¹ Rawls, John, A Theory of Justice (Cambridge: Harvard University Press, 1971), p. 100.Google Scholar The role of luck in Rawls's thinking is also noted by Cohen, G.A.: “They [i.e. Rawls and Dworkin] say that, because it is a matter of brute luck that people have the talents they do, their talents do not, morally speaking, belong to them, but are, properly regarded, resources over which society as a whole may legitimately dispose.”Google ScholarCohen, G.A., “Self-Ownership, World Ownership, and Equality: Part II,” Social Philosophy and Policy, vol. 3, no. 2 (Spring 1986), p. 79.CrossRefGoogle Scholar

² For a general discussion, see Epstein, R., “Possession as the Root of Title,” Georgia Law Review, vol. 13 (Summer 1979), p. 1221.Google Scholar

³ The actual operation of insurance is far more complex. Life insurance, for example, cannot equalize a person's wealth after death, and may not be desired by persons who have no dependents or heirs. Even disability insurance may not be desired by persons who think that they get very little value out of money when sick or injured. But these examples only show that complete insurance need not be desired. They are not inconsistent with the proposition that insurance, when purchased, tends to equalize wealth across different states of the world. See generally, Epstein, R.A., “Products Liability as an Insurance Market,” Journal of Legal Studies, vol. 14 (December 1985), p. 645CrossRefGoogle Scholar; Schwartz, Alan, “Proposals for Product Liability Reform: A Theoretical Synthesis,” Yale Law Journal, vol. 97 (Feb. 1988), pp. 353, 362–367, and the materials cited in note 13, p. 362.CrossRefGoogle Scholar

⁴ Again, the statement is somewhat incomplete. Some insurance is purchased in order to bond the loss prevention services that the insurer provides the insured. Thus, the vast bulk of boiler insurance goes for inspection, not payouts for losses.

⁵ See Lawson, F.H., Negligence in the Civil Law (Oxford: Clarendon Press, 1950), p. 14Google Scholar: “We know that as late as the classical period and, so far as appearances go, in the time of Justinian an action could be brought on the lex [Aquilia] itself only if the death or injury resulted from direct contact between the body of the wrongdoer and thing (corpore corpori). Translated into the language of the English law, this means that the lex penalized only trespasses.” The reference to the “lex itself” notes that analogous actions, not explicitly under the language of the statute (here occidere, to kill, literally by force) were allowed for those who furnished a cause of death, e.g., supplied a victim with poison.

⁶ See, e.g., Beven, T., Negligence in Law, (London: Stevens & Haynes, 3rd edition, 1908), p. 45.Google Scholar

⁷ The literature on this debate is legion. See, e.g., Epstein, R.A., “A Theory of Strict Liability,” Journal of Legal Studies, vol. 2 (January 1973), p. 223Google Scholar; Posner, R.A., “A Theory of Negligence,” Journal of Legal Studies, vol. 1 (January 1972), p. 29CrossRefGoogle Scholar; Shavell, S., “Strict Liability versus Negligence,” Journal of Legal Studies, vol. 9 (January 1980), p. 1.CrossRefGoogle Scholar See also, Symposium on Causation in the Law of Torts, with articles by Coleman, Jules, Cooter, Robert, Epstein, Richard, Kelman, Mark, Thompson, Judith, and Weinrib, Ernest, Chicago-Kent Law Review, vol. 63 (1987), p. 397.Google Scholar

⁸ The most extreme illustration of the principle is a variation of the famous case of Summers v. Tice, 33 Cal.2d 80, 199 P.2d 1 (1948), where two defendants shot at a plaintiff in likelihood that each person's bullet hit. The court there held that the loss was joint and several, so that each defendant was liable for the whole, but had an action against the other for SO percent contribution, if he were solvent. If it could have been shown that defendant A had a 51 percent chance of having fired the bullet, his would have been the total loss, notwithstanding the equal culpability of the two defendants. The same result would hold if the defendant who shot the plaintiff was guilty only of ordinary negligence, while the one who missed had engaged in reckless conduct.

⁹ There was some liability for omissions, but here too only with respect to breach of some special relationship that linked plaintiff to defendant. See, e.g., Restatement Second of Torts, §323. Kline v. 1500 Massachusetts Avenue Apartment Corp., 439 F.2d 477 (D.C. Cir. 1970).

¹⁰ Blyth v. Birmingham Water Works, 11 Exch. 781, 156 Eng. Rep. 1047 (1856).

¹¹ Note that it is possible to adopt an intermediate position which recognizes the specific defense of compulsion by acts of God, without committing one's self to the more general principle of no liability without negligence. The compulsion cases are the easiest to identify as situations in which a person's will is overborne by external events. Allowing this defense does not open up all stranger cases to the complex cost-benefit calculations that many rules of negligence invite.

