This case raises issues concerning the scope of judicial immunity from a civil suit that seeks injunctive and declaratory relief under § 1 of the Civil Rights Act of 1871, as amended, 42 U.S.C. § 1983, and from fee awards made under the Civil Rights Attorney's Fees Awards Act of 1976, 90 Stat. 2641, as amended, 42 U.S.C. § 1988.

Petitioner Gladys Pulliam is a state Magistrate in Culpeper County, Va. Respondents Richmond R. Allen and Jesse W. Nicholson were plaintiffs in a § 1983 action against Pulliam brought in the United States District Court for the Eastern Distict of Virginia. They claimed that Magistrate Pulliam's practice of imposing bail on persons arrested for nonjailable offenses under Virginia law and of incarcerating those persons if they could not meet the bail was unconstitutional. The District Court agreed and enjoined the practice. That court also awarded respondents $7,691.09 in costs and attorney's fees under § 1988. The United States Court of Appeals for the Fourth Circuit rejected petitioner's claim that the award of attorney's fees against her should have been barred by principles of judicial immunity. We agree with the Court of Appeals and affirm the award.

* Respondent Allen was arrested in January 1980 for allegedly using abusive and insulting language, a Class 3 misdemeanor under Va.Code § 18.2-416 (1982). The maximum penalty for a Class 3 misdemeanor is a $500 fine. See § 18.2-11(c). Petitioner set a bond of $250. Respondent Allen was unable to post the bond, and petitioner committed Allen to the Culpeper County jail, where he remained for 14 days. He was then tried, found guilty, fined, and released. The trial judge subsequently reopened the judgment and reversed the conviction. Allen then filed his § 1983 claim, seeking declaratory and injunctive relief against petitioner's practice of incarcerating persons waiting trial for nonincarcerable offenses.1

Respondent Nicholson was incarcerated four times within the 2-month period immediately before and after the filing of Allen's complaint. His arrests were for alleged violations of Va.Code § 18.2-388 (1982), being drunk in public. Section 18.2-388 is a Class 4 misdemeanor for which the maximum penalty is a $100 fine. See § 18.2-11(d). Like Allen, respondent Nicholson was incarcerated for periods of two to six days for failure to post bond. He intervened in Allen's suit as a party plaintiff.

The District Court found it to be petitioner's practice to require bond for nonincarcerable offenses. The court declared the practice to be a violation of due process and equal protection and enjoined it.2 The court also found that respondents, having substantially prevailed on their claims, were entitled to costs, including reasonable attorney's fees, in accordance with § 1988. It directed respondents to submit a request for costs to petitioner within 10 days. App. 23. Petitioner did not appeal this order.

Respondents submitted a request for fees and costs totalling $7,691.09. The fee component of this figure was $7,038. Petitioner filed objections and prayed "that the Court reduce the request of Plaintiffs for attorney's fees." Id., at 33. The court found the fees figure reasonable and granted fees and costs in the requested amount.

Petitioner took an appeal from the order awarding attorney's fees against her. She argued that, as a judicial officer, she was absolutely immune from an award of attorney's fees. The Court of Appeals reviewed the language and legislative history of § 1988. It concluded that a judicial officer is not immune from an award of attorney's fees in an action in which prospective relief properly is awarded against her. Since the court already had determined that judicial immunity did not extend to injunctive and declaratory relief under § 1983,3 the court concluded that prospective relief properly had been awarded against petitioner. It therefore affirmed the award of attorney's fees. Allen v. Burke, 690 F.2d 376 (1982).

We granted certiorari in this case, 461 U.S. 904, 103 S.Ct. 1873, 76 L.Ed.2d 806 (1983), to determine, as petitioner phrased the question, "[w] hether Judicial Immunity Bars the Award of Attorney's Fees Pursuant to 42 U.S.C. § 1988 Against a Member of the Judiciary Acting in his Judicial Capacity." See the initial leaf of the petition for certiorari. As the Court of Appeals recognized, the answer to that question depends in part on whether judicial immunity bars an award of injunctive relief under § 1983. The legislative history of § 1988 clearly indicates that Congress intended to provide for attorney's fees in cases where relief properly is granted against officials who are immune from damages awards. H.R.Rep. No. 94-1558, p. 9 (1976).4 There is no indication, however, that Congress intended to provide for a fee award if the official was immune from the underlying relief on which the award was premised. See Supreme Court of Virginia v. Consumers Union of United States, Inc., 446 U.S. 719, 738-739, 100 S.Ct. 1967, 1977-1978, 64 L.Ed.2d 641 (1980). Before addressing the specific provisions of § 1988, therefore, we turn to the more fundamental question, that is, whether a judicial officer acting in her judicial capacity should be immune from prospective injunctive relief.5

