meanings may sometimes "evolve" with our changing society. Even if the latter is true in some instances, it seems problematic to make that evolution turn upon the morals and values of other societies. Presumably, we decide to adopt a view of a U.S. law different from its original meaning because we feel that changes in our own society make the original rule no longer appropriate. It would seem odd, therefore, to say that, although American society has not changed in a way that would require an evolving interpretation of a U.S. law, that foreign societies have done so.11 To return to the First Amendment context, we may feel confident that strong protections of anti-government speech are contained in the intent of the Amendment itself, and that U.S. society has not evolved in a way to bring them into question; yet we might also note that many countries around the world have more restrictive limits on anti-government speech. 12 It is hard to see how the latter evidence would justify a departure from an interpretation of the First Amendment that is consistent with both its original meaning and with modern American values. Nonetheless, this is what some recent Supreme Court cases, and some academic commentary, seem to be suggesting.

SPECIFIC EXAMPLES OF SUPREME COURT PRACTICE

I now turn to specific evaluations of two recent Supreme Court cases that have excited much attention for their use of foreign materials: Atkins v. Virginia, concerning the constitutionality of executing mentally handicapped defendants, and Lawrence v. Texas, concerning the constitutionality of criminalizing homosexual sodomy.13 In each case the Court found the challenged law unconstitutional, and relied in part upon evidence of foreign practices. In each case several Justices registered strong objections to the use of such materials. And in each case some of the briefs made extensive use of foreign materials, urging an even greater reliance upon them. 14

In Atkins, the Court relied in part upon the opinion of the “world community" that mentally handicapped defendants should be exempt from the death penalty, in deciding that executing the mentally handicapped violated the Eighth Amendment's ban on "cruel and unusual punishment." 15 As I have described elsewhere, there are serious methodological problems with how the Court determined the "opinion of the world community"-including the fact that the court did not cite any foreign judgments, but only the amicus briefs of one of the parties, which were in turn either misleading or inaccurate in important respects. 16 But leaving this aside, the relevant question here is, assuming that in general most nations do not execute the mentally handicapped, whether that should be relevant to the meaning of the Eighth Amendment.

The Court made no attempt to show why foreign practice should be relevant (the citation was in an footnote, made almost as an aside). There is no legal text parallel to the Eighth Amendment that has been interpreted in a foreign country in any way that is helpful to discerning the original meaning of the Eighth Amendment. Even if most foreign countries disapprove such executions, they do not do so as a result of an interpretation of the language of the Eighth Amendment, or anything upon which the Eighth Amendment was based. The Court's interpretation of the Eighth Amendment in Atkins did not turn on facts or predictions about effects that could be influenced by practice in foreign countries. 17 And third, the Court did not show

11 For the foreign materials to have any relevance to the decision beyond mere window-dressing, we must posit a situation in which the court's evaluation of the values of American society (however those may be determined) lead to a different result from its evaluation of foreign materials. Otherwise, the foreign materials are not truly a factor in the decision.

12 For key European decisions on free speech that may be less protective than U.S. law, see, e.g., Zana v. Turkey, 27 E.H.R.R. 667 (1997); Observer and Guardian v. United Kingdom, 14 E.H.R.R. 153 (1991); Barfod v. Denmark, 13 E.H.R.R. 493 (1998).

13 Atkins v. Virginia, 536 U.S. 304 (2002); Lawrence v. Texas, No. 02-102 (June 26, 2003). 14 To be clear, in the subsequent discussion I am not taking any position on the correct outcome of either case-only upon the type of evidence that should and should not have influenced the outcome.

15 Atkins, 536 U.S. at 316 n. 21 (“Moreover, within the world community, the imposition of the death penalty for crimes committed by mentally retarded offenders is overwhelmingly disapproved.").

16 Michael D. Ramsey, International Materials and Domestic Rights: Reflections on Atkins and Lawrence, Amer. J. Int'l L. (forthcoming 2004).

17 Justice Scalia in dissent suggested that a categorical rule against executing the mentally handicapped was a bad one because of the dangers of undetectable faking. Assuming that this should be relevant to the outcome, this is something that could be tested empirically by examining the experiences of jurisdictions that have a categorical rule.

that the Eighth Amendment itself, in its original understanding, depended upon the scope of punishments in foreign countries.