¹² The recent “wrongful life” cases are not an exception to this general rule. These cases fall into two categories. In the former, parents bring suits against a physician for having a baby at all. The grounds of the action are quite traditional in that it is alleged typically that the physician was careless in an operation designed to sterilize either the father or mother, so that the costs of having the baby were a consequence of the breach of duty. These suits have met with mixed success with normal babies, but have generated some large rewards where the child born suffers from serious birth defects, imposing extraordinary costs of care on the parents. The more exotic class of wrongful birth cases is brought by the defective children themselves. Here courts have been reluctant to reward damages to the injured party because they cannot quite figure out the baseline for compensation when the child claims that he was better off never having been born. The first case in the wrongful birth line was Gleitman v. Cosgrove, 49 NJ. 22, 227 A.2d 689 (1967). See generally, Prosser and Keeton on the Law of Torts (W.P., Keeton gen. ed., 5th ed. 1984), p. 370.Google Scholar

¹³ Aristotle, , The Nicomachean Ethics, Bk. 4, Ch. 4, trans. Rackham, H. (Cambridge: Harvard University Press, 1926), p. 275.Google Scholar

¹⁴ Rawls, , A Theory of Justice, pp. 10–11.Google Scholar

¹⁵ There can be an extensive debate over whether such philosophers as Locke and Nozick follow this pattern of common law rights. Locke, for example, appears to allow some limited place for redistribution, at least in cases of “extreme want,” albeit one that falls short of a robust commitment to some guaranteed level of social support. I discuss some of these issues in Epstein, R.A., “Taxation in a Lockean World,” Social Philosophy & Policy, vol. 4, no. 1 (Autumn 1986), p. 49, esp. p. 69, n. 36.CrossRefGoogle Scholar

¹⁶ See, on this point, Hoffman, E. & Spitzer, M.L., “Entitlements, Rights, and Fairness: An Experimental Examination of Subjects' Concepts of Distributive Justice,” Journal of Legal Studies, vol. 14 (June 1985), p. 259.CrossRefGoogle Scholar Their experiments show that there is at least some widespread sentiment for the sharing of windfall gains (at least as the Lockean would define them). The finding is of itself great importance, for it helps explain why programs of redistribution have such broad appeal. But the finding accounts for redistribution by voluntary means as well as redistribution by coercion. It helps explain why both these means might be tried, but does not afford a ground to choose between them.

¹⁷ See, e.g., G.A. Cohen's example of a world composed only of Able and Infirm, with joint ownership and external goods and individual ownership of talents. Cohen, G.A., “Self-Ownership,” pp. 84–87.Google Scholar Cohen's point is that this ownership structure places very powerful restraints on Able's ability to deal with his own talents, and thereby reveals the awkwardness of having separate systems of ownership for different forms of resources. But once it is granted that a system of property rights has to work in a world with large numbers of persons, then the simpler common law system, whereby external things in the original position are regarded as unowned, has the virtue of eliminating the enormous collective action problems that would otherwise arise.

¹⁸ See Kitch, E., “The Nature and Function of the Patent System,” Journal of Law and Economics, vol. 20 (October 1977), p. 265.CrossRefGoogle Scholar Kitch's point is that the system of patent law extends its protection to many sketchy inventions in order to encourage further labor by the patentee. Note that recognizing ownership by first possession improves the likelihood that labor will be expended on external resources, so that in the end the value of most resources will be in fact enhanced by the labor so often thought to justify private ownership.

¹⁹ This last point is a problem for Jules Coleman, who wishes to separate, at least analytically, the obligation to pay for wrongdoers from the right to collect for victims. See Coleman, Jules L., “Mental Abnormality, Personal Responsibility, and Tort Liability,” Brody, B.A. and Tristram, Engelhardt H. Jr., eds., Mental Illness: Law and Public Policy (Boston: Kluwer, 1980), p. 107.CrossRefGoogle Scholar See also Coleman, Jules L., “Property, Wrongfulness and the Duty to Compensate,” Chicago-Kent Law Review, vol. 63 (1988), pp. 460–61Google Scholar (“the central claim I make is that liability and recovery are conceptually and normatively distinguishable”).

²⁰ This last point is important because it helps explain why the insurance companies have some “give” in their classification system. Overcharged individuals have to pay a price to find coverage elsewhere, or to go it alone. And this will keep them in place, at least for some limited time.

²¹ Most, but not all. One element of loss that is not socialized is the insured's time. The standard insurance policy requires the tort defendant to cooperate in the defense with the company, at his own expense. These costs provide a useful break against the greater willingness to engage in risky activities that arises once insurance is in place.

²² I do, sort of. That is, I think that it is possible for individuals to make these comparisons in their ordinary lives, which is why marriages and friendships last. But it is quite a different question whether they can be made with similar accuracy in the political context where information is hard to come by, and the power of partisan zeal great.

²³ See Friedman, D.D., “Does Altruism Produce Efficient Outcomes? Marshall versus Kaldor,” Journal of Legal Studies, vol. 17 Qanuary 1988), p. 1.CrossRefGoogle Scholar

²⁴ See Schmalz, R.A., “On the Financing of Compensation Systems,” Journal of Legal Studies, vol. 14 (December 1985), p. 807.CrossRefGoogle Scholar

²⁵ See, for discussion, R.A. Epstein, “Taxation in a Lockean World.”

²⁶ I have discussed these at greater length in Epstein, Richard A., Takings: Private Property and the Power of Eminent Domain (Cambridge: Harvard University Press, 1985), pp. 314–324Google Scholar; and “The Uncertain Quest for Welfare Rights,” Brigham Young Law Review (1985), p. 201, reprinted in G.C., Bryner & N.B., Reynolds, eds., Constitutionalism and Rights (Provo: Brigham Young University, 1987), p. 33.Google Scholar

²⁷ See for discussion, Proceedings of the Conference on Takings of Properly and the Constitution, 41 Miami Latv Review 49, 79–82 (1986).