Although injunctive relief against a judge rarely is awarded, the United States Courts of Appeals that have faced the issue are in agreement that judicial immunity does not bar such relief.6 This Court, however, has never decided the question.7

The starting point in our own analysis is the common law. Our cases have proceeded on the assumption that common-law principles of legislative and judicial immunity were incorporated into our judicial system and that they should not be abrogated absent clear legislative intent to do so. See Pierson v. Ray, 386 U.S. 547, 554-555, 87 S.Ct. 1213, 1217-1218, 18 L.Ed.2d 288 (1967); Tenney v. Brandhove, 341 U.S. 367, 71 S.Ct. 783, 95 L.Ed. 1019 (1951). Accordingly, the first and crucial question is whether the common law recognized judicial immunity from prospective collateral relief.

At the common law itself, there was no such thing as an injunction against a judge. Injunctive relief was an equitable remedy that could be awarded by the Chancellor only against the parties in proceedings before other courts. See 2 J. Story, Equity Jurisprudence ¶ 875, p. 72 (11th ed. 1873). This limitation on the use of the injunction, however, says nothing about the scope of judicial immunity. And the limitation derived not from judicial immunity, but from the substantive confines of the Chancellor's authority. Ibid.

Although there were no injunctions against common-law judges, there is a common-law parallel to the § 1983 injunction at issue here. That parallel is found in the collateral prospective relief available against judges through the use of the King's prerogative writs. A brief excursion into common-law history helps to explain the relevance of these writs to the question whether principles of common-law immunity bar injunctive relief against a judicial officer.

The doctrine of judicial immunity and the limitations on prospective collateral relief with which we are concerned have related histories. Both can be traced to the successful efforts of the King's Bench to ensure the supremacy of the common-law courts over their 17th- and 18th-century rivals. See 5 W. Holdsworth, A History of English Law 159-160 (3d ed. 1945) (Holdsworth).

A number of courts challenged the King's Bench for authority in those days. Among these were the Council, the Star Chamber, the Chancery, the Admiralty, and the ecclesiastical courts. Ibid. In an effort to assert the supremacy of the common-law courts, Lord Coke forbade the interference by courts of equity with matters properly triable at common law. See Heath v. Rydley, Cro.Jac. 335, 79 Eng.Rep. 286 (K.B.1614). Earlier, in Floyd and Barker, 12 Co.Rep. 23, 77 Eng.Rep. 1305 (1607), Coke and his colleagues of the Star Chamber had declared the judges of the King's Bench immune from prosecution in competing courts for their judicial acts. In doing so, they announced the theory upon which the concept of judicial immunity was built. The judge involved in Floyd and Barker was a common-law Judge of Assize who had presided over a murder trial. He was then charged in the Star Chamber with conspiracy. The court concluded that the judges of the common law should not be called to account "before any other Judge at the suit of the King." Id., at 24, 77 Eng.Rep., at 1307.

"[A]nd it was agreed, that insomuch as the Judges of the realm have the administration of justice, under the King, to all his subjects, they ought not to be drawn into question for any supposed corruption, which extends to the annihilating of a record, or of any judicial proceedings before them, or tending to the slander of the justice of the King, which will trench to the scandal of the King himself, except it be before the King himself; for they are only to make an account to God and the King, and not to answer to any suggestion in the Star-Chamber." Id., at 25, 77 Eng.Rep., at 1307.

As this quoted language illustrates, Coke's principle of immunity extended only to the higher judges of the King's courts. See 5 Holdsworth, at 159-160. In time, Coke's theory was expanded beyond his narrow concern of protecting the common-law judges from their rival courts, so that judges of all courts were accorded immunity, at least for actions within their jurisdiction.8 See Scott v. Stansfield, 3 L.R.Ex. 220 (1868) (immunity extended to a county court, an inferior court of record; reliance placed on precedent extending immunity to the court of a coroner and to a court-martial, an inferior court and a court not of record); Haggard v. Pelicier Freres [1892] A.C. 61 (1891) (judge of Consular Court of Madagascar given same immunity as judge of a court of record). In addition, the theory itself was refined, its focus shifting from the need to preserve the King's authority to the public interest in independent judicial decisionmaking. See Taaffe v. Downes, reprinted in footnote in Calder v. Halket, 13 Eng.Rep. 12, 18, n. (a) (P.C.1840) ("An action before one Judge for what is done by another, is in the nature of an Appeal; and is the Appeal from an equal to an equal. It is a solecism in the law . . . that the Plaintiff's case is against the independence of the Judges").