On the third point, it is of course possible that the drafters of the Eighth Amendment intended that its scope be affected by the severity of punishments in foreign countries, but I think that unlikely. For example, suppose a certain punishment was thought repugnant by Americans at the time the Amendment was adopted, and continues to be thought repugnant by most Americans today, but the punishment has been widely adopted throughout the world. Would that justify allowing the punishment in the few American jurisdictions that sought to adopt it? I think not, because the founding generation in America in many cases (including, I would say, in the Eighth Amendment) defined their values in opposition to what was practiced in much of the world. Most jurisdictions in the Framers' day did not protect their citizens from brutal punishments; the point of the Eighth Amendment was to establish a uniquely American standard. But if the practices of the world do not permit us to diminish the protections of the Eighth Amendment, they also should not permit us to enlarge its protections. In any event, there is no evidence that the Framers expected or condoned such an approach.

Instead, what the Court seemed to be saying in Atkins is that other jurisdictions' decisions not to execute the mentally handicapped (whether for moral, constitutional, practical or other reasons) should influence our decision whether to permit such executions in the United States. As a matter of social policy, I agree with that proposition: we should surely consider (though not feel bound by) other nations' approaches to similar social problems (just as, in our federal system, individual states should consider, though not feel bound by, approaches to similar social problems by other states). Thus the Congress, and individual state legislatures should consider foreign practices in deciding whether there should be a categorical rule against executing the mentally handicapped.

However, it is not the role of the Supreme Court to set U.S. social policy, with respect to executions or otherwise: the Court's role, in the Atkins case, was to interpret the Eighth Amendment. That means that the Court should base its decision upon the original meaning of the Eighth Amendment, or (perhaps) upon an evolving meaning that resonates with modern American values. În any event, its decision should turn upon the interpretation of the legal text. Congress, and the state legislatures, are the appropriate bodies to determine social policy (and thus to consider the relevance of social policies of foreign jurisdictions).

The Court's decision in Lawrence shows some similar problems. The issue there was whether a state law criminalizing homosexual sodomy violated the Due Process Clause of the Fourteenth Amendment. According to prior precedent, the question should have been decided by asking (a) whether homosexual sodomy was a fundamental right, and (b) if not, whether the state had a rational basis in banning it. 18 Since the Court did not appear to find a fundamental right, the rationality of the state law was the central constitutional question. That issue had already been decided by the Court in its prior decision in Bowers v. Hardwick,19 but the Court in Lawrence decided that Bowers should be overruled on this point.

In addressing this question, the Court discussed several decisions of the European Court of Human Rights (ECHR), and referred to an amicus brief that described the law in some foreign countries.20 There are two ways to view this approach, one of which is much more limited and defensible than the other. First, the state in Lawrence (and to some extent the Court's prior discussion in Bowers) relied in part upon a claim that bans on homosexual sodomy were pervasive in Western civilization. To the extent that such a claim is relevant, it seems appropriate to look at foreign jurisdictions to show that this claim is not true. That is, the actual practice of foreign nations can be used to refute arguments based upon unfounded claims about supposed foreign practice. Though this defensive use of foreign materials by the Court does not seem too objectionable, I would prefer if the Court had simply rejected the state's claims as irrelevant. The fact (if it is a fact) that many nations currently ban homosexual sodomy does not show that such bans are rational, or otherwise inform the original meaning or modern meaning of the U.S. Due Process Clause.

Another way of looking at Lawrence, however, is that the Court used foreign practice as an affirmative argument in favor of striking down the statute. That is, it

20 "[I]t should be noted that the reasoning and holding of Bowers have been rejected elsewhere [citing three decisions of the ECHR]. Other nations, too, have taken action consistent with an affirmation of the protected right of homosexual adults to engage in intimate, consensual conduct. [citing an amicus brief]. The right petitioners seek in this case has been accepted as an integral part of human freedom in many other countries." Lawrence, slip op. at 16.

thought that because other jurisdictions had de-criminalized homosexual sodomy, the U.S. should do so as well. This resembles the Court's claim in Atkins, and is similarly problematic because it is a statement of social policy rather than an interpretation of a legal text.

