¹ Of course, the lawyer will be less concerned with the objective justice or injustice of those principles than with whether he can formulate rhetorically appealing arguments about justice that can maximize the client's wealth position. Or at least, if the lawyer intends to pursue abstract justice rather than maximization of the client's wealth, he ought to tell the (ex-)client before collecting his fee.

² Legal justification is simply justification within the confines of a given legal system. The principles that will justify a proposition for purposes of legal adjudication can be very different from the principles that will justify a metaphysical truth claim. For example, legal systems can and do, for policy reasons, employ evidentiary rules of admissibility that purport to forbid decision makers to consider much evidence that is metaphysically relevant. See Posner, Richard A., The Problems of Jurisprudence (Cambridge: Harvard University Press, 1990), pp. 205–6.Google Scholar Accordingly, legal justification is always relative to a specific legal system.

³ At least, I am unfamiliar with any such discussions. I would welcome correction.

⁴ In the American legal system, that standard can range from proof by a preponderance of the evidence (the ordinary rule in civil cases) to proof beyond a reasonable doubt (the rule in criminal cases for questions relevant to guilt).

⁵ This discussion draws heavily on a prior essay in which I discuss in more detail the formal structure of justification and its significance for legal analysis. See Lawson, Gary, “Proving the Law,” Northwestern University Law Review, vol. 86, no. 4 (Summer 1992), pp. 859–904.Google Scholar

⁶ Of course, the practical requirements may be very different from the law's formal requirements. In some legal contexts, the best, or even only, way to “prove” a factual proposition may be to get the trier of fact either to ignore the evidence or to consider evidence that the law deems formally inadmissible. My concern here is exclusively with the formal structure of legal justification.

⁷ There is substantial debate within the legal community about how to define the factual propositions that need to be proved in adjudication. The conventional theory identifies a set of distinct prepositional elements that, in conjunction, establish the plaintiff's case. Some scholars have suggested, however, that this conjunction-of-elements approach generates logical absurdities and cannot explain actual legal practice. See generally Allen, Ronald J., “The Nature of Juridical Proof,” Cardozo Law Review, vol. 13, no. 2 (11 1991), pp. 373–422.Google Scholar These scholars suggest that the factual “propositions” that are actually passed upon at trial are comprehensive, integrated stories about events. For my purposes, it does not matter how one defines the relevant factual propositions that must be justified, as the formal structure of justification will be the same in any instance.

⁸ The burden of proof is distinct from the standard of proof. The burden of proof identifies the party who must, as a normative matter, provide the relevant legal authorities with reasons to act or not act. The standard of proof, together with principles of admissibility and significance, defines the extent of that burden. The American legal system typically assigns the burden of proof to the party seeking legal action; a criminal defendant, for example, goes free unless proven guilty. Different legal systems could adopt different default positions.

⁹ See Lawson, , “Proving the Law” (supra note 5), pp. 867–71.Google Scholar

¹⁰ For an illustration of how standards of proof necessarily inform general philosophical discussions of epistemology, see ibid., pp. 871–74 (examining the assumptions about standards of proof implicit in the Cartesian argument for skepticism and some of the responses thereto).

¹¹ Ibid., p. 871.

¹² A wag might suggest at this point that we also need a meta-meta-theory that tells us how to evaluate meta-theories, and so on in an infinite regress. As with all epistemological regresses, the inquiry into foundations must end with criteria that are presupposed by any attempt to question, deny, or justify them (and which are therefore true beyond a rational doubt). In other words, we must ultimately ground all inquiry into property rights in axioms. See Becker, Lawrence C., Property Rights: Philosophic foundations (London: Routlcdge & Kegan Paul, 1977), pp. 45–46.Google Scholar

¹³ In nonlegal settings, inquiries into admissibility and significance tend to merge. The law often deems evidence inadmissible for reasons having nothing to do with metaphysical truth. For the philosopher, however, the only criterion of admissibility is relevance. If evidence has relevance, that means that it is epistemologically entitled to some significance in the decision maker's cognitive process. Nonetheless, one might have reason even in non-legal settings to distinguish whether something counts as evidence for a proposition from how much it counts as evidence. A deontologist, for example, might want to say that consequentialist considerations are simply inadmissible in the proof of moral propositions, even if that is functionally identical to saying that such considerations are admissible but entitled to no significance.

¹⁴ Obviously, we cannot know whether any theory of property rights can meet any specific standard of proof unless we know the admissibility and significance principles that ought to be employed to judge theories. My goal in this essay, however, is solely to explore the standard-of-proof problem and to show that it exists independently of the particular admissibility and significance rules that one selects.

¹⁵ See, e.g., Machan, Tibor R., individuals and Their Rights (La Salle, IL: Open Court, 1989), p. xxxiGoogle Scholar: “The evaluation of a substantive theory must… be comparative.….” But see Becker, , Property Rights (supra note 12), p. 33Google Scholar: “[T]he absence of a valid alternative proves nothing about the validity of the case at hand.”

¹⁶ In reality, no such numerical measure is available, and any comparative assessment of theories must be ordinal. But an example cast wholly in ordinal terms would be much more difficult to construct and grasp. For a discussion of how standards of proof can be expressed in ordinal terms, see Lawson, , “Proving the Law” (supra note 5), pp. 869–70.Google Scholar

¹⁷ See ibid., pp. 898–99.

¹⁸ See Batey, Robert, “Techniques of Strict Construction: The Supreme Court and the Gun Control Act of 1968,” American Journal of Criminal Law, vol. 13, no. 2 (Winter 1986), pp. 133–35.Google Scholar I am profoundly indebted to Professor Batey for bringing these jurisdictions to my attention. It should be noted that Professor Batey saw before I did that the justification of legal propositions, at least in criminal cases, requires specification of a standard of proof.

¹⁹ Pierson v. Post, 3 Cai. R. 175 (N.Y. Sup. Ct. 1805).

²⁰ See Restatement of Property (Philadelphia: American Law Institute, 1944), section 450, comment g, p. 2906.Google Scholar

²¹ An “absolute” standard of proof is simply a standard which relies on criteria other than a relative ranking of the alternative propositions facing the decision maker. Functionally, a standard is absolute if it is possible in principle for all of the available alternatives to fail to meet it.

²² This point is strongly made—and urged as a criticism of the use of absolute proof standards for legal propositions—by Larry Alexander in an essay responding to my earlier, more general treatment of standards of proof in the law. See Alexander, Larry, “Proving the Law: Not Proven,” Northwestern University Law Review, vol. 86, no. 4 (Summer 1992), pp. 907–11.Google Scholar

²³ See Epstein, Richard A., “Property and Necessity,” Harvard Journal of Law and Public Policy, vol. 13, no. 1 (Winter 1990), p. 3.Google Scholar

²⁴ For a comprehensive discussion of this general rule of the relativity of title, and the exceptions thereto, see Donahue, Charles Jr., Kauper, Thomas E., and Martin, Peter W., Cases and Materials on Property: An Introduction to the Concept and the Institution, 2nd ed. (St. Paul, MN: West Publishing Co., 1983), pp. 56–108.Google Scholar

²⁵ One can argue at length about when a legal alternative is “available” to a court in a particular case. Suppose, in our example, that the owner of the land, who we know with the requisite degree of certainty is the true owner of the fox, is not a formal party to the lawsuit but is known to the parties and the court. That is, we not only know that there is a true owner whose claim is superior to the parties' claims, but we know the identity of that true owner. Is such knowledge alone sufficient to defeat the claims of all of the actual parties? Some courts would say yes. See, e.g., Barwick v. Barwick, 33 N.C. 80, 81–83 (1850)Google Scholar; Annotation, “Mere Possession in Plaintiff as Basis of Action for Wrongfully Taking or Damaging Personal Property,” American Law Reports Annotated, vol. 150 (1944), pp. 186–89.Google Scholar

²⁶ One could argue that in these circumstances it is a mistake to discuss the issue in terms of ownership, as the law would merely protect possession as an independent interest. That is no doubt true, but it does not alter the analysis. The law still must determine which acts and events constitute possession, and that determination is subject to the same standard-of-proof considerations as are determinations regarding ownership in any larger sense.

²⁷ Or conceivably the law could allow Post, Peters, and Petrov to pool their claims and split the proceeds. It is clear, after all, that as a group their joint claim is superior to Pierson's.

²⁸ This position, however, requires not merely an aversion to state action but an aversion to organized legal action of any kind. Even an anarchist who believes that law should be privately administered has an interest in the law's dispute-resolving capacity.