 |
|||||
|
|
|
|
|
||
|
|
|||||
|
|
|
|
|
||
GONZALEZ v. CROSBY
United States Supreme Court No. 04-6432
Can Rule 60(b) of the Federal Rules of Civil Procedure be used to reopen a habeas corpus case after final disposition by the Court of Appeals, when (1) the petitioner cannot meet the criteria for a successive petition under AEDPA, and (2) the state-court judgment is not a miscarriage of justice under the pre-AEDPA case law standard?
ORAL ARGUMENT DATE: April 25, 2005
Summary of facts and case
Summary of argumentI. The Federal Rules of Civil Procedure do not automatically apply to habeas cases
II. FRCP 60(b) does not apply to habeas cases where its effect is to evade the successive petition statute
A. Civil Rule 60
B. The successive petition ruleIII. The state's legitimate interest in finality of the judgment is the starting point for determining when use of Rule 60(b) is contrary to the spirit of AEDPA
A. Finality and AEDPA
B. Applying Harris to Rule 60(b)
1. Mistake, inadvertence, surprise, and
excusable neglect
2. Newly discovered evidence
3. Fraud on the court
4. Void judgment
5. Satisfaction, reversed prior judgment, prospective application
6. Catch-all 22IV. There is no constitutional doubt in this case
V. Use of Rule 60(b) in the present case would be contrary to the spirit of AEDPA
IN THE |
SUPREME COURT OF THE UNITED STATES |
|
|
| Aurelio O. Gonzalez, |
|
Petitioner,
|
| vs. |
|
Florida Department of Corrections, |
|
Respondent.
|
|
|
BRIEF AMICUS CURIAE OF THE
|
The Criminal Justice Legal Foundation (CJLF) (1) is a non-profit California corporation organized to participate in litigation relating to the criminal justice system as it affects the public interest. CJLF seeks to bring the constitutional protections of the accused into balance with the rights of the victim and of society to rapid, efficient, and reliable determination of guilt and swift execution of punishment.
This case involves an attempt to evade Congress's landmark reform of habeas corpus law in the Antiterrorism and Effective Death Penalty Act of 1996 ("AEDPA"). This law, if properly implemented, will greatly reduce unnecessary delay in the enforcement of capital punishment and reduce the number of correct criminal judgments erroneously overturned on federal habeas corpus. These changes would advance the rights of victims and society which CJLF was formed to protect.
In 1982, petitioner Aurelio Gonzalez pleaded guilty to multiple felonies: three counts of robbery with a firearm, one of armed burglary, and one of armed kidnapping. Brief in Opp. 2. For these crimes, he was sentenced to 99 years in prison. He did not appeal, but many years later filed a postconviction relief motion, denied on September 14, 1994. See ibid. Florida has a two-year time limitation, with exceptions, and a limit on successive motions. See Fla. Rule Crim. Proc. 3.850(b), (f). Gonzalez filed a successive motion in November 1996, claiming newly discovered evidence regarding the validity of his guilty plea. This motion was denied. Brief for Petitioner 2. The state appellate court affirmed, and it denied rehearing on May 8, 1997. See ibid.
Petitioner filed his federal habeas petition in June 1997. See Brief for Petitioner 3 (June 17); Brief in Opp. 3 (June 23). The precise date is not material. See Brief in Opp. 3, n. 3. The state asserted both procedural default and statute of limitations, 28 U. S. C. §2244(d), and the District Court dismissed on statute of limitations on Sept. 9, 1998. Brief in Opp. 3. A certificