with whom Mr. Justice Marshall joins, dissenting.
Over the course of the last decade, the deliberate-bypass standard announced in Fay v. Noia, 372 U. S. 391, 438-439 (1963), has played a central role in efforts by the federal judiciary to accommodate the constitutional rights of the individual with the States’ interests in the integrity of their judicial procedural regimes. The Court today decides that this standard should no longer apply with respect to procedural defaults occurring during the trial of a criminal defendant. In its place, the Court adopts the two-part “cause”- and-“prejudice” test originally developed in Davis v. United States, 411 U. S. 233 (1973), and Francis v. Henderson, 425 U. S. 536 (1976). As was true with these earlier cases,1 *100however, today’s decision makes no effort to provide concrete guidance as to the content of those terms. More particularly, left unanswered is the thorny question that must be recognized to be central to a realistic rationalization of this area of law: How should the federal habeas court treat a procedural default in a state court that is attributable purely and simply to the error or negligence of a defendant’s trial counsel? Because this key issue remains unresolved, I shall attempt in this opinion a re-examination of the' policies2 that should *101inform — and in Fay did inform — the selection of the standard governing the availability of federal habeas corpus jurisdiction in the face of an intervening procedural default in the state court.
I
I begin with the threshold question: What is the meaning and import of a procedural default? If it could be assumed that a procedural default more often than not is the product of a defendant’s conscious refusal to abide by the duly constituted, legitimate processes of the state courts, then I might agree that a regime of collateral review weighted in favor of a State’s procedural rules would be warranted.3 Fay, however, recognized that such rarely is the case; and therein lies Fay’s basic unwillingness to embrace a view of habeas jurisdiction that results in “an airtight system of [procedural] forfeitures.” 372 U. S., at 432.
This, of course, is not to deny that there are times when the failure to heed a state procedural requirement stems from an intentional decision to avoid the presentation of constitutional claims to the state forum. Fay was not insensitive to this possibility. Indeed, the very purpose of its bypass test is to detect and enforce such intentional procedural *102forfeitures of outstanding constitutionally based claims. Fay does so through application of the longstanding rule used to test whether action or inaction on the part of a criminal defendant should be construed as a decision to surrender the assertion of rights secured by the Constitution: To be an effective waiver, there must be “an intentional relinquishment or abandonment of a known right or privilege.” Johnson v. Zerbst, 304 U. S. 458, 464 (1938). Incorporating this standard, Fay recognized that if one “understandingly and knowingly forewent the privilege of seeking to vindicate his federal claims in the state courts, whether for strategic, tactical or any other reasons that can fairly be described as the deliberate by-passing of state procedures, then it is open to the federal court on habeas to deny him all relief . . ..” 372 U. S., at 439. For this reason, the Court’s assertion that it “think [s]” that the Fay rule encourages intentional “sandbagging” on the part of the defense lawyers is without basis, ante, at 89; certainly the Court points to no cases or commentary arising during the past 15 years of actual use of the Fay test to support this criticism. Rather, a consistent reading of case law demonstrates that the bypass formula has provided a workable vehicle for protecting the integrity of state rules in those instances when such protection would be both meaningful and just.4
*103But having created the bypass exception to the availability of collateral review, Fay recognized that intentional, tactical forfeitures are not the norm upon which to build a rational system of federal habeas jurisdiction. In the ordinary case, litigants simply have no incentive to slight the state tribunal, since constitutional adjudication on the state and federal levels are not mutually exclusive. Brown v. Allen, 344 U. S. 443 (1953); Brewer v. Williams, 430 U. S. 387 (1977); Castaneda v. Partida, 430 U. S. 482 (1977). Under the regime of collateral review recognized since the days of Brown v. Allen, and enforced by the Fay bypass test, no rational lawyer would risk the “sandbagging” feared by the Court.5 If a constitutional challenge is not properly raised *104on the state level, the explanation generally will be found elsewhere than in an intentional tactical decision.