By 1868, one of the judges of the Court of Exchequer explained judicial immunity in language close to our contemporary understanding of the doctrine:

"It is essential in all courts that the judges who are appointed to administer the law should be permitted to administer it under the protection of the law, independently and freely, without favor and without fear. This provision of the law is not for the protection or benefit of a malicious or corrupt judge, but for the benefit of the public, whose interest it is that the judges should be at liberty to exercise their functions with independence, and without fear of consequences." Scott v. Stansfield, 3 L.R.Ex., at 223, quoted in Bradley v. Fisher, 13 Wall. 335, 350, n., 20 L.Ed. 646 (1872).

It is in the light of the common law's focus on judicial independence that the collateral control exercised by the King's Bench over rival and inferior courts has particular significance.

The King's Bench exercised significant collateral control over inferior and rival courts through the use of prerogative writs. The writs included habeas corpus, certiorari, prohibition, mandamus, quo warranto, and ne exeat regno. 1 Holdsworth, at 226-231 (7th ed. 1956). Most interesting for our current purposes are the writs of prohibition and mandamus.9 The writs issued against a judge, in theory to prevent him from exceeding his jurisdiction or to require him to exercise it. Id., at 228-229. In practice, controlling an inferior court in the proper exercise of its jurisdiction meant that the King's Bench used and continues to use the writs to prevent a judge from committing all manner of errors, including departing from the rules of natural justice, proceeding with a suit in which he has an interest, misconstruing substantive law, and rejecting legal evidence. See 1 Halsbury's Laws of England &Par; 76, 81, 130 (4th ed. 1973); Gordon, The Observance of Law as a Condition of Jurisdiction, 47 L.Q.Rev. 386, 394 (1931).10

Examples are numerous in which a judge of the King's Bench, by issuing a writ of prohibition at the request of a party before an inferior or rival court, enjoined that court from proceeding with a trial or from committing a perceived error during the course of that trial. See generally Dobbs, The Decline of Jurisdiction by Consent, 40 N.C.L.Rev. 49, 60-61 (1961). The writs were particularly useful in exercising collateral control over the ecclesiastical courts, since the King's Bench exercised no direct review over those tribunals. In Shatter v. Friend, 1 Show. 158, 89 Eng.Rep. 510 (K.B.1691), for example, the court granted a prohibition against the Spiritual Court for refusing to allow the defendant's proof of payment of a 10-pound legacy, one of the justices concluding that "it was an unconscionable unreasonable thing to disallow the proof." Id., at 161, 89 Eng.Rep., at 512.11

In Gould v. Gapper, 5 East. 345, 102 Eng.Rep. 1102 (K.B.1804), the court made explicit what had been implicit in a number of earlier decisions. It held that a writ of prohibition would be granted not only when a court had exceeded its jurisdiction, but also when the court, either a noncommon-law court or an inferior common-law court, had misconstrued an Act of Parliament or, acting under the rules of the civil law, had decided otherwise than the courts of common law would upon the same subject. The fact that the error might be corrected on appeal was deemed to be irrelevant to the availability of a writ of prohibition. In the court's view, the reason for prohibition in such a case was "[n]ot that the Spiritual Court had not jurisdiction to construe [the statute], but that the mischiefs of misconstruction were to be prevented by prohibition." Id., at 368, 102 Eng.Rep., at 1111.12

Although the King's Bench exercised direct review of the inferior common-law courts, it also used the writ of prohibition to control those courts. See, e.g., In re Hill, 10 Exch. 726 (1855) (prohibition issued to prevent judge from proceeding in a case in which he, of his own accord, had amended a claim to an amount within his jurisdiction).13

The practice has continued into modern times. In King v. Emerson, [1913] 2 Ir.R. 377, for instance, the court granted a writ of prohibition preventing a justice of the peace, acting in a judicial capacity, from proceeding with a deposition, because of a likelihood that a reasonable public might conclude that the magistrate's statements indicated bias in favor of the Crown. The court directed the magistrate to pay costs to the complaining party, leaving him to settle with the Crown the matter of indemnification.

The relationship between the King's Bench and its collateral and inferior courts is not precisely paralleled in our system by the relationship between the state and federal courts. To the extent that we rely on the common-law practice in shaping our own doctrine of judicial immunity, however, the control exercised by the King's Bench through the prerogative writs is highly relevant. It indicates that, at least in the view of the common law, there was no inconsistency between a principle of immunity that protected judicial authority from "a wide, wasting, and harassing persecution," Taaffe v. Downes, 13 Eng.Rep., at 18, n. (a), and the availability of collateral injunctive relief in exceptional cases. Nor, as indicated above, did the common law deem it necessary to limit this collateral relief to situations where no alternative avenue of review was available. See Gould v. Gapper, supra.

