"Though the tiers of scrutiny have become a ubiquitous feature of constitutional law, they are of recent vintage. Only in the 1960’s did the Court begin in earnest to speak of “strict scrutiny” versus reviewing legislation for mere rationality, and to develop the contours of these tests. See Fallon, Strict Judicial Scrutiny, 54 UCLA L. Rev. 1267, 1274, 1284–1285 (2007). In short order, the Court adopted strict scrutiny as the standard for reviewing everything from race-based classifications under the Equal Protection Clause to restrictions on constitutionally protected speech. Id., at 1275–1283. Roe v. Wade, 410 U. S. 113 , then applied strict scrutiny to a purportedly “fundamental” substantive due process right for the first time. Id., at 162–164; see Fallon, supra, at 1283; accord, Casey, supra, at 871 (plurality opinion) (noting that post-Roe cases interpreted Roe to demand “strict scrutiny”). Then the tiers of scrutiny proliferated into ever more gradations. See, e.g., Craig, 429 U. S., at 197–198 (intermediate scrutiny for sex-based classifications); Lawrence v. Texas, 539 U. S. 558, 580 (2003) (O’Connor, J., concurring in judgment) (“a more searching form of rational basis review” applies to laws reflecting “a desire to harm a politically unpopular group”); Buckley v. Valeo, 424 U. S. 1, 25 (1976) (per cu-riam) (applying “ ‘closest scrutiny’ ” to campaign-finance contribution limits). Casey’s undue-burden test added yet another right-specific test on the spectrum between rational-basis and strict-scrutiny review. The illegitimacy of using “made-up tests” to “displace longstanding national traditions as the primary determinant of what the Constitution means” has long been apparent. United States v. Virginia, 518 U. S. 515, 570 (1996) (Scalia, J., dissenting). The Constitution does not prescribe tiers of scrutiny. The three basic tiers—“rational basis,” intermediate, and strict scrutiny—“are no more scientific than their names suggest, and a further element of randomness is added by the fact that it is largely up to us which test will be applied in each case.” Id., at 567; see also Craig, supra, at 217–221 (Rehnquist, J., dissenting)."