with whom Justice Ginsburg joins, concurring in part and dissenting in part.
I agree with the way today’s opinion describes a district court’s tasks in sentencing under the Guidelines, and the role of a court of appeals in reviewing sentences, but I part company from the Court in applying its standard on two specific points. I would affirm the Court of Appeals’s rejection of the downward departures based on susceptibility to abuse in prison and on successive prosecution, for to do otherwise would be to attribute an element of irrationality to the Commission and to its “heartland” concept. Accordingly, I join the Court’s opinion except Part IV-B-3.
As the majority notes, ante, at 106, “Congress did not grant federal courts authority to decide what sorts of sentencing considerations are inappropriate in every circumstance.” In *115fact, Congress allowed district courts to depart from the Guidelines only if “the court finds that there exists an aggravating or mitigating circumstance of a kind, or to a degree, not adequately taken into consideration by the Sentencing Commission in formulating the guidelines that should result in a sentence different from that described.” 18 U. S. C. § 3553(b); see also ante, at 92-93. While discussing departures, the Commission quotes this language from § 3553(b), before stating that “[w]hen a court finds an atypical case,... the court may consider whether a departure is warranted.” United States Sentencing Commission, Guidelines Manual ch. 1, pt. A, intro, comment. 4(b) (Nov. 1995) (1995 USSG). Thus, both Congress and the Commission envisioned that departures would require some unusual factual circumstance, but would be justified only if the factual difference “should” result in a different sentence. Departures, in other words, must be consistent with rational normative order.
As to the consideration of susceptibility to abuse in prison, the District Court departed downward because it believed that “the widespread publicity and emotional outrage which have surrounded this case from the outset, in addition to the [petitioners’] status as police officers, lead the Court to find that Koon and Powell are particularly likely to be targets of abuse during their incarceration.” 833 F. Supp. 769, 788 (CD Cal. 1993). That is, the District Court concluded that petitioners would be subject to abuse not simply because they were former police officers, but in large part because of the degree of publicity and condemnation surrounding their crime.1 But that reasoning overlooks the fact that the publicity stemmed from the remarkable brutality of petitioners’ proven behavior, which it was their misfortune to have pre*116cisely documented on film. To allow a departure on this basis is to reason, in effect, that the more serious the crime, and the more widespread its consequent publicity and condemnation, the less one should be punished; the more egregious the act, the less culpable the offender. In the terminology of the Guidelines, such reasoning would take the heartland to be the domain of the less, not the more, deplorable of the acts that might come within the statute. This moral irrationality cannot be attributed to the heartland scheme, however, and rewarding the relatively severe offender could hardly have been in the contemplation of a Commission that discouraged downward departures for susceptibility to prison abuse even when the nonculpable reason is an unusual “[pjhysical . . . appearance, including physique.” 1995 USSG § 5H1.4; see also ante, at 107; 1995 USSG ch. 1, pt. A, intro, comment. 3 (discussing the principle of “ ‘just deserts,’” which the Commission describes as a concept under which “punishment should be scaled to the offender’s culpability and the resulting harms”).2
The Court of Appeals appreciated the significance of the requisite moral calculus when it wrote that “[a]ny public outrage was the direct result of [petitioners’] criminal acts. It is incongruous and inappropriate to reduce [petitioners’] sentences specifically because individuals in society have condemned their acts as criminal and an abuse of the trust that society placed in them.” 34 F. 3d 1416, 1456 (CA9 1994). The Court of Appeals should be affirmed on this point.
I believe that it was also an abuse of discretion for the District Court to depart downward because of the successive prosecutions.3 In these cases, there were facial showings *117that the state court system had malfunctioned when the petitioners were acquitted (or, in the case of one charge, had received no verdict), and without something more one cannot accept the District Court’s conclusion that there was no demonstration that a “clear miscarriage of justice” caused the result in the state trial. 833 F. Supp., at 790. This is so simply because the federal prosecutors, in proving their cases, proved conduct constituting the crimes for which petitioners had been prosecuted unsuccessfully in the state court. See Powell v. Superior Court, 232 Cal. App. 3d 785, 789, 283 Cal. Rptr. 777, 779 (1991) (noting that petitioners were charged, inter alia, with assault by force likely to produce great bodily injury, Cal. Penal Code Ann. § 245(a)(1) (West 1988), and being an officer unnecessarily assaulting or beating any person in violation of § 149); § 149 (“Every public officer who, under color of authority, without lawful necessity, assaults or beats any person” commits an offense); § 245(a)(1) (“Every person who commits an assault upon the person of another ... by any means of force likely to produce great bodily injury” commits an offense); ante, at 87-88 (observing that petitioners were tried in state court for assault with a deadly weapon and excessive use of force by a police officer and tried in federal court for willfully using or willfully allowing others to use unreasonable force in arresting King); 833 F. Supp., at 790 (stating that the “same underlying conduct” was involved in both cases). While such a facial showing resulting from the identity of factual predicates for the state and federal prosecutions might in some cases be overcome (by demonstrating, say, that a crucial witness for *118the State was unavailable in the state trial through no one’s fault), there was no evidence to overcome it here.
As a consequence, reading the Guidelines to suggest that those who profit from state-court malfunctions should get the benefit of a downward departure would again attribute a normative irrationality to the heartland concept. The sense of irrationality here is, to be sure, different from what was presupposed by the District Court’s analysis on the issue of susceptibility to abuse in prison, for the incongruity produced by downward departures here need not depend on the defendant’s responsibility for the particular malfunction of the state system. But the fact remains that it would be a normatively obtuse sentencing scheme that would reward a defendant whose federal prosecution is justified solely because he has obtained the advantage of injustice produced by the failure of the state system.
This is not, of course, to say that a succession of state and federal prosecutions may never justify a downward departure. If a comparison of state and federal verdicts in relation to their factual predicates indicates no incongruity, a downward departure at federal sentencing could well be consistent with an application of a rational heartland concept. But these are not such cases.
Although it is not essential to my analysis, I note in passing that the unusual extent of outside publicity is probably irrelevant in the prison environment. Given any amount of outside publicity, prison inmates quickly learn about new arrivals, including former police officers, and the crimes of which they were convicted.
The requirement of normative order does not, of course, say anything one way or the other about considering exceptionally unusual physical appearance as a basis to anticipate abuse.
It is true, factually, that successive federal prosecutions after state proceedings occur very rarely even in criminal civil rights prosecutions, U. S. Commission on Civil Rights, Who is Guarding the Guardians?, 112, 116 (Oct. 1981) (noting that between 50 and 100 police misconduct cases *117are brought each year and that from March 1977 to September 1980 only seven successive prosecutions were authorized); United States v. Davis, 906 F. 2d 829, 832 (CA2 1990) (“In practice, successive prosecutions for the same conduct remain rarities”). Those figures do not, however, demonstrate that all convictions on successive federal prosecutions under 18 U. S. C. §242 should for that reason be subject to discretion to depart downward, for they do not take account of the normative ordering, discussed below.