It is true that the King's Bench was successful in insulating its judges from collateral review. But that success had less to do with the doctrine of judicial immunity than with the fact that only the superior judges of the King's Bench, not the ecclesiastical courts or the inferior common-law courts, had authority to issue the prerogative writs.14

Our own experience is fully consistent with the common law's rejection of a rule of judicial immunity from prospective relief. We never have had a rule of absolute judicial immunity from prospective relief, and there is no evidence that the absence of that immunity has had a chilling effect on judicial independence. None of the seminal opinions on judicial immunity, either in England or in this country, has involved immunity from injunctive relief.15 No Court of Appeals ever has concluded that immunity bars injunctive relief against a judge. See n. 6, supra. At least seven Circuits have indicated affirmatively that there is no immunity bar to such relief, and in situations where in their judgment an injunction against a judicial officer was necessary to prevent irreparable injury to a petitioner's constitutional rights, courts have granted that relief.16

For the most part, injunctive relief against a judge raises concerns different from those addressed by the protection of judges from damages awards. The limitations already imposed by the requirements for obtaining equitable relief against any defendant a showing of an inadequate remedy at law and of a serious risk of irreparable harm, see Beacon Theatres, Inc. v. Westover, 359 U.S. 500, 506-507, 79 S.Ct. 948, 954-955, 3 L.Ed.2d 988 (1959)17 severely curtail the risk that judges will be harassed and their independence compromised by the threat of having to defend themselves against suits by disgruntled litigants.18 Similar limitations serve to prevent harassment of judges through use of the writ of mandamus. Because mandamus has "the unfortunate consequence of making the judge a litigant, obliged to obtain personal counsel or to leave his defense to one of the litigants before him," the Court has stressed that it should be "reserved for really extraordinary causes." Ex parte Fahey, 332 U.S. 258, 260, 67 S.Ct. 1558, 1559, 91 L.Ed. 2041 (1947). Occasionally, however, there are "really extraordinary causes" and, in such cases, there has been no suggestion that judicial immunity prevents the supervising court from issuing the writ.19

The other concern raised by collateral injunctive relief against a judge, particularly when that injunctive relief is available through § 1983, relates to the proper functioning of federal-state relations. Federal judges, it is urged, should not sit in constant supervision of the actions of state judicial officers, whatever the scope of authority under § 1983 for issuing an injunction against a judge.

The answer to this concern is that it is not one primarily of judicial independence, properly addressed by a doctrine of judicial immunity. The intrusion into the state process would result whether the action enjoined were that of a state judge or of another state official. The concern, therefore, has been addressed as a matter of comity and federalism, independent of principles of judicial immunity.20 We reaffirm the validity of those principles and the need for restraint by federal courts called on to enjoin the actions of state judicial officers. We simply see no need to reinterpret the principles now as stemming from the doctrine of judicial immunity.

If the Court were to employ principles of judicial immunity to enhance further the limitations already imposed by principles of comity and federalism on the availability of injunctive relief against a state judge, it would foreclose relief in situations where, in the opinion of a federal judge, that relief is constitutionally required and necessary to prevent irreparable harm. Absent some basis for determining that such a result is compelled, either by the principles of judicial immunity, derived from the common law and not explicitly abrogated by Congress, or by Congress' own intent to limit the relief available under § 1983, we are unwilling to impose those limits ourselves on the remedy Congress provided.

As illustrated above, there is little support in the common law for a rule of judicial immunity that prevents injunctive relief against a judge. There is even less support for a conclusion that Congress intended to limit the injunctive relief available under § 1983 in a way that would prevent federal injunctive relief against a state judge. In Pierson v. Ray, 386 U.S. 547, 87 S.Ct. 1213, 18 L.Ed.2d 288 (1967), the Court found no indication of affirmative congressional intent to insulate judges from the reach of the remedy Congress provided in § 1983. The Court simply declined to impute to Congress the intent to abrogate common-law principles of judicial immunity. Absent the presumption of immunity on which Pierson was based, nothing in the legislative history of § 1983 or in this Court's subsequent interpretations of that statute supports a conclusion that Congress intended to insulate judges from prospective collateral injunctive relief.

Congress enacted § 1983 and its predecessor, § 2 of the Civil Rights Act of 1866, 14 Stat. 27, to provide an independent avenue for protection of federal constitutional rights. The remedy was considered necessary because "state courts were being used to harass and injure individuals, either because the state courts were powerless to stop deprivations or were in league with those who were bent upon abrogation of federally protected rights." Mitchum v. Foster, 407 U.S. 225, 240, 92 S.Ct. 2151, 2161, 32 L.Ed.2d 705 (1972). See also Pierson v. Ray, 386 U.S., at 558-564, 87 S.Ct., at 1219-1222 (dissenting opinion) (every Member of Congress who spoke to the issue assumed that judges would be liable under § 1983).

