with whom The Chief Justice, Mr. Justice Stewart, and Mr. Justice Stevens join, concurring in part and dissenting in part.
I concur in that part of the Court’s judgment which affords absolute immunity to those persons performing adjudicatory functions within a federal agency, ante, at 514, *518those who are responsible for the decision to initiate or continue a proceeding subject to agency adjudication, ante, at 516, and those agency personnel who present evidence on the record in the course of an adjudication, ante, at 517. I cannot agree, however, with the Court’s conclusion that in a suit for damages arising from allegedly unconstitutional action federal executive officials, regardless of their rank or the scope of their responsibilities, are entitled to only qualified immunity even when acting within the outer limits of their authority. The Court’s protestations to the contrary notwithstanding, this decision seriously misconstrues our prior decisions, finds little support as a matter of logic or precedent, and perhaps most importantly, will, I fear, seriously “dampen the ardor of all but the most resolute, or the most irresponsible, in the unflinching discharge of their duties,” Gregoire v. Biddle, 177 F. 2d 579, 581 (CA2 1949) (Learned Hand, J.).
Most noticeable is the Court’s unnaturally constrained reading of the landmark case of Spalding v. Vilas, 161 U. S. 483 (1896). The Court in that case did indeed hold that the actions taken by the Postmaster General were within the authority conferred upon him by Congress, and went on to hold that even though he had acted maliciously in carrying out the duties conferred upon him by Congress he was protected by official immunity. But the Court left no doubt that it would have reached the same result had it been alleged the official acts were unconstitutional.
“We are of the opinion that the same general considerations of public policy and convenience which demand for judges of courts of superior jurisdiction immunity from civil suits for damages arising from acts done by them in the course of the performance of their judicial functions, apply to a large extent to official communications made by heads of Executive Departments when engaged in the discharge of duties imposed upon them by law. The interests of the people require that due protection be *519accorded to them in respect of their official acts.” Id., at 498.
The Court today attempts to explain away that language by observing that Spalding indicated no intention to overrule Kendall v. Stokes, 3 How. 87 (1845), or Wilkes v. Dinsman, 7 How. 89 (1849). See ante, at 493 n. 18. But as the Court itself observes, the Postmaster General was held not “liable in an action for an error of judgment” in Kendall, supra, at 98. The Court in Wilkes, supra, likewise exonerated the defendant. The Court did indicate in dictum in both those cases that a federal officer might be liable if he acted with malice, Kendall, supra, at 99; Wilkes, supra, at 131, but the holding in Spalding was, as even the Court is forced to admit today, see ante, at 492-493, directly contrary to those cases on that point. In short, Spalding clearly and inescapably stands for the proposition that high-ranking executive officials acting within the outer limits of their authority are absolutely immune from suit.
Indeed, the language from Spalding quoted above unquestionably applies with equal force in the case at bar. No one seriously contends that the Secretary of Agriculture or the Assistant Secretary, who are being sued for $32 million in damages, had wandered completely off the official reservation in authorizing prosecution of respondent for violation of regulations promulgated by the Secretary for the regulation of “futures commission merchants,” 7 U. S. C. § 6 (1976 ed.). This is precisely what the Secretary and his assistants were empowered and required to do. That they would on occasion be mistaken in their judgment that a particular merchant had in fact violated the regulations is a necessary concomitant of any known system of administrative adjudication; that they acted “maliciously” gives no support to respondent’s claim against them unless we are to overrule Spalding.
The Court’s attempt to distinguish Spalding may be predi*520cated on a simpler but equally erroneous concept of immunity. At one point the Court observes that even under Spalding “an executive officer would be vulnerable if he took action 'manifestly or palpably’ beyond his authority or ignored a clear limitation on his enforcement powers.” Ante, at 493 n. 18. From that proposition, which is undeniably accurate, the Court appears to conclude that anytime a plaintiff can paint his grievance in constitutional colors, the official is subject to damages unless he can prove he acted in good faith. After all, Congress would never “authorize” an official to engage in unconstitutional conduct. That this notion in fact underlies the Court’s decision is strongly suggested by its discussion of numerous cases which supposedly support its position, but all of which in fact deal not with the question of what level of immunity a federal official may claim when acting within the outer limits of his authority, but rather with the question of whether he was in fact so acting. See ante, at 489-491.
