0 it is 8 (2² + 2²).