The Court's citation of the ECHR (and especially its claim that the ECHR had "rejected" the "reasoning and holding in Bowers") suggests that constitutional courts are all engaged in a common interpretive enterprise (as in fact they are when they are interpreting a common legal text such as the Warsaw Convention). But as a matter of legal interpretation, there is no direct connection between the U.S. Constitution and foreign court opinions that address the interpretation of different documents written in different times and different countries. The mandate of the ECHR, for example, is to interpret the European Convention for the Protection of Human Rights and Fundamental Freedoms, a treaty among European nations drafted in the 1950s. Under the Convention, the question is whether sodomy laws violate the right (in Article 8(1)) to "privacy and family life" and are not justified under Article 8(2) (restrictions that are "necessary" to protect listed social values). Under the U.S. Constitution, as discussed, the question is whether the right is "fundamental" and, if not, whether the law is rationally related to a legitimate governments interest. Thus in confronting sodomy laws the ECHR and the U.S. Supreme Court faced entirely distinct texts, with a distinct body of precedent elaborating upon the meaning of key phrases. It is too simplistic to say that both are doing constitutional law, and so doing the same thing. Rather, they are both interpreting texts, but the texts they are interpreting are distinct.

Dudgeon v. United Kingdom, the leading European case cited in Lawrence, confirms this point. According to Dudgeon, the principal question it faced was whether the sodomy law was "necessary for the protection of health or morals" (the quoted language being the text of Article 8(2) of the Convention). The ECHR emphasized that in this context "necessary" meant a "pressing social need" or a "particularly serious reason" and not merely "reasonable." 21 In the U.S. case, in contrast, assuming that the Lawrence Court was following its own precedents in other respects, the Court was asking not whether sodomy laws were "necessary" but whether they were reasonable-that is, exactly the question Dudgeon said it was not asking.

The question, then, is how the conclusions of a European Court, interpreting a legal document totally distinct in language and context from the U.S. Constitution, could have implications for the correct interpretation of the U.S. Constitution. In a strictly legal sense, the answer should be that they do not, because the two courts are engaged in a distinct legal enterprise. Contrary to the observations of one U.S. Supreme Court Justice, there is no such thing as a "global legal enterprise in constitutional law,' ,"22 because there is no single global constitution. There is broad commonality among constitutional courts only if one thinks that the courts are not really interpreting texts, but deciding whether sodomy laws are justifiable as a matter of moral and social policy.

As in Atkins, under our constitutional system legislatures not courts should make decisions regarding matters of moral and social policy. It is appropriate for legislatures to consider the moral and social policy decisions of foreign jurisdictions with respect to anti-sodomy laws to guide their own moral and social decisionmaking on that issue. Courts, on the other hand, make (or should make) decisions concerning interpretation of specific legal texts. It is appropriate for courts to consider the interpretive decisions of foreign jurisdictions to guide their own interpretive decisions on the same legal texts. As the foregoing discussion illustrates, in relying on the Dudgeon case in Lawrence, the Supreme Court was not looking to the European court for interpretative guidance as to the meaning of a legal text, but was looking to the European court for guidance as to moral and social policy.

In sum, in both Lawrence and Atkins the Supreme Court did not appear to be looking to foreign materials to aid in legal interpretation of the text of the U.S. Constitution, but rather it looked to foreign materials to provide what Professor Gerald Neuman has called "normative insight." But it is contrary to the constitutional role of courts for courts (rather than legislatures) to be making moral and social policy in this way. Courts should decide what a text means, not what the best moral and social outcome should be. The meaning of a text that forms part of U.S. law is not affected by what other jurisdictions have decided about matters of moral and social policy, or by what other courts have decided about the meaning of different legal texts.

21 Dudgeon v. United Kingdom, 45 E.C.H.R., para. 49-52 (1981).

22 Justice Breyer, quoted in Roger Alford, Misusing International Sources to Interpret the Constitution, Am. J. Int'l L. (forthcoming 2004).