Subsequent interpretations of the Civil Rights Acts by this Court acknowledge Congress' intent to reach unconstitutional actions by all state actors, including judges. In Ex parte Virginia, 100 U.S. 339, 25 L.Ed. 676 (1880), § 4 of the Civil Rights Act of 1875, 18 Stat. 336, was employed to authorize a criminal indictment against a judge for excluding persons from jury service on account of their race. The Court reasoned that the Fourteenth Amendment prohibits a State from denying any person within its jurisdiction the equal protection of the laws. Since a State acts only by its legislative, executive, or judicial authorities, the constitutional provision must be addressed to those authorities, including the State's judges. Section 4 was an exercise of Congress' authority to enforce the provisions of the Fourteenth Amendment and, like the Amendment, reached unconstitutional state judicial action.21

The interpretation in Ex parte Virginia of Congress' intent in enacting the Civil Rights Acts has not lost its force with the passage of time. In Mitchum v. Foster, supra, the Court found § 1983 to be an explicit exception to the anti-injunction statute, citing Ex parte Virginia for the proposition that the "very purpose of § 1983 was to interpose the federal courts between the States and the people, as guardians of the people's federal rights to protect the people from unconstitutional action under color of state law, 'whether that action be executive, legislative, or judicial.' " 407 U.S., at 242, 92 S.Ct., at 2162.

Much has changed since the Civil Rights Acts were passed. It no longer is proper to assume that a state court will not act to prevent a federal constitutional deprivation or that a state judge will be implicated in that deprivation. We remain steadfast in our conclusion, nevertheless, that Congress intended § 1983 to be an independent protection for federal rights and find nothing to suggest that Congress intended to expand the common-law doctrine of judicial immunity to insulate state judges completely from federal collateral review.

We conclude that judicial immunity is not a bar to prospective injunctive relief against a judicial officer acting in her judicial capacity. In so concluding, we express no opinion as to the propriety of the injunctive relief awarded in this case. Petitioner did not appeal the award of injunctive relief against her. The Court of Appeals therefore had no opportunity to consider whether respondents had an adequate remedy at law, rendering equitable relief inappropriate,22 or whether the order itself should have been more narrowly tailored. On the record before us and without the benefit of the Court of Appeals' assessment, we are unwilling to speculate about these possibilities. We proceed, therefore, to the question whether judicial immunity bars an award of attorney's fees, under § 1988, to one who succeeds in obtaining injunctive relief against a judicial officer.

Petitioner insists that judicial immunity bars a fee award because attorney's fees are the functional equivalent of monetary damages and monetary damages indisputably are prohibited by judicial immunity. She reasons that the chilling effect of a damages award is no less chilling when the award is denominated attorney's fees.

There is, perhaps, some logic to petitioner's reasoning.

The weakness in it is that it is for Congress, not this Court, to determine whether and to what extent to abrogate the judiciary's common-law immunity. See Pierson v. Ray, 386 U.S., at 554, 87 S.Ct., at 1217. Congress has made clear in § 1988 its intent that attorney's fees be available in any action to enforce a provision of § 1983. See also Hutto v. Finney, 437 U.S. 678, 694, 98 S.Ct. 2565, 2575, 57 L.Ed.2d 522 (1978). The legislative history of the statute confirms Congress' intent that an attorney's fee award be available even when damages would be barred or limited by "immunity doctrines and special defenses, available only to public officials." H.R.Rep. No. 94-1558, p. 9 (1976).23 See also Supreme Court of Virginia v. Consumers Union of United States, Inc., 446 U.S., at 738-739, 100 S.Ct., at 1977-1978 ("The House Committee Report on [§ 1988] indicates that Congress intended to permit attorney's fees awards in cases in which prospective relief was properly awarded against defendants who would be immune from damages awards").

Congress' intent could hardly be more plain. Judicial immunity is no bar to the award of attorney's fees under 42 U.S.C. § 1988.

The judgment of the Court of Appeals, allowing the award of attorney's fees against petitioner, is therefore affirmed.

It is so ordered.

Justice POWELL, with whom THE CHIEF JUSTICE, Justice REHNQUIST, and Justice O'CONNOR join, dissenting.

The Court today reaffirms the rule that judges are immune from suits for damages, but holds that they may be sued for injunctive and declaratory relief and held personally liable for money judgments in the form of costs and attorney's fees merely on the basis of erroneous judicial decisions. The basis for the Court's distinction finds no support in common law and in effect eviscerates the doctrine of judicial immunity that the common law so long has accepted as absolute.