In brief then, any realistic system of federal habeas corpus jurisdiction must be premised on the reality that the ordinary procedural default is born of the inadverténce, negligence, inexperience, or incompetence of trial counsel. See, e. g., Hill, The Inadequate State Ground, 65 Colum. L. Rev. 943, 997 (1965). The case under consideration today is typical. The Court makes no effort to identify a tactical motive for the failure of Sykes’ attorney to challenge the admissibility or reliability of a highly inculpatory statement. While my Brother Stevens finds a possible tactical advantage, I agree with the Court of Appeals that this reading is most implausible: “We can find no possible advantage which the defense might have gained, or thought they might gain, from the failure to conform with Florida Criminal Procedure Rule 3.190 (i).” 528 F. 2d 522, 527 (1976). Indeed, there is no basis for inferring that Sykes or his state trial lawyer was even aware of the existence of his claim under the Fifth Amendment; for this is not a case where the trial judge expressly drew the attention of the defense to a possible constitutional contention or procedural requirement, e. g., Murch v. Mottram, 409 U. S. 41 (1972); cf. Henry v. Mississippi, 379 U. S. 443, 448 n. 3 (1965), or where the defense signals its knowledge of a constitutional claim by abandoning a challenge previously raised, e. g., Sanders v. United States, 373 U. S. 1, *10518 (1963). Rather, any realistic reading of the record demonstrates that we are faced here with a lawyer’s simple error.6
Fay’s answer thus is plain: the bypass test simply refuses to credit what is essentially a lawyer’s mistake as a forfeiture of constitutional rights. I persist in the belief that the interests of Sykes and the State of Florida are best rationalized by adherence to this test, and by declining to react to inadvertent defaults through the creation of an “airtight system of forfeitures.’’
II
What are the interests that Sykes can assert in preserving the availability of federal collateral relief in the face of his inadvertent state procedural default? Two are paramount.
As is true with any federal habeas applicant, Sykes seeks access to the federal court for the determination of the validity of his federal constitutional claim. Since at least Brown v. Allen, it has been recognized that the “fair effect [of] the habeas corpus jurisdiction as enacted by Congress” entitles a state prisoner to such federal review. 344 U. S., at 500 (opinion of Frankfurter, J.). While some of my Brethren may feel uncomfortable with this congressional choice of policy, see, e. g., Stone v. Powell, 428 U. S. 465 (1976), the Legislative Branch nonetheless remains entirely free to determine that the constitutional rights of an individual subject to state custody, like those of the civil rights *106plaintiff suing under 42 U. S. C. § 1983, are best preserved by “interpos [ing] the federal courts between the States and the people, as guardians of the people’s federal rights . . . Mitchum v. Foster, 407 U. S. 225, 242 (1972).
With respect to federal habeas corpus jurisdiction, Congress explicitly chose to effectuate the federal court’s primary responsibility for preserving federal rights and privileges by authorizing the litigation of constitutional claims and defenses in a district court after the State vindicates its own interest through trial of the substantive criminal offense in the state courts.7 This, of course, was not the only course that Congress might have followed: As an alternative, it might well have decided entirely to circumvent all state procedure through the expansion of existing federal removal statutes such as 28 U. S. C. §§ 1442 (a)(1) and 1443, thereby authorizing the pretrial transfer of all state criminal cases to the federal courts whenever federal defenses or claims are in issue.8 But liberal post-trial federal review is the redress *107that Congress ultimately chose to allow and the consequences of a state procedural default should be evaluated in conformance with this policy choice. Certainly, we can all agree that once a state court has assumed jurisdiction of a criminal case, the integrity of its own process is a matter of legitimate concern. The Fay bypass test, by seeking to discover intentional abuses of the rules of the state forum, is, I believe, compatible with this state institutional interest. See Part III, infra. But whether Fay was correct in penalizing a litigant solely for his intentional forfeitures properly must be read in light of Congress’ desired norm of widened post-trial access to the federal courts. If the standard adopted today is later construed to require that the simple mistakes of attorneys are to be treated as binding forfeitures, it would serve to subordinate the fundamental rights contained in our constitutional charter to inadvertent defaults of rules promulgated by state agencies, and would essentially leave it to the States, through the enactment of procedure and the certification of the competence of local attorneys, to determine whether a habeas applicant will be permitted the access to the federal forum that is guaranteed him by Congress.9
*108Thus, I remain concerned that undue deference to local procedure can only serve to undermine the ready access to a federal court to which a state defendant otherwise is entitled. But federal review is not the full measure of Sykes' interest, for there is another of even greater immediacy: assuring that his constitutional claims can be addressed to some court. For the obvious consequence of barring Sykes from the federal courthouse is to insulate Florida’s alleged constitutional violation from any and all judicial review because of a lawyer’s mistake. From the standpoint of the habeas petitioner, it is a harsh rule indeed that denies him “any review at all where the state has granted none,” Brown v. Allen, 344 U. S., at 552 (Black, J., dissenting) — particularly when he would have enjoyed both state and federal consideration had his attorney not erred.
