90
O'Connor, J., dissenting
Clause, even when some statistical support can be conjured up for the generalization"); Craig, 429 U. S., at 201 (invali-dating a sex-based classification even though the evidence supporting the distinction was "not trivial in a statistical sense"); id., at 202 (noting that "prior cases have consistently rejected the use of sex as a decisionmaking factor even though the statutes in question certainly rested on far more predictive empirical relationships than this"); Wiesenfeld, 420 U. S., at 645 (invalidating a sex-based classification even though the underlying generalization was "not entirely without empirical support"). Indeed, the stereotypes that underlie a sex-based classification "may hold true for many, even most, individuals." Miller, 523 U. S., at 460 (Ginsburg, J., dissenting). But in numerous cases where a measure of truth has inhered in the generalization, "the Court has rejected official actions that classify unnecessarily and overbroadly by gender when more accurate and impartial functional lines can be drawn." Ibid.
Nor do stereotypes consist only of those overbroad generalizations that the reviewing court considers to "show disrespect" for a class, ante, at 73. Cf., e. g., Craig, supra, at 198-201. The hallmark of a stereotypical sex-based classification under this Court's precedents is not whether the classification is insulting, but whether it "relie[s] upon the simplistic, outdated assumption that gender could be used as a 'proxy for other, more germane bases of classification.' " Mississippi Univ. for Women, supra, at 726 (quoting Craig, supra, at 198).
It is also important to note that, while our explanations of many decisions invalidating sex-based classifications have pointed to the problems of "stereotypes" and "overbroad generalizations," these explanations certainly do not mean that the burden is on the challenger of the classification to prove legislative reliance on such generalizations. Indeed, an arbitrary distinction between the sexes may rely on no identifiable generalization at all but may simply be a de-
Page: Index Previous 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 NextLast modified: October 4, 2007