JUSTICE O'CONNOR, dissenting.
Although I agree with the result reached in JUSTICE REHNQUIST's dissent, I write separately because I cannot agree with the approach taken by either the Court or JUSTICE REHNQUIST. Both opinions engage in exhaustive, but ultimately unilluminating, exegesis of the common law of the availability of punitive damages in 1871. Although both the Court and JUSTICE REHNQUIST display admirable skills in legal research and analysis of great numbers of musty cases, the results do not significantly further the goal of the inquiry: to establish the intent of the 42d Congress. In interpreting § 1983, we have often looked to the common law as it existed in 1871, in the belief that, when Congress was silent on a point, it intended to adopt the principles of the common law with which it was familiar. See, e.g., 453 U. S. Fact Concerts, Inc., 453 U. S. 247, 453 U. S. 258 (1981); Carey v. Piphus, 435 U. S. 247, 435 U. S. 255 (1978). This approach makes sense when there was a generally prevailing rule of common law, for then it is reasonable to assume that Congressmen were familiar with that rule and imagined that it would cover the cause of action that they were creating. But when a significant split in authority existed, it strains credulity to argue that Congress simply assumed that one view, rather than the other, would govern. Particularly in a case like this one, in which those interpreting the common law of 1871 must resort to dictionaries in an attempt to translate the language of the late 19th century into terms that judges of the late 20th century can understand, see ante at 461 U. S. 39 -41, n. 8; 461 U. S. 61 -64, nn. 3, 4, and in an area in which the courts of the earlier period frequently used inexact and contradictory language, see ante at 461 U. S. 45 -47, n. 12, we cannot safely infer anything about congressional intent from the divided contemporaneous judicial opinions. The battle of the string citations can have no winner.
Once it is established that the common law of 1871 provides us with no real guidance on this question, we should turn to the policies underlying § 1983 to determine which rule best accords with those policies. In Fact Concerts, we identified the purposes of § 1983 as preeminently to compensate victims of constitutional violations and to deter further violations. 453 U.S. at 453 U. S. 268. See also Robertson v. Wegmann, 436 U. S. 584, 436 U. S. 590 -591 (1978); Carey v. Piphus, supra, at 435 U. S. 254 -257, and n. 9. The conceded availability of compensatory damages, particularly when coupled with the availability of attorney's fees under § 1988, completely fulfills the goal of compensation, leaving only deterrence to be served by awards of punitive damages. We must then confront the close question whether a standard permitting an award of unlimited punitive damages on the basis of recklessness will chill public officials in the performance of their duties more than it will deter violations of the Constitution, and whether the availability of punitive damages for reckless violations of the Constitution in addition to attorney's fees will create an incentive to bring an ever-increasing flood of § 1983 claims, threatening the ability of the federal courts to handle those that are meritorious. Although I cannot concur in JUSTICE REHNQUIST's wholesale condemnation of awards of punitive damages in any context, or with the suggestion that punitive damages should not be available even for intentional or malicious violations of constitutional rights, I do agree with the discussion in 461 U. S. Since awards of compensatory damages and attorney's fees already provide significant deterrence, I am persuaded that the policies counseling against awarding punitive damages for the recklessness of public officials outweigh the desirability of any incremental deterrent effect that such awards may have. Consequently, I dissent.