Putting to one side the illogic and impracticability of distinguishing between constitutional and common-law claims for purposes of immunity, which will be discussed shortly, this sort of immunity analysis badly misses the mark. It amounts to saying that an official has immunity until someone alleges he has acted unconstitutionally. But that is no immunity at all: The “immunity” disappears at the very moment when it is needed. The critical inquiry in determining whether an official is entitled to claim immunity is not whether someone has in fact been injured by his action; that is part of the plaintiff’s case in chief. The immunity defense turns on whether the action was one taken “when engaged in the discharge of duties imposed upon [the official] by law,” Spalding, 161 U. S., at 498, or in other words, whether the official was acting within the outer bounds of his authority. Only if the immunity inquiry is approached in this manner does it have any meaning. That such a rule may occasionally result in individual injustices has never been doubted, but at least until *521today, immunity has been accorded nevertheless. As Judge Learned Hand said in Gregoire v. Biddle, 177 F. 2d, at 581:
“The justification for doing so is that it is impossible to know whether the claim is well founded until the case has been tried, and that to submit all officials, the innocent as well as the guilty, to the burden of a trial and to the inevitable danger of its outcome, would dampen the ardor of all but the most resolute, or the most irresponsible, in the unflinching discharge of their duties. Again and again the public interest calls for action which may turn out to be founded on a mistake, in the face of which an official may later find himself hard put to it to satisfy a j ury of his good faith. There must indeed be means of punishing public officers who have been truant to their duties; but that is quite another matter from exposing such as have been honestly mistaken to suit by anyone who has suffered from their errors. As is so often the case, the answer must be found in a balance between the evils inevitable in either alternative. In this instance it has been thought in the end better to leave unredressed the wrongs done by dishonest officers than to subject those who try to do their duty to the constant dread of retaliation. . . .”
Indeed, in that very case Judge Hand laid bare the folly of approaching the question of immunity in the manner suggested today by the Court.
“The decisions have, indeed, always imposed as a limitation upon the immunity that the official’s act must have been within the scope of his powers; and it can be argued that official powers, since they exist only for the public good, never cover occasions where the public good is not their aim, and hence that to exercise a power dishonestly is necessarily to overstep its bounds. A moment’s reflection shows, however, that that cannot be the meaning of the limitation without defeating the *522whole doctrine. What is meant by saying that the officer must be acting within his power cannot be more than that the occasion must be such as would have justified the act, if he had been using his power for any of the purposes on whose account it was vested in him. . . .” Ibid.
Barr v. Matteo, 360 U. S. 564 (1959), unfortunately fares little better at the Court’s hand than Spalding. Here the Court at least recognizes and reaffirms the minimum proposition for which Barr stands — that executive officials are absolutely immune at least from actions predicated on common-law claims as long as they are acting within the outer limits of their authority. See ante, at 495. Barr is distinguished, however, on the ground that it did not involve a violation of “those fundamental principles of fairness embodied in the Constitution.” Ibid. But if we allow a mere allegation of unconstitutionality, obviously unproved at the time made, to require a Cabinet-level official, charged with the enforcement of the responsibilities to which the complaint pertains, to lay aside his duties and defend such an action on the merits, the defense of official immunity will have been abolished in fact if not in form. The ease with which a constitutional claim may be pleaded in a case such as this, where a violation of statutory or judicial limits on agency action may be readily converted by any legal neophyte into a claim of denial of procedural due process under the Fifth Amendment, will assure that. The fact that the claim fails when put to trial will not prevent the consumption of time, effort, and money on the part of the defendant official in defending his actions on the merits. The result can only be damage to the “interests of the people,” Spalding, supra, at 498, which “require^] that due protection be accorded to [Cabinet officials] in respect of their official acts.”