The Court recognizes that the established principle of judicial immunity serves as the bulwark against threats to "independent judicial decisionmaking," ante, at 531. Yet, at the same time it concludes that judicial immunity does not bar suits for injunctive or declaratory relief with the attendant claims for costs and attorney's fees. The Court reasons that "[f] or the most part, injunctive relief against a judge raises concerns different from those addressed by the protection of judges from damages awards." Ante, at 1978. This case illustrates the unsoundness of that reasoning. The Court affirms a $7,691.09 money judgment awarded against a state Magistrate on the determination that she made erroneous judicial decisions with respect to bail and pretrial detentions. Such a judgment poses the same threat to independent judicial decisionmaking whether it be labeled "damages" of $7,691.09 or "attorney's fees" in that amount. Moreover, as was held a century and a half ago, an "action before one Judge for what is done by another . . . [is a] case . . . against the independence of the Judges." Taaffe v. Downes, reprinted in footnote in Calder v. Halket, 13 Eng.Rep. 12, 18, n. (a) (P.C. 1840). The burdens of having to defend such a suit are identical in character and degree, whether the suit be for damages or prospective relief. The holding of the Court today subordinates realities to labels. The rationale of the common-law immunity cases refutes the distinction drawn by the Court.

* Since 1869, this Court consistently has held that judges are absolutely immune from civil suits for damages. See, e.g., Stump v. Sparkman, 435 U.S. 349, 98 S.Ct. 1099, 55 L.Ed.2d 331 (1978); Pierson v. Ray, 386 U.S. 547, 87 S.Ct. 1213, 18 L.Ed.2d 288 (1967); Bradley v. Fisher, 13 Wall. 335, 20 L.Ed. 646 (1872); Randall v. Brigham, 7 Wall. 523, 19 L.Ed. 285 (1869). We have had no occasion, however, to determine whether judicial immunity bars a § 1983 suit for prospective relief. See Supreme Court of Virginia v. Consumers Union of United States, Inc., 446 U.S. 719, 735, 100 S.Ct. 1967, 1976, 64 L.Ed.2d 641 (1980).1 It is clear that Congress did not limit the scope of common-law immunities in either § 19832 o§ 1988.3 We, therefore, have looked to the common law to determine when absolute immunity should be available. A review of the common law reveals nothing that suggests—much less requires—the distinction the Court draws today between suits for prospective relief (with the attendant liability for costs and attorney's fees) and suits for damages.

The doctrine of judicial immunity is one of the earliest products of the English common law.4 It was established to protect the finality of judgments from continual collateral attack in courts of competing jurisdiction5 and to protect judicial decisionmaking from intimidation and outside interference.6 Gradually, the protection of judicial independence became its primary objective. The specific source of intimidation articulated by the English common-law cases was the threat of vexatious litigation should judges be required to defend their judicial acts in collateral civil proceedings. In Taaffe v. Downes, supra, at 18, n. (a), the justices observed: "If you once break down the barrier . . . and subject [judges] to an action, you let in upon the judicial authority a wide, wasting, and harassing persecution. . . ." The common-law cases made no reference to the effect on judicial independence of particular remedies such as an award of damages.

The early opinions of this Court echo the principal justification for the immunity doctrine articulated at English common law. In Bradley v. Fisher, supra, the emphasis was on the burden of harassing and vexatious litigation. The Court observed:

"If . . . a judge could be compelled to answer in a civil action for his judicial acts, . . . he would be subjected for his protection to the necessity of preserving a complete record of all the evidence produced before him in every litigated case, and of the authorities cited and arguments presented, in order that he might be able to show to the judge before whom he might be summoned by the losing party . . . that he had decided as he did with judicial integrity; and the second judge would be subjected to a similar burden, as he in his turn might also be held amenable by the losing party." Id., at 349.

Addressing the need for judicial independence, the Court therefore concluded:

" 'The public are deeply interested in th[e] rule [of judicial immunity], which . . . was established in order to secure the independence of the judges, and prevent them being harassed by vexatious actions.' " Ibid. (quoting Fray v. Blackburn, 3 B. & S. 576, 578, 122 Eng.Rep. 217 (1863)).

The justification for the immunity doctrine emphasized in Bradley has been repeated in subsequent decisions by this Court. See, e.g., Pierson v. Ray, 386 U.S., at 554, 87 S.Ct., at 1217; Butz v. Economou, 438 U.S. 478, 512, 98 S.Ct. 2894, 2913, 57 L.Ed.2d 895 (1978). In these cases as well, the burdens of litigation, rather than the threat of pecuniary loss, are cited as posing a threat to judicial independence and occasioning the need for immunity. These burdens apply equally to all suits against judges for allegedly erroneous or malicious conduct. It is immaterial whether the relief sought is an injunction as in this case, or damages as in Pierson v. Ray or Stump v. Sparkman. Indeed, the Court today, largely ignoring that it was the burden of litigation that motivated the common-law immunity, makes no argument to the contrary. Unless the rationale of Bradley and the common-law cases is rejected, judicial immunity from suits against judges for injunctive relief must be coextensive with immunity from suits for damages.

