Justice Kennedy, dissenting.

As the majority recognizes, the touchstone for interpreting the Assimilative Crimes Act (ACA) is the intent of Congress. Ante, at 9–10. One of Congress’ purposes in enacting the ACA was to fill gaps in federal criminal law. Ante, at 3. The majority fails to weigh, however, a second, countervailing policy behind the ACA: the value of federalism. The intent of Congress was to preserve state law except where it is “displaced by specific laws enacted by Congress.” Franklin v. United States, 216 U. S. 559, 568 (1910). In other words, the ACA embodies Congress’ “policy of general conformity to local law.” United States v. Sharpnack, 355 U. S. 286, 289 (1958). The majority quotes these passages with approval, ante, at 3–4, yet ignores the principles of federalism upon which they rest.

A central tenet of federalism is concurrent jurisdiction over many subjects. See McCulloch v. Maryland, 4 Wheat. 316, 425, 435 (1819). One result of concurrent jurisdiction is that, outside federal enclaves, citizens can be subject to the criminal laws of both state and federal sovereigns for the same act or course of conduct. See Heath v. Alabama, 474 U. S. 82, 88–89 (1985). The ACA seeks to mirror the results of concurrent jurisdiction in enclaves where, but for its provisions, state laws would be suspended in their entirety. Congress chose this means to recognize and respect the power of both sovereigns. We should implement this principle by assimilating state law except where Congress has manifested a contrary intention in “specific [federal] laws.” Franklin, supra, at 568. But see ante, at 7 (suggesting that persons within federal enclaves should not be “randomly subject” to state as well as federal law, even though both sovereigns regulate those outsideenclaves).

The majority recognizes that assimilation is not barred simply because the conduct at issue could be punished under a federal statute. It is correct, then, to assume that assimilation depends on whether Congress has proscribed the same offense. Ante, at 4–5. Yet in trying to define the same offense, the majority asks whether assimilation would interfere with a federal policy, rewrite a federal offense, or intrude upon a field occupied by the Federal Government. Ante, at 8. The majority’s standards are a roundabout way to ask whether specific federal laws conflict with state laws. The standards take too little note of the value of federalism and the concomitant presumption in favor of assimilation. And for many concrete cases, they are too vague to be of help.

A more serious problem with the majority’s approach, however, is that it undervalues the best indicia of congressional intent: the words of the criminal statutes in question and the factual elements they define. There is a methodology at hand for this purpose, and it is the Blockburger test we use in double jeopardy law. See Blockburger v. United States, 284 U. S. 299 (1932); see also Missouri v. Hunter, 459 U. S. 359, 366–367 (1983) (Blockburger is a rule for divining congressional intent). Under Blockburger, we examine whether “[e]ach of the offenses created requires proof of a different element.” 284 U. S., at 304. In other words, does “each provision requir[e] proof of a fact which the other does not”? Ibid.

The same-elements test turns on the texts of the statutes in question, the clearest and most certain indicators of the will of Congress. The test is straightforward, and courts and Congress are already familiar with its dynamic. Following Blockburger, a same-elements approach under the ACA would respect federalism by allowing a broad scope for assimilation of state law. The majority rejects this approach, however, because federal and state statutes may have trivial differences in wording or may differ in jurisdictional elements. Ante, at 6–7, 9.

It would be simpler and more faithful to federalism to use a same-elements inquiry as the starting point for the ACA analysis. Courts could use this standard and still accommodate the majority’s concerns. Under this view, we would look beyond slight differences in wording and jurisdictional elements to discern whether, as a practical matter, the elements of the two crimes are the same. The majority frets that a small difference in the definitions of purses in federal and state purse-snatching laws would by itself permit assimilation. Ante, at 7. But a slight difference in definition need not by itself allow assimilation. See Amar & Marcus, Double Jeopardy Law After Rodney King, 95 Colum. L. Rev. 1, 38–44 (1995) (advocating a similar approach for double jeopardy claims involving combinations of federal and state offenses). The majority also wonders whether one could assimilate state laws forbidding robbery of state-chartered banks because a federal bank-robbery law did not require a state charter. Ante, at 7. But again, a jurisdictional element need not by itself allow assimilation, if all substantive elements of the offenses are identical.

Because the purposes of the ACA and double jeopardy law differ, some other adjustments to Blockburger may be necessary. For instance, Blockburger treats greater and lesser included offenses as the same to protect the finality of a single prosecution, but finality is not the purpose of the ACA. Congress chooses to allow greater and lesser included offenses to coexist at the federal level, though a particular offender cannot be convicted of both. So too the existence of a lesser included federal offense does not prevent the assimilation of a greater state offense under the ACA, or vice versa. See ante, at 16 (citing cases finding federal assault statute does not prevent assimilation of state child-abuse laws).

Another way in which the ACA differs from double jeopardy law is compelled by our own precedent interpreting the ACA. See Williams v. United States, 327 U. S. 711 (1946). Congress sometimes adverts to a specific element of an offense and sets it at a level different from the level set by state law. When the federal and state offenses have otherwise identical elements, assimilation is not proper. In the Williams case, for example, a state statutory-rape law set the age of majority at 18. Id., at 716. Congress had enacted a federal carnal-knowledge statute, setting the age of majority at 16. Id., at 714, n. 6. Once Congress had adverted to and set the age of majority, state law could not be used to rewrite and broaden this particular element. See id., at 717–718, 724–725. Because Congress had manifested a clear intent to the contrary, assimilation was improper. The same would be true if a state grand-larceny law required a theft of at least $200, while a federal grand-larceny law required a theft of $250 or more.

Congress could have defined first-degree murder to include the killing of children younger than 3, even though state law set the requisite age at 12. Had Congress done so, Williams would apply and assimilation of state law would be improper if all other elements were the same. Here, on the other hand, Congress has not taken a victim’s age into account at all in defining first-degree murder. The state offense includes a substantive age element missing from the federal statute, so the two do not share the same elements and assimilation is proper. The majority’s analysis is more obscure and leads it to an incorrect conclusion. For these reasons, and with all respect, I dissent.