Fay’s answer to Sykes’ predicament, measuring the existence and extent of his procedural waiver by the Zerbst standard is, I submit, a realistic one. The Fifth Amendment assures that no person “shall be compelled in any criminal case to be a witness against himself . . . .” A defendant like Sykes can forgo this protection in two ways: He may decide to waive his substantive self-incrimination right at the point that he gives an inculpatory statement to the police authorities, Miranda v. Arizona, 384 U. S. 436, 478 (1966), or he and his attorney may choose not to challenge the admissibility of an incriminating statement when such a challenge would be effective under state trial procedure. See Estelle v. Williams, 425 U. S. 501, 524 (1976) (dissenting opinion). With few exceptions in the past 40 years, e. g., Estelle v. Williams, supra; Schneckloth v. Bustamonte, 412 U. S. 218 (1973), this Court has required that the substantive waiver, to be valid, must be a knowing and intelligent one. *109See, e. g., Brewer v. Williams, 430 U. S., at 404; Brookhart v. Janis, 384 U. S. 1, 4 (1966); Escobedo v. Williams, 378 U. S. 478, 490 n. 14 (1964); Green v. United States, 365 U. S. 184, 191-192 (1957); Smith v. United States, 337 U. S. 137, 149-150 (1949); Adams v. United States ex rel. McCann, 317 U. S. 269, 275 (1942). It has long been established that such is the case for the waiver of the protections of the Miranda rule. See 384 U. S., at 475; Schneckloth v. Bustamonte, supra, at 240. Fay simply evaluates the procedural waiver of Sykes’ Fifth Amendment rights by the same standard.
From the standpoint of the habeas petitioner this symmetry is readily understandable. To him, the inevitable consequence of either type of forfeiture — be it substantive or procedural — is that the protection of the Fifth Amendment is lost and his own words are introduced at trial to the prejudice of his defense. The defendant’s vital interest in preserving his Fifth Amendment privilege entitles him to informed and intelligent consideration of any decision leading to its forfeiture. It may be, of course, that the State’s countervailing institutional interests are more compelling in the case of eliciting a procedural default, thereby justifying a relaxation of the Zerbst standard. I discuss this possibility in greater detail in Part III, infra. It is sufficient for present purposes, however, that there is no reason for believing that this necessarily is true. That the State legitimately desires to preserve an orderly and efficient judicial process is undeniable. But similar interests of efficiency and the like also can be identified with respect to other state institutions, such as its law enforcement agencies. Yet, as was only recently reconfirmed, we would not permit and have not permitted the state police to enhance the orderliness and efficiency of their law enforcement activities by embarking on a campaign of acquiring inadvertent waivers of important constitutional rights. Brewer v. Williams, supra, at 401-406; see generally Francis v. Henderson, 425 U. S., at 548-549, n. 2 (dissenting opinion).
*110A procedural default should be treated accordingly. Indeed, a recent development in the law of habeas corpus suggests that adherence to the deliberate-bypass test may be more easily justified today than it was when Fay was decided. It also suggests that the “prejudice” prong of the Court’s new test may prove to be a redundancy. Last Term the Court ruled that alleged violations of the Fourth Amendment in most circumstances no longer will be cognizable in habeas corpus. Stone v. Powell, 428 U. S. 465 (1976). While, for me, the principle that generated this conclusion was not readily apparent, I expressed my concern that the Stone decision contains the seeds for the exclusion from collateral review of a variety of constitutional rights that my Brethren somehow deem to be unimportant — perhaps those that they are able to conclude are not “guilt-related.” See id., at 517-518 (dissenting opinion). If this trail is to be followed, it would be quite unthinkable that an unintentional procedural default should be allowed to stand in the way of vindication of constitutional rights bearing upon the guilt or innocence of a defendant. Indeed, if as has been argued, a key to decision in this area turns upon a comparison of the importance of the constitutional right at stake with the state procedural rule, Sandalow, Henry v. Mississippi and the Adequate State Ground: Proposals for a Revised Doctrine, 1965 Sup. Ct. Rev. 187, 236-237, then the Court’s threshold effort to identify those rights of sufficient importance to be litigated collaterally should largely predetermine the outcome of this balance.
In sum, I believe that Fay’s commitment to enforcing intentional but not inadvertent procedural defaults offers a realistic measure of protection for the habeas corpus petitioner seeking federal review of federal claims that were not litigated before the State. The threatened creation of a more “airtight system of forfeitures” would effectively deprive habeas petitioners of the opportunity for litigating *111their constitutional claims before any forum and would disparage the paramount importance of constitutional rights in our system of government. Such a restriction of habeas corpus jurisdiction should be countenanced, I submit, only if it fairly can be concluded that Fay’s focus on knowing and voluntary forfeitures unduly interferes with the legitimate interests of state courts or institutions. The majority offers no suggestion that actual experience has shown that Fay’s bypass test can be criticized on this score. And, as I now hope to demonstrate, any such criticism would be unfounded.