The Court nevertheless argues that the common law of England can be viewed as supporting the absence of immunity where the suit is for injunctive relief. The Court concedes, as it must, that suits for injunctive relief against a judge could not be maintained either at English common law or in the English courts of equity. Ante, at 529. Injunctive relief from inequitable proceedings at common law was available in equity "to stay [a common-law] trial; or, after verdict, to stay judgment; or, after judgment, to stay execution." J. Story, Equity Jurisprudence ¶ 874, p. 72 (11th ed. 1873). But such relief was available only against the parties to the common-law proceedings and not against the judge. Id., ¶ 875, at 72. The suit for injunctive relief at issue here is precisely the type of suit that the Court concedes could not have been maintained either at common law or in equity. The Court, however, reasons that the writs of prohibition and mandamus present a "common-law parallel to the § 1983 injunction at issue here." Ante, at 529.

The prerogative writs of mandamus and prohibition are simply not analogous to suits for injunctive relief from the judgments of common-law courts, and the availability of these writs against judicial officials has nothing to do with judicial immunity. It has long been recognized at common law that judicial immunity protects only those acts committed within the proper scope of a judge's jurisdiction, but provides no protection for acts committed in excess of jurisdiction.7 Because writs of prohibition and mandamus were intended only to control the proper exercise of jurisdiction,8 they posed no threat to judicial independence and implicated none of the policies of judicial immunity. Thus, the judges of England's inferior courts were subject to suit for writs of mandamus and prohibition, but judicial immunity barred all suits attacking judicial decisions made within the proper scope of their jurisdiction.9 There is no allegation in this case that petitioner exceeded her jurisdiction. The suit for injunctive relief is based solely on an erroneous construction and application of law. It is precisely this kind of litigation that the common-law doctrine of judicial immunity was intended to prohibit.

The Court's observation that prerogative writs may have been used at English common law to correct errors of judgment rather than excesses of jurisdiction is irrelevant to the case at bar. We "rely on the common-law practice in shaping our own doctrine of judicial immunity," ante, at 536, only to the extent that the common-law practices consulted are consistent with our own judicial systems. The Court's reliance on English common-law practice ignores this constraint. It was the rivalry between the English temporal and spiritual courts that induced the King's Bench to adopt the myth that misapplication of substantive common law affects the court's jurisdiction.10 As the Court points out, the relationship between the King's Bench and its rival ecclesiastical courts finds no parallel in our judicial system. Ante, at 535. There is no indication that the courts of this country ever resorted to the fictional use of prerogative writs found at English common law. To the contrary, our courts expressly have rejected the fiction and have limited the use of mandamus and prohibition to jurisdictional issues or to cases where the court has a clear duty to act. See Roche v. Evaporated Milk Assn., 319 U.S. 21, 26, 63 S.Ct. 938, 941, 87 L.Ed. 1185 (1943). See also Bankers Life & Casualty Co. v. Holland, 346 U.S. 379, 382-383, 74 S.Ct. 145, 147-148, 98 L.Ed. 106 (1953); Will v. United States, 389 U.S. 90, 103-104, 88 S.Ct. 269, 278-279, 19 L.Ed.2d 305 (1967).

Nor is there any indication that the expansive use of prerogative writs in England modified the doctrine of judicial immunity in this country.11 Indeed, the sparing use of the writs of prohibition and mandamus in American jurisprudence has been motivated in large part by the concern for judicial independence. Cases counseling restraint in the use of prerogative writs repeatedly have observed that such writs have "the unfortunate consequence" of "plac[ing] trial judges in the anomalous position of being litigants without counsel other than uncompensated volunteers." La Buy v. Howes Leather Co., 352 U.S. 249, 258, 77 S.Ct. 309, 314, 1 L.Ed.2d 290 (1957). See also Kerr v. United States District Court, 426 U.S. 394, 402, 96 S.Ct. 2119, 2123, 48 L.Ed.2d 725 (1976); Bankers Life & Casualty Co., supra, at 384-385, 74 S.Ct., at 148-149; Ex parte Fahey, 332 U.S. 258, 259-260, 67 S.Ct. 1558, 1559, 91 L.Ed. 2041 (1947). In response to this concern, the Federal Rules of Appellate Procedure have provided that the respondent judge in a proceeding for mandamus or prohibition may elect not to appear in the proceeding without conceding the issues raised in the petition. Fed.Rule App.Proc. 21(b).12 Finally, courts consistently have held that concerns for judicial independence require that any award of costs to a prevailing party in an action for mandamus or prohibition be made only against the party at interest and not against the judge. The United States Court of Appeals for the First Circuit explained:

"It would be contrary to the fundamental rules protecting the freedom of judicial action to tax costs against a judge of any one of the constitutional courts of the United States by reason of any failure to apprehend the law correctly." In re Haight & Freese Co., 164 F. 688, 690 (1908).

Accord, Cotler v. Inter-County Orthopaedic Assn., 530 F.2d 536, 538 (CA3 1976).

In sum, the perceived analogy to the use of prerogative writs at English common law simply does not withstand analysis. As shown above, the analogy rests on a peculiar practice at English common law that was occasioned by the unique relationship between the King's Bench and England's ecclesiastical courts. That relationship finds no parallel in this country. Moreover, our courts, and the Federal Rules of Appellate Procedure, have sought to limit the use of mandamus and prohibition for the very purpose of protecting judicial immunity. It is extraordinary, therefore, that the Court today should rely on the use of prerogative writs in England to justify exposing judicial officials in this country to harassing litigation and to subject them to personal liability for money judgments in the form of costs and attorney's fees.

The Court suggests that the availability of injunctive relief under § 1983 poses no serious "risk that judges will be harassed and their independence compromised by the threat of having to defend themselves against suits by disgruntled litigants." Ante, at 537-538. The reasons advanced for this optimism are that equitable relief will be unavailable unless the plaintiff can show "an inadequate remedy at law and . . . a serious risk of irreparable harm." Ibid. Again, this suit refutes the Court's argument. Adequate remedies were expressly available to each of the respondents under state law.13 Nor was there any showing in this case of irreparable harm in the absence of injunctive relief. Nevertheless, petitioner was forced to bear the burdens of extended litigation, making clear the need for absolute judicial immunity.14

As discussed, both the English common-law cases and the decisions of this Court identify the burdens of harassing litigation, rather than the threat of pecuniary loss, as threatening judicial independence. In suits for injunctive relief, just as in suits for damages, the likely scenario was well stated by one of the justices in Taaffe v. Downes:

"[Without the doctrine of judicial immunity, judges] become amenable to every other species of correction by a Court. . . . One hour at the bar—the next at the bench, of the same or some other Court. They would have a busy and harassing time, getting from one station to the other—from the Judge to the accused—from the corrector to the corrected." 13 Eng.Rep., at 20, n. (a).

The ever-present threat of burdensome litigation, made realistic by today's decision, may well influence judicial determinations, particularly in close cases where the decision is likely to be unpopular.

Suits for injunctive relief may pose even greater threats to judicial independence if they are successful and an injunction is ordered. The specter of contempt proceedings for alleged violations of injunctive orders is likely to inhibit unbiased judicial decisionmaking as much as the threat of liability for damages. Again, this suit is a case in point. The injunctive order entered here was of unlimited duration and enjoined petitioner from authorizing the pretrial detention of any person charged with a certain class of misdemeanor, unless that person was "lawfully deemed likely to be a danger to himself or to others," and "only so long as such danger persists." App. 22. Whether a particular defendant is "likely to be a danger to himself or to others" and how "long [that danger will] last" are questions normally and necessarily left to the discretion of the presiding judge. The threat of contempt—with the possibility of a fine or even imprisonment—could well deter even the most courageous judge from exercising this discretion independently and free from intimidation.15

Finally, harassing litigation and its potential for intimidation increases in suits where the prevailing plaintiff is entitled to attorney's fees. Perhaps for understandable reasons, the Court's opinion passes lightly over the effect of § 1988. In fact, that provision has become a major additional source of litigation. Since its enactment in 1976, suits against state officials under § 1983 have increased geometrically.16 Congress enacted § 1988 for the specific purpose of facilitating and encouraging citizens of limited means to obtain counsel to pursue § 1983 remedies. But §§ 1983 and 1988 are available regardless of the financial ability of a plaintiff to engage private counsel. The lure of substantial fee awards,17 now routinely made to prevailing § 1983 plaintiffs, assures that lawyers will not be reluctant to recommend and press these suits.18 The Court again ignores reality when it suggests that the availability of injunctive relief under § 1983, combined with the prospect of attorney's fees under § 1988, poses no serious threat of harassing litigation with its potentially adverse consequences for judicial independence.

In sum, I see no principled reason why judicial immunity should bar suits for damages but not for prospective injunctive relief. The fundamental rationale for providing this protection to the judicial office—articulated in the English cases and repeated in decisions of this Court—applies equally to both types of asserted relief. The underlying principle, vital to the rule of law, is assurance of judicial detachment and independence. Nor is the Court's decision today in the broader public interest that the doctrine of absolute immunity is intended to serve. Bradley, 13 Wall., at 349.