It likewise cannot seriously be argued that an official will be less deterred by the threat of liability for unconstitutional *523conduct than for activities which might constitute a common-law tort. The fear that inhibits is that of a long, involved lawsuit and a significant money judgment, not the fear of liability for a certain type of claim. Thus, even viewing the question functionally — indeed, especially viewing the question functionally — the basis for a distinction between constitutional and common-law torts in this context is open to serious question. Even the logical justification for raising such a novel distinction is far from clear. That the Framers thought some rights sufficiently susceptible of legislative derogation that they should be enshrined in the Constitution does not necessarily indicate that the Framers likewise intended to establish an immutable hierarchy of rights in terms of their importance to individuals. The most heinous common-law tort surely cannot be less important to, or have less of an impact on, the aggrieved individual than a mere technical violation of a constitutional proscription.
The Court purports to find support for this distinction, and therefore this result, in the principles supposedly underlying Marbury v. Madison, 1 Cranch 137 (1803) and Bivens v. Six Unknown Fed. Narcotics Agents, 403 U. S. 388 (1971), and the fact that cognate state officials are not afforded absolute immunity for actions brought under 42 U. S. C. § 1983. Undoubtedly these rationales have some superficial appeal, but none withstands careful analysis. Marbury v. Madison, supra, leaves no doubt that the high position of a Government official does not insulate his actions from judicial review. But that case, like numerous others which have followed, involved equitable-type relief by way of mandamus or injunction. In the present case, respondent sought damages in the amount of $32 million. There is undoubtedly force to the argument that injunctive relief, in these cases where a court determines that an official defendant has violated a legal right of the plaintiff, sets the matter right only as to the future. But there is at least as much force to the argument *524that the threat of injunctive relief without the possibility of damages in the case of a Cabinet official is a better tailoring of the competing need to vindicate individual rights, on the one hand, and the equally vital need, on the other, that federal officials exercising discretion will be unafraid to take vigorous action to protect the public interest.
The Court also suggests in sweeping terms that the cause of action recognized in Bivens would be “ 'drained of meaning’ if federal officials were entitled to absolute immunity for their constitutional transgressions.” Ante, at 501. But Bivens is a slender reed' on which to rely when abrogating official immunity for Cabinet-level officials. In the first place, those officials most susceptible to claims under Bivens have historically been given only a qualified immunity. As the Court observed in Pierson v. Ray, 386 U. S. 547, 555 (1967), “[t]he common law has never granted police officers an absolute and unqualified immunity ....” In any event, it certainly does not follow that a grant of absolute immunity to the Secretary and Assistant Secretary of Agriculture requires a like grant to federal law enforcement officials. But even more importantly, on the federal side, when Congress thinks redress of grievances is appropriate, it can and generally does waive sovereign immunity, allowing an action directly against the United States. This allows redress for deprivations of rights, while at the same time limiting the outside influences which might inhibit an official in the free and considered exercise of his official powers. In fact, Congress, making just these sorts of judgments with respect to the very causes of action which the Court suggests require abrogation of absolute immunity, has amended the Federal Tort Claims Act, see 28 U. S. C. § 2680 (h) (1976 ed.), to allow suits against the United States on the basis of certain intentional torts if committed by federal “investigative or law enforcement officers.”
The Court also looks to the question of immunity of state officials for causes arising under § 1983 and, quoting a con*525curring opinion in Anderson v. Nosser, 438 F. 2d 183, 205 (CA5 1971), to the effect that there should not be “one law for Athens and another for Rome,” finds no reason why those principles should not likewise apply when federal officers are the target. Homilies cannot replace analysis in this difficult area, however. And even a moment’s reflection on the nature of the Bivens-type action and the purposes of § 1983, as made abundantly clear in this Court’s prior cases, supplies a compelling reason for distinguishing between the two different situations. In the first place, as made clear above, a grant of absolute immunity to high-ranking executive officials on the federal side would not eviscerate the cause of action recognized in Bivens. The officials who are the most likely defendants in a Bivens-type action have generally been accorded only a qualified immunity. But more importantly, Congress has expressly waived sovereign immunity for this type of suit. This permits a direct action against the Government, while limiting those risks which might “dampen the ardor of all but the most resolute, or the most irresponsible, in the unflinching discharge of their duties.” And the Federal Government can internally supervise and check its own officers. The Federal Government is not so situated that it can control state officials or strike this same balance, however. Hence the necessity of § 1983 and the differing standards of immunity. As the Court observed in District of Columbia v. Carter, 409 U. S. 418 (1973):
“Although there are threads of many thoughts running through the debates on the 1871 Act, it seems clear that § 1 of the Act, with which we are here concerned, was designed primarily in response to unwillingness or inability of the state governments to enforce their own laws against those violating the civil rights of others.” Id., at 426.
