concurring in part and concurring in the judgment.
I agree with the plurality that even if obscenity on the Internet is defined in terms of local community standards, respondents have not shown that the Child Online Protection Act (COPA) is overbroad solely on the basis of the variation in the standards of different communities. See ante, at 577-579. Like Justice Breyer, however, see post, at 589 (opinion concurring in part and concurring in judgment), I write separately to express my views on the constitutionality and desirability of adopting a national standard for obscenity for regulation of the Internet.
The plurality’s opinion argues that, even under local community standards, the variation between the most and least restrictive communities is not so great with respect to the narrow category of speech covered by COPA as to, alone, render the statute substantially overbroad. See ante, at 577-579. I agree, given respondents’ failure! to provide examples of materials that lack literary, artistic, political, and scientific value for minors, which would nonetheless result in variation among communities judging the other elements of the test. Respondents’ examples of materia;, for which com*587munity standards would vary include such things as the appropriateness of sex education and the desirability of adoption by same-sex couples. Brief for Respondents 43. Material addressing the latter topic, however, seems highly unlikely to be seen to appeal to the prurient interest in any community, and educational material like the former must, on any objective inquiry, see ante, at 579, have scientific value for minors.
But respondents’ failure to prove substantial overbreadth on a facial challenge in this case still leaves open the possibility that the use of local community standards will cause problems for regulation of obscenity on the Internet, for adults as well as children, in future cases. In an as-applied challenge, for instance, individual litigants may still dispute that the standards of a community more restrictive than theirs should apply to them. And in future facial challenges to regulation of obscenity on the Internet, litigants may make a more convincing case for substantial overbreadth. Where adult speech is concerned, for instance, there may in fact be a greater degree of disagreement about what is patently offensive or appeals to the prurient interest.
Nor do I think such future cases can be resolved by application of the approach we took in Hamling v. United States, 418 U. S. 87 (1974), and Sable Communications of Cal., Inc. v. FCC, 492 U. S. 115 (1989). I agree with JUSTICE KENNEDY that, given Internet speakers’ inability to control the geographic location of their audience, expecting them to bear the burden of controlling the recipients of their speech, as we did in Hamling and Sable, may be entirely too much to ask, and would potentially suppress an inordinate amount of expression. See post, at 594-596 (opinion concurring in judgment); contra, ante, at 580-584. For these reasons, adoption of a national standard is necessary in my view for any reasonable regulation of Internet obscenity.
Our precedents do not forbid adoption of a national standard. Local community-based standards originated with *588Miller v. California, 413 U. S. 15 (1973). In that case, we approved jury instructions that based the relevant “community standards” on those of the State of California rather than on the Nation as a whole. In doing so, we held that “[njothing in the First Amendment requires” that a jury consider national standards when determining if something is obscene as a matter of fact. Id., at 31. The First Amendment, we held, did not require that “the people of Maine or Mississippi accept public depiction of conduct found tolerable in Las Vegas, or New York City.” Id., at 32. But we said nothing about the constitutionality of jury instructions that would contemplate a national standard — i. e., requiring that the people who live in all of these places hold themselves to what the nationwide community of adults would find was patently offensive and appealed to the prurient interest.
Later, in Jenkins v. Georgia, 418 U. S. 153, 157 (1974), we confirmed that “Miller approved the use of [instructions based on local standards]; it did not mandate their use.” The instructions we approved in that case charged the jury with applying “community standards” without designating any particular “community.” In holding that a State may define the obscenity standard by stating the Miller standard without further specification, 418 U. S., at 157, Jenkins left open the possibility that jurors would apply any number of standards, including a national standard, in evaluating material’s obscenity.
To be sure, the Court in Miller also stated that a national standard might be “unascertainable,” 413 U. S., at 31, and “[un]realistic,” id., at 32. But where speech on the Internet is concerned, I do not share that skepticism. It is true that our Nation is diverse, but many local communities encompass a similar diversity. For instance, in Miller itself, the jury was instructed to consider the standards of the entire State of California, a large (today, it has a population of greater than 33 million people, see U. S. Bept. of Commerce, Bureau of Census, Statistical Abstract of the United States 23 (120th *589ed. 2000) (Table 20)) and diverse State that includes both Berkeley and Bakersfield. If the Miller Court believed generalizations about the standards of the people of California were possible, and that jurors would be capable of assessing them, it is difficult to believe that similar generalizations are not also possible for the Nation as a whole. Moreover, the existence of the Internet, and its facilitation of national dialogue, has itself made jurors more aware of the views of adults in other parts of the United States. Although jurors asked to evaluate the obscenity of speech based on a national standard will inevitably base their assessments to some extent on their experience of their local communities, I agree with Justice Breyer that the lesser degree of variation that would result is inherent in the jury system and does not necessarily pose a First Amendment problem. See post, at 591. In my view, a national standard is not only constitutionally permissible, but also reasonable.
While I would prefer that the Court resolve the issue before it by explicitly adopting a national standard for defining obscenity on the Internet, given respondents’ failure to demonstrate substantial overbreadth due solely to the variation between local communities, I join Parts I, II, III — B, and IV of Justice Thomas’ opinion and the judgment.