Ill
A regime of federal habeas corpus jurisdiction that permits the reopening of state procedural defaults does not invalidate any state procedural rule as such;10 Florida’s courts remain entirely free to enforce their own rules as they choose, and to deny any and all state rights and remedies to a defendant who fails to comply with applicable state procedure. The relevant inquiry is whether more is required — specifically, whether the fulfillment of important interests of the State necessitates that federal courts be called upon to impose additional sanctions 'for inadvertent noncompliance with state procedural requirements such as the contemporaneous-objection rule involved here.
*112Florida, of course, can point to a variety of legitimate interests in seeking allegiance to its reasonable procedural requirements, the contemporaneous-objection rule included. See Henry v. Mississippi, 379 U. S., at 448. As Fay recognized, a trial, like any organized activity, must conform to coherent process, and “there must be sanctions for the flouting of such procedure.” 372 U. S., at 431. The strict enforcement of procedural defaults, therefore, may be seen as a means of deterring any tendency on the part of the defense to slight the state forum, to deny state judges their due opportunity for playing a meaningful role in the evolving task of constitutional adjudication, or to mock the needed finality of criminal trials. All of these interests are referred to by the Court in various forms.11
The question remains, however, whether any of these policies or interests are efficiently and fairly served by enforcing both intentional and inadvertent defaults pursuant to the identical stringent standard. I remain convinced that when one pierces the surface justifications for a harsher rule posited by the Court, no standard stricter than Fay’s deliberate-bypass test is realistically defensible.
*113Punishing a lawyer’s unintentional errors by closing the federal courthouse door to his client is both a senseless and misdirected method of deterring the slighting of state rules. It is senseless because unplanned and unintentional action of any kind generally is not subject to deterrence; and, to the extent that it is hoped that a threatened sanction addressed to the defense will induce greater care and caution on the part of trial lawyers, thereby forestalling negligent conduct or error, the potential loss of all valuable state remedies would be sufficient to this end.12 And it is a misdirected sanction because even if the penalization of incompetence or carelessness will encourage more thorough legal training and trial preparation, the habeas applicant, as opposed to his lawyer, hardly is the proper recipient of such a penalty. Especially with fundamental constitutional rights *114at stake, no fictional relationship of principal-agent or the like can justify holding the criminal defendant accountable for the naked errors of his attorney.13 This is especially true when so many indigent defendants are without any realistic choice in selecting who ultimately represents them at trial.14 Indeed, if responsibility for error must be apportioned between the parties, it is the State, through its attorney’s admissions and certification policies, that is more fairly held to blame for the fact that practicing lawyers too often are ill-prepared or ill-equipped to act carefully and knowledgeably when faced with decisions governed by state procedural requirements.
*115Hence, while I can well agree that the proper functioning of our system of criminal justice, both federal and state, necessarily places heavy reliance on the professionalism and judgment of trial attorneys, I cannot accept a system that ascribes the absolute forfeiture of an individual’s constitutional claims to situations where his lawyer manifestly exercises no professional judgment at all — where carelessness, mistake, or ignorance is the explanation for a procedural default. Of course, it is regrettable that certain errors that might have been cured earlier had trial counsel acted expeditiously must be corrected collaterally and belatedly. I can understand the Court’s wistfully wishing for the day when the trial was the sole, binding and final “event” of the adversarial process — although I hesitate to agree that in the eyes of the criminal defendant it has ever ceased being the “main” one, ante, at 90. But it should be plain that in the real world, the interest in finality is repeatedly compromised in numerous ways that arise with far greater frequency than do procedural defaults. The federal criminal system, to take one example, expressly disapproves of interlocutory review in the generality of cases even though such a policy would foster finality by permitting the authoritative resolution of all legal and constitutional issues prior to the convening of the “main event.” See generally Abney v. United States, 431 U. S. 651 (1977). Instead, it relies on the belated correction of error, through appeal and collateral review, to ensure the fairness and legitimacy of the criminal sanction. Indeed, the very existence of the well-established right collaterally to reopen issues previously litigated before the state courts, Brown v. Allen, 344 U. S. 443 (1953), represents a congressional policy choice that is inconsistent with notions of strict finality — and probably more so than authorizing the litigation of issues that, due to inadvertence, were never addressed to any court. Ultimately, all of these limitations on the finality of criminal convictions emerge from the tension between justice *116and efficiency in a judicial system that hopes to remain true to its principles and ideals. Reasonable people may disagree on how best to resolve these tensions. But the solution that today’s decision risks embracing seems to me the most unfair of all: the denial of any judicial consideration of the constitutional claims of a criminal defendant because of errors made by his attorney which lie outside the power of the habeas petitioner to prevent or deter and for which, under no view of morality or ethics, can he be held responsible.