“[T]he [basic] rationale underlying Congress’ decision not to enact legislation similar to § 1983 with respect to *526federal officials [was] the assumption that the Federal Government could keep its own officers under control....” Id., at 429M30.
The Court attempts to avoid the force of this argument by suggesting that the statute which vests federal courts with general federal-question jurisdiction is basically the equivalent of § 1983. Ante, at 502 n. 30. But that suggestion evinces a basic misunderstanding of the difference between a statute which vests jurisdiction in federal courts, which are, as a constitutional matter, courts of limited jurisdiction, and a statute, or even a constitutional provision, which creates a private right of action. As even the Court’s analysis in Bivens made clear, a statute giving jurisdiction to federal courts does not, in and of itself, create a right of action. And to date, the Court has not held that the Constitution itself creates a private right of action for damages except when federal law enforcement officials arrest someone and search his premises in violation of the Fourth Amendment. Thus, the Court’s attempt to equate § 1983 and 28 U. S. C. § 1331 (1976 ed.) simply fails, and its further observation — that there should be no difference in immunity between state and federal officials— remains subject to serious doubt.
My biggest concern, however, is not with the illogic or impracticality of today’s decision, but rather with the potential for disruption of Government that it invites. The steady increase in litigation, much of it directed against governmental officials and virtually all of which could be framed in constitutional terms, cannot escape the notice of even the most casual observer. From 1961 to 1977, the number of cases brought in the federal courts under civil rights statutes increased from 296 to 13,113. See Director of the Administrative Office of the United States Courts Ann. Rep. 189, Table 11 (1977); Ann. Rep. 173, Table 17 (1976). It simply defies logic and common experience to suggest that officials will not have this in the back of their minds when considering *527what official course to pursue. It likewise strains credulity to suggest that this threat will only inhibit officials from taking action which they should not take in any event. It is the cases in which the grounds for action are doubtful, or in which the actor is timid, which will be affected by today’s decision.
The Court, of course, recognizes this problem and suggests two solutions. First, judges, ever alert to the artful pleader, supposedly will weed out insubstantial claims. Ante, at 507. That, I fear, shows more optimism than prescience. Indeed, this very case, unquestionably frivolous in the extreme, belies any hope in that direction. And summary judgment on affidavits and the like is even more inappropriate when the central, and perhaps only, inquiry is the official’s state of mind. See C. Wright, Law of Federal Courts 493 (3d ed. 1976) (It “is not feasible to resolve on motion for summary judgment cases involving state of mind”); Subin v. Goldsmith, 224 F. 2d 753 (CA2 1955).
The second solution offered by the Court is even less satisfactory. The Court holds that in those special circumstances “where it is demonstrated that absolute immunity is essential for the conduct of the public business,” absolute immunity will be extended. Ante, at 507. But this is a form of “absolute immunity” which in truth exists in name only. If, for example, the Secretary of Agriculture may never know until inquiry by a trial court whether there is a possibility that vexatious constitutional litigation will interfere with his deci-sionmaking process, the Secretary will obviously think not only twice but thrice about whether to prosecute a litigious commodities merchant who has played fast and loose with the regulations for his own profit. Careful consideration of the rights of every individual subject to his jurisdiction is one thing; a timorous reluctance to prosecute any of such individuals who have a reputation for using litigation as a defense weapon is quite another. Since Cabinet officials are mortal, *528it is not likely that we shall get the precise judgmental balance desired in each of them, and it is because of these very human failings that the principles of Spalding, 161U. S., at 498, dictate that absolute immunity be accorded once it be concluded by a court that a high-level executive official was “engaged in the discharge of duties imposed upon [him] by law.”*
Today’s opinion has shouldered a formidable task insofar as it seeks to justify the rejection of the views of the first Mr. Justice Harlan expressed in his opinion for the Court in Spalding v. Vilas, supra, and those of the second Mr. Justice Harlan expressed in his opinions in Barr v. Matteo, 360 U. S. 564 (1959), and its companion case of Howard v. Lyons, 360 U. S. 593 (1959). In terms of juridical jousting, if not in terms of placement in the judicial hierarchy, it has taken on at least as formidable a task when it disregards the powerful statement of Judge Learned Hand in Gregoire v. Biddle, 177 F. 2d 579 (CA2 1949).