PRINCIPLED ADJUDICATION AND THE DANGER OF USING FOREIGN MATERIALS

While realists may say that courts routinely make decisions of moral and social policy, there are particular dangers of U.S. courts relying (or purporting to rely) upon foreign materials in this process. As part of our constitutional system, we expect courts to make decisions on the basis of neutral, generally applicable legal principles.23 If U.S. courts adopt a practice of relying on foreign materials, we would expect that foreign materials be treated as authoritative guides as a general matter, not merely in cases in which the foreign materials happen to support moral and policy intuitions arising from other sources. But this principle leads to one of two outcomes, each unsatisfactory.

First, courts might in fact treat foreign materials as authoritative across the board. The result, though, would likely be a lessening of U.S. rights. The recent push for foreign materials has come most strongly from rights advocates, and in Lawrence and Atkins the United States lagged at least parts of the world, and parts of world opinion, in guaranteeing the rights at issue. But there is nothing necessarily rights-enhancing about foreign materials. In many areas, it seems likely that the United States is an outlier in protecting rights that few other societies recognize such as the First Amendment. As I have suggested, freedom of speech is one important example. Another is freedom of religion: many countries have much greater establishment of religion (as in Europe, where many countries have an established church or explicitly "Christian" parties); at the same time, many countries have lesser protections for the free exercise of religion (as the controversy in France over headscarves and other religious headgear suggests).24

Beyond the First Amendment, it seems clear that many foreign nations lack the rights, for example, to bear arms and own property guaranteed in the U.S. Constitution-indeed, as with many of our constitutional provisions, the framers' intent was to guarantee rights that were not traditionally recognized elsewhere. In addition, the United States has elaborate procedural protections for criminal defendants, as a matter of the Court's interpretation of open-ended constitutional clauses such as "unreasonable" search and "due" process, that likely go far beyond those existing in most foreign nations. For example, it appears that the "exclusionary rule" of the Fourth Amendment, which excludes from trial evidence obtained in unconstitutional searches, has few counterparts worldwide. 25 Should each of these rights be re-evaluated to see if they are generally recognized by foreign nations, and abandoned if they are not? If we are serious about the project of using foreign materials, we must "take the bitter with the sweet" and use foreign materials to contradict, not merely to confirm, our own view of rights.

I doubt, though, that there is the moral and political will to apply foreign materials in this way. More likely, then, is the selective use of foreign materials to support judgments reached for other reasons. One can already see this developing in Supreme Court advocacy and jurisprudence. First, there is selective citation to countries whose practices happen to support a particular result, but not to those that contradict it. In Lawrence, for example, the Court discussed some jurisdictions that had overturned or repealed their sodomy laws, but did not discuss anything close to a general practice of nations. Though I have not made systematic inquiries, it seems likely that quite a number of foreign jurisdictions criminalize sodomy. This went unmentioned in Lawrence. In Atkins, the Court claimed, without adequate support, that "world opinion" opposed execution of the mentally handicapped. In fact, it appears that many leading death penalty jurisdictions do not make such a categorical exception, and that opposition comes mostly from countries and scholars that oppose the death penalty across the board.

Of course, one might say that some countries are better moral models than others. Should it matter, for example, that Chinese law apparently permits the execution of the mentally handicapped? But attempting to articulate a legal principle justifying this sort of selectivity, if done explicitly, leads courts into another unsatisfactory choice. Presumably we do not want attorneys arguing, and the Supreme Court

23 See Herbert Wechsler, Toward Neutral Principles of Constitutional Law, 73 Harv. L. Rev. 1 (1959). As the Court's plurality put it in Planned Parenthood of Pennsylvania v. Casey, the Court's legitimacy arises from it "making legally principled decisions under circumstances in which their principled character is sufficiently plausible to be accepted by the Nation." 505 U.S. 833, 866 (1992).

24 See Christopher Marquis, U.S. Chides France on Effort to Bar Religious Garb in Schools, N.Y. Times, Dec. 19, 2003, at A8.

25 See Erik J. Luna & Douglas Sylvester, Beyond Breard, 17 Berkeley J. Int'l. L. 147, 177– 79 (1999) ("Legal rules suppressing relevant probative evidence from criminal trials are few and far between outside the United States.").

deciding, which of (say) Japan, Thailand, Pakistan, China, etc., are sufficiently "civilized" to serve as moral precedents.

A further selection problem is that the Supreme Court has invoked foreign materials only in some cases, and not others. As Professor Roger Alford has pointed out, the Court's recent decision invalidating the previous federal law against late-term abortions under the due process clause, Stenberg v. Carhart, made no reference to foreign materials.26 Yet it seems likely that foreign jurisdictions have grappled with this issue. Moreover, it seems at least possible that the weight of foreign practice (which generally does not embrace abortion rights as fully as U.S. jurisprudence) does not permit late-term abortions. It is hard to square Stenberg's disregard for foreign practice with Lawrence, which involved the same clause of the U.S. Constitution, other than on the ground that in Lawrence the Court approved of the foreign practice and in Stenberg it did not.

This selectivity confirms that courts are not really being guided by foreign materials in their readings of specific texts, but are using foreign materials to support decisions of moral and social policy reached on other grounds.27 And this further confirms that considering foreign practice as a guide to moral and social policy decisionmaking is properly a legislative, not a judicial, function. Legislatures acknowledge that their decisions are policymaking that is not based on interpretive principles. Thus they are free to consider the views and practices of foreign jurisdictions, adopting what they like and discarding what they do not like, for policy reasons without the need to justify their decisions in judicial terms. When courts behave in this way (as it seems inevitable that they will in dealing with foreign materials), the rule of law and the role of courts is undermined.

CONCLUSION

In Lawrence and Atkins, the use of foreign materials, while open to serious question, probably did not affect the ultimate outcome of either case. To see the potential scope of the use of foreign materials, it may be useful to consider recent comments by Professor Harold Koh of Yale Law School. In an article published in the U.C. Davis Law Review, Professor Koh urged that human rights advocates use foreign materials to persuade the Supreme Court to abolish the death penalty.28

It seems plain that the Framers did not intend to exclude the death penalty through the Eighth Amendment. It also seems plain that the death penalty, in appropriate circumstances, is consistent with modern American social values, based on the broad acceptance of the death penalty in the United States. But Professor Koh is correct that many countries, particularly in Europe, reject the death penalty as a matter of moral and social policy.29 The question is whether and how we should take that into account.

As indicated above, I think it appropriate for Americans to consider Europe's abolition of the death penalty in deciding whether we should retain it. The key, though, is that the legislatures (and the people, acting through their legislatures) should consider it, not the courts. The courts' role is limited to deciding whether the death penalty is consistent with the meaning of the Eighth Amendment-either its original meaning, or, in some versions, its "evolving" meaning as informed by the evolving values of American society. Europe's current view of the death penalty as a matter of moral and social policy does not inform the original meaning of the Eighth Amendment nor the values of modern American society, and so should not figure in the courts' view of the Eighth Amendment. Professor Koh's suggestion that we give consideration to Europe's views is correct, but addressed to the wrong forum. The decision whether or not to change American moral and social policy to abolish

26 Stenberg v. Carhart, 530 U.S. 914 (2000); see Roger Alford, Misusing International Sources to Interpret the Constitution, Am. J. Int'l L. (forthcoming 2004).

27 See Diane Marie Amann, Raise the Flag and Let it Talk: On the Use of External Norms in Constitutional Decisionmaking, 2 Int'l J. Const. L. (forthcoming 2004). Professor Amann predicts, as I do, that courts will likely behave in this way, adopting "external norms" (i.e., foreign views of moral and social policy) that they like and discarding those they do not like, in an essentially legislative fashion. We differ on whether this is appropriate.

28 Harold Hongju Koh, Paying "Decent Respect" to World Opinion on the Death Penalty, 35 U.C. Davis L. Rev. 1085 (2002).

29 It is worth noting that Professor Koh's use of foreign materials is-like the Court's-selective. The death penalty has not been abolished in all countries, including in liberal democracies such as Japan. Moreover, polls suggest that in Europe the death penalty is much more popular among the ordinary population that among elites. See Crime Uncovered, The Observer, April 27, 2003 (reporting poll showing 67% in Britain support re-introduction of the death penalty). Moreover, I doubt Professor Koh would endorse using foreign materials to guide courts' decisionmaking on abortion or criminal procedure matters where the United States is more protective of rights than other nations.