In short, I believe that the demands of our criminal justice system warrant visiting the mistakes of a trial attorney on the head of a habeas corpus applicant only when we are convinced that the lawyer actually exercised his expertise and judgment in his client’s service, and with his client’s knowing and intelligent participation where possible. This, of course, is the precise system of habeas review established by Fay v. Noia.
IV
Perhaps the primary virtue of Fay is that the bypass test at least yields a coherent yardstick for federal district courts in rationalizing their power of collateral review. See n. 4, supra. In contrast, although some four years have passed since its introduction in Davis v. United States, 411 U. S. 233 (1973), the only thing clear about the Court’s “cause”-and-“prejudice” standard is that it exhibits the notable tendency of keeping prisoners in jail without addressing their constitutional complaints. Hence, as of today, all we know of the “cause” standard15 is its requirement that habeas applicants bear an undefined burden of explanation for the failure to obey the state rule, ante, at 91. Left unresolved is whether a habeas petitioner like Sykes can adequately discharge this burden by *117offering the commonplace and truthful explanation for his default: attorney ignorance or error beyond the client’s control. The “prejudice” inquiry, meanwhile, appears to bear a strong resemblance to harmless-error doctrine. Compare ante, at 91, with Chapman v. California, 386 U. S. 18, 24 (1967). I disagree with the Court’s appraisal of the harmlessness of the admission of respondent’s confession, but if this is what is meant by prejudice, respondent’s constitutional contentions could be as quickly and easily disposed of in this regard by permitting federal courts to reach the merits of his complaint. In the absence of a persuasive alternative formulation to the bypass test, I would simply affirm the judgment of the Court of Appeals and allow Sykes his day in court on the ground that the failure of timely objection in this instance was not a tactical or deliberate decision but stemmed from a lawyer’s error that should not be permitted to bind his client.
One final consideration deserves mention. Although the standards recently have been relaxed in various jurisdictions,16 it is accurate to assert that most courts, this one included,17 traditionally have resisted any realistic inquiry into the competency of trial counsel. There is nothing unreasonable, *118however, in adhering to the proposition that it is the responsibility of a trial lawyer who takes on the defense of another to be aware of his client's basic legal rights and of the legitimate rules of the forum in which he practices his profession.18 If he should unreasonably permit such rules to bar the assertion of the colorable constitutional claims of his client, then his conduct may well fall below the level of competence that can fairly be expected of him.19 For almost 40 years it has been established that inadequacy of counsel undercuts the very competence and jurisdiction of the trial court and is always open to collateral review. Johnson v. Zerbst, 304 U. S. 458 (1938).20 Obviously, as a practical matter, a trial counsel cannot procedurally waive his own inadequacy. If the scope of habeas jurisdiction previously governed by Fay v. Noia is to be redefined so as to enforce the errors and neglect of lawyers with unnecessary and unjust rigor, the time may come when conscientious and fairminded federal and state courts, in adhering to the teaching of Johnson v. Zerbst, will have to reconsider whether they can continue to indulge the comfortable fiction that all lawyers are skilled or even competent craftsmen in representing the fundamental rights of their clients.
The Court began its retreat from the deliberate-bypass standard of Fay in Davis v. United States, where a congressional intent to restrict *100the bypass formulation with respect to collateral review under 28 U. S. C. § 2256 was found to inhere in Fed. Rule Crina. Proc. 12 (b) (2). By relying upon Congress’ purported intent, Davis managed to evade any consideration of the justifications and any shortcomings of the bypass test. Subsequently, in Francis v. Henderson, a controlling congressional expression of intent no longer was available, and the Court therefore employed the shibboleth of “considerations of comity and federalism” to justify application of Davis to a § 2254 proceeding. 425 U. S., at 541. Again, any coherent analysis of the bypass standard or the waivability of constitutional rights was avoided — as it was that same day in Estelle v. Williams, 425 U. S. 501 (1976), which proceeded to find a surrender of a constitutional right in an opinion that was simply oblivious to some 40 years of existing case law. See infra, at 108-109. Thus, while today’s opinion follows from Davis, Francis, and Estelle, the entire edifice is a mere house of cards whose foundation has escaped any systematic inspection.