*529History will surely not condemn the Court for its effort to achieve a more finely ground product from the judicial mill, a product which would both retain the necessary ability of public officials to govern and yet assure redress to those who are the victims of official wrongs. But if such a system of redress for official wrongs was indeed capable of being achieved in practice, it surely would not have been rejected by this Court speaking through the first Mr. Justice Harlan in 1896, by this Court speaking through the second Mr. Justice Harlan in 1959, and by Judge Learned Hand speaking for the Court of Appeals for the Second Circuit in 1948. These judges were not inexperienced' neophytes who lacked the vision or the ability to define immunity doctrine to accomplish that result had they thought it possible. Nor were they obsequious toadies in their attitude toward high-ranking officials of coordinate branches of the Federal Government. But they did see with more prescience than the Court does today, that there are inevitable trade-offs in connection with any doctrine of official liability and immunity.. They forthrightly accepted the possibility that an occasional failure to redress a claim of official wrongdoing would result from the doctrine of absolute immunity which they espoused, viewing it as a lesser evil than the impairment of the ability of responsible public officials to govern.
But while I believe that history will look approvingly on the motives of the Court in reaching the result it does today, I do not believe that history will be charitable in its judgment of the all but inevitable result of the doctrine espoused by the Court in this case. That doctrine seeks to gain and hold a middle ground which, with all deference, I believe the teachings of those who were at least our equals suggest cannot long be held. That part of the Court’s present opinion from which I dissent will, I fear, result in one of two evils, either one of which is markedly worse than the effect of according absolute immunity to the Secretary and the Assistant Secretary in this *530case. The first of these evils would be a significant impairment of the ability of responsible public officials to carry out the duties imposed upon them by law. If that evil is to be avoided after today, it can be avoided only by a necessarily unprincipled and erratic judicial “screening” of claims such as those made in this case, an adherence to the form of the law while departing from its substance. Either one of these evils is far worse than the occasional failure to award damages caused by official wrongdoing, frankly and openly justified by the rule of Spalding v. Vilas, Barr v. Matteo, and Gregoire v. Biddle.
The ultimate irony of today's decision is that in the area of common-law official immunity, a body of law fashioned and applied by judges, absolute immunity within the federal system is extended only to judges and prosecutors functioning in the judicial system. See Bradley v. Fisher, 13 Wall. 335 (1872); Yaselli v. Goff, 12 F. 2d 396 (CA2 1926), summarily aff’d, 275 U. S. 503 (1927). Similarly, where this Court has interpreted 42 U. S. C. § 1983 in the light of common-law doctrines of official immunity, again only judges and prosecutors are accorded absolute immunity. See Pierson v. Ray, 386 U. S. 547 (1967); Stump v. Sparkman, 435 U. S. 349 (1978); Imbler v. Pachtman, 424 U. S. 409 (1976). If one were to hazard an informed guess as to why such a distinction in treatment between judges and prosecutors, on the one hand, and other public officials on the other, obtains, mine would be that those who decide the common law know through personal experience the sort of pressures that might exist for such decisionmakers in the absence of absolute immunity, but may not know or may have forgotten that similar pressures exist in the case of nonjudicial public officials to whom difficult decisions are committed. But the cynical among us might not unreasonably feel that this is simply another unfortunate example of judges treating those who are not part of the judicial machinery as “lesser breeds without the law.”