1 use the term “policies” advisedly, for it is important to recognize the area of my disagreement with the Court. This Court has never taken issue with the foundation principle established by Fay v. Noia — that in considering a petition for the writ of habeas corpus, federal courts possess the power to look beyond a state procedural forfeiture in order to entertain the contention that a defendant’s constitutional rights have been abridged. 372 U. S., at 398-399. Indeed, only last Term, the Court reiterated: “There can be no question of a federal district court’s power to entertain an application for a writ of habeas corpus in a case such as this.” Francis v. Henderson, 425 U. S., at 538. Today’s decision reconfirms this federal power by authorizing federal intervention under the “cause”-and-"prejudice” test. Were such power unavailable, federal courts would be bound by Fla. Rule Crim. Proc. 3.190, which contains no explicit provision for relief from procedural defaults. Our disagreement, there*101fore, centers upon the standard that should govern a federal district court in the exercise of this power to adjudicate the constitutional claims of a state prisoner — which, in turn, depends upon an evaluation of the competing policies and values served by collateral review weighted against those furthered through strict deference to a State’s procedural rules.
It is worth noting that because we deal with the standards governing the exercise of the conceded power of federal habeas courts to excuse a state procedural default, Congress, as the primary expositor of federal-court jurisdiction, remains free to undo the potential restrictiveness of today’s decision by expressly defining the standard of intervention under 28 U. S. C. § 2254. Cf. Davis v. United States, 411 U. S., at 241-242.
Even this concession to procedure would, in my view, be unnecessary so long as the habeas court is capable of distinguishing between intentional and inadvertent defaults with acceptable accuracy — as I believe it can. See n. 4, infra.
Over the years this Court has without notable difficulty applied the Fay rule to a variety of contexts. E. g., Lefkowitz v. Newsome, 420 U. S. 283 (1975); Humphrey v. Cady, 405 U. S. 504, 517 (1972); Anderson v. Nelson, 390 U. S. 523, 525 (1968); Warden v. Hayden, 387 U. S. 294, 297 n. 3 (1967); cf. Chambers v. Mississippi, 410 U. S. 284, 290 n. 3 (1973). Similarly, the standard has been capable of intelligent application by the lower federal courts in order to bar the collateral reconsideration of tactical decisions by the defense, e. g., United States ex rel. Green v. Bundle, 452 F. 2d 232, 236 (CA3 1971) (counsel concedes tactical decision) ; Whitney v. United States, 513 F. 2d 326, 329 (CA8 1974) (counsel forgoes challenge to seized evidence in order to avoid concession of any possessory interest in searched premises), while otherwise permitting federal review, Henderson v. Kibbe, 534 F. 2d 493, 496-497 (CA2 1976), rev’d *103on other grounds, 431 U. S. 145 (1977); Paine v. McCarthy, 527 F. 2d 173 (CA9 1975). And in cases similar to the present one where Fifth Amendment violations were in issue, Fay has afforded a meaningful standard governing the scope of federal collateral review. Compare United States ex rel. Terry v. Henderson, 462 F. 2d 1125, 1129 (CA2 1972) (bypass found where counsel relied on confession to rebut premeditation in murder trial); and United States ex rel. Cruz v. LaVallee, 448 F. 2d 671, 679 (CA2 1971) (bypass found where trial strategy called for confessing to killing but arguing that mitigating circumstances exist), with Moreno v. Beto, 415 F. 2d 154 (CA5 1969) (defense not held to bypass where defense counsel deliberately chose not to raise and submit voluntariness issue to jury due to unwillingness to expose client to unconstitutional procedure).
In brief, the defense lawyer would face two options: (1) He could elect to present his constitutional claims to the state courts in a proper fashion. If the state trial court is persuaded that a constitutional breach has occurred, the remedies dictated by the Constitution would be imposed, the defense would be bolstered, and the prosecution accordingly weakened, perhaps precluded altogether. If the state court rejects the properly tendered claims, the defense has lost nothing: Appellate review before the state courts and federal habeas consideration are preserved. (2) He could elect to “sandbag.” This presumably means, first, that he would hold back the presentation of his constitutional claim to the trial court, thereby increasing the likelihood of a conviction since the prosecution would be able to present evidence that, while arguably constitutionally deficient, may be highly prejudicial to the defense. Second, he would thereby have forfeited all state review and remedies with respect to these claims (sub*104ject to whatever “plain error” rule is available). Third, to carry out his scheme, he would now be compelled to deceive the federal habeas court and to convince the judge that he did not "deliberately bypass” the state procedures. If he loses on this gamble, all federal review would be barred, and his “sandbagging” would have resulted in nothing but the forfeiture of all judicial review of his client's claims. The Court, without substantiation, apparently believes that a meaningful number of lawyers are induced into option 2 by Fay. I do not. That belief simply offends common sense.
The likelihood that we are presented with a lawyer’s simple mistake is not answered by respondent’s stipulation to his trial counsel’s competency. At oral argument it was made clear that Sykes so stipulated solely because of the position expressed by the habeas court that a challenge to his prior legal representation would require the return to the state courts and the further exhaustion of state remedies, a detour that respondent insisted on avoiding. Tr. of Oral Arg. 49. Furthermore, in light of the prevailing standards, or lack of standards, for judging the competency of trial counsel, see infra, at 117, it is perfectly consistent for even a lawyer who commits a grievous error — whether due to negligence or ignorance — to be deemed to have provided competent representation.
Congress’ grant of post-trial access to the federal courts was reconfirmed by its modification of 28 U. S. C. § 2254 following our decisions in Fay and Townsend v. Sain, 372 U. S. 293 (1963). This legislative amendment of the habeas statute essentially embraced the relitigation standards outlined in Townsend without altering the broad framework for collateral review contained in Brown v. Allen, 344 U. S. 443 (1953), Fay, and like eases. See, e. g., Stone v. Powell, 428 U. S. 465, 528-529 (1976) (Brennan, J., dissenting).
Whether in a civil or criminal case, Congress’ broad authority to allocate federal issues for decision in its choice of forum is clear. See, e. g., Tennessee v. Davis, 100 U. S. 257 (1880); Greenwood v. Peacock, 384 U. S. 808, 833 (1966): “We have no doubt that Congress, if it chose, could provide for exactly such a system. We may assume that Congress has constitutional power to provide that all federal issues be tried in the federal courts, that all be tried in the courts of the States, or that jurisdiction of such issues be shared. And in the exercise of that power, we may assume that Congress is constitutionally fully free to establish the conditions under which civil or criminal proceedings involving federal issues may be removed from one court to another.” The same day as Greenwood the Court applied § 1443 (1) as authorizing (subject to further fact-*107finding) the removal of a state trespass prosecution to the United States District Court. Georgia v. Rachel, 384 U. S. 780 (1966). Once a criminal case is thus removed to the federal court, the State no longer can assert any interest in having trial of the state substantive offense governed by the State’s choice of procedure, for this Court has long provided that federal procedure then- obtains. Tennessee v. Davis, supra, at 272. In this sense, the prevailing system of post-trial federal collateral review is more generous to state procedure than would be required, and, some would say, desired. See generally Amsterdam, Criminal Prosecutions Affecting Federally Guaranteed Civil Rights: Federal Removal and Habeas Corpus Jurisdiction to Abort State Court Trial, 113 U. Pa. L. Rev. 793 (1965).
Of course, even under the Court’s new standard, traditional principles continue to apply, and the federal judiciary is not bound by state rules of procedure that are unreasonable on their face, or that are either unrea*108sonably or inconsistently applied. See, e. g., Henry v. Mississippi, 379 U. S. 443 (1965); NAACP v. Alabama, 377 U. S. 288 (1964); Staub v. Baxley, 355 U. S. 313 (1958); Williams v. Georgia, 349 U. S. 375 (1955).
This is no-t to suggest that the availability of collateral review has no bearing on the States’ selection and enforcement of procedural requirements. On the contrary, to the extent that a State desires to have input into the process of developing federal law, and seeks to guarantee its primary factfinding role as authorized by § 2254 and Townsend v. Sain, 372 U. S. 293 (1963), the existence of broad federal habeas power will tend to encourage the liberalizing and streamlining of state rules that otherwise might serve to bar such state participation. From every perspective, I would suppose that any such effect of Fay would be considered a salutary one, see, e. g., Shapiro, Federal Habeas Corpus: A Study in Massachusetts, 87 Harv. L. Rev. 321, 348 (1973), although the Court implies the contrary, ante, at 89-90.
In my view, the strongest plausible argument for strict enforcement of a contemporaneous-objection rule is one that the Court barely relies on at all: the possibility that the failure of timely objection to the admissibility of evidence may foreclose the making of a fresh record and thereby prejudice the prosecution in later litigation involving that evidence. There may be force to this contention, but it rests on the premise that the State in fact has suffered actual prejudice because of a procedural lapse. Florida demonstrates no such injury here. Sykes’ trial occurred in June 1972. He subsequently filed his petition for a writ of habeas corpus in April 1973, thereby apprising Florida of his constitutional objection. There is no basis in the record for concluding that lost evidence or other form of prejudice, see, e. g., Barker v. Wingo, 407 U. S. 514, 532 (1972), arising during this 10%-month interval effectively forestalls Florida’s defense of the Fifth Amendment claim or the reprosecution of Sykes should his constitutional challenge prevail.
Under § 2254, the availability of federal review is not limited or dependent on forgoing litigation in the state courts. Because the state forum thus affords purely an additional measure of protection, Fay recognized: “A man under conviction for crime has an obvious inducement to do his very best to keep his state remedies open, and not stake his all on the outcome of a federal habeas proceeding which, in many respects, may be less advantageous to him than a state court proceeding. . . . And if because of inadvertence or neglect he runs afoul of a state procedural requirement, and thereby forfeits his state remedies, appellate and collateral, as well as direct review thereof in this Court, those consequences should be sufficient to vindicate the State’s valid interest in orderly procedure.” 372 U. S., at 433. See Reitz, Federal Habeas Corpus: Impact of an Abortive State Proceeding, 74 Harv. L. Rev. 1315, 1351 (1961). This Court’s recent decision in Stone v. Powell, 428 U. S. 465 (1976), seems to subscribe to a similar view that deterrence is not meaningfully furthered by adopting an overkill of sanctions. There the Court reasoned that police misconduct under the Fourth Amendment will be deterred by state review of any search-and-seizure claim, and that further federal-court consideration would have but an “incremental” and “isolated” deterrent impact. Id., at 494. Assuming that criminal defendants and lawyers are no less rational than police, they should be deterred from risking the unnecessary forfeiture of all state remedies and the initial opportunity for judicial victory before the state courts.
Traditionally, the rationale for binding a criminal defendant by his attorney’s mistakes has rested on notions akin to agency law. See, e. g., Comment, Criminal Waiver: The Requirements of Personal Participation, Competence and Legitimate State Interest, 54 Calif. L. Rev. 1262, 1278-1281 (1966). With respect to ordinary commercial matters, the common law established and recognized principal-agent relationships for the protection of innocent third parties who deal with the latter. In the context of a criminal trial, this analogy is not apt, for the State, primarily in control of the criminal process and responsible for qualifying and assigning attorneys to represent the accused, is not a wholly innocent bystander. Consequently, the dominant relationship of the trial counsel with respect to his client more recently has been found simply to inhere in “our legal system” or “our adversary system.” Estelle v. Williams, 425 U. S., at 512. There is undoubted truth in this; obviously “our legal system” presupposes that attorneys will function competently, that their clients cannot participate in all decisions, Henry v. Mississippi, 379 U. S., at 451, and that the trial of a criminal defendant will not inevitably be followed by a trial of his attorney’s performance. Fay reacts to this institutional demand by enforcing both action and inaction of attorneys — even if they prove to backfire in actual practice — provided that it is found that the lawyer was aware of his client’s rights and knowingly applied his professional judgment in his client’s behalf. In brief, the bypass test rightfully defers to the attorney’s “vast array of trial decisions, strategic and tactical,” Estelle v. Williams, supra, at 512, but not to sheer inadvertence where no decision was made.
See generally Tague, An Indigent’s Right to the Attorney of His Choice, 27 Stan. L. Rev. 73 (1974).
The earlier cases of Davis v. United States, 411 U. S. 233 (1973), and Francis v. Henderson, 425 U. S., at 542, similarly are not instructive in defining “cause,” since both decisions appear to have disposed of the habeas application primarily on the “prejudice” aspect of the test.
A majority of courts have now passed beyond the standard of attorney competence embodied in the so-called “mockery” test, which abdicates any judicial supervision over attorney performance so long as the attorney does not make a farce of the trial. See, e. g., United States v. Katz, 425 F. 2d 928 (CA2 1970) (attorney who was prone to fall asleep during trial held to have provided competent representation). The new emerging rule essentially requires that the attorney provide assistance within a reasonable range of professional competence, see United States v. DeCoster, 159 U. S. App. D. C. 326, 487 F. 2d 1197 (1973); United States ex rel. Williams v. Twomey, 510 F. 2d 634, 640, (CA7 1975).
See, e. g., Chambers v. Maroney, 399 U. S. 42, 55-60 (1970) (Harlan, J., dissenting). Recently, this Court, however, has made clear that attorneys are expected to perform “within the range of competence demanded of attorneys in criminal cases,” McMann v. Richardson, 397 U. S. 759, 771 (1970); Tolleti v. Henderson, 411 U. S. 258, 266 (1973).
Indeed, at least this level of knowledge and proficiency would seem to be a prerequisite for the provision of “effective and substantial aid” as guaranteed by the Sixth Amendment. Powell v. Alabama, 287 U. S. 45, 53 (1932).
“Counsel’s failure to evaluate properly facts giving rise to a constitutional claim, or his failure properly to inform himself of facts that would have shown the existence of a constitutional claim, might in particular fact situations meet this standard of proof [of incompetent counsel].” Tollett v. Henderson, supra, at 266-267.
Zerbst dealt specifically with an instance where trial counsel was altogether lacking, but “[i]t has long been recognized that the right to counsel is the right to the effective assistance of counsel.” McMann v. Richardson, supra, at 771 n. 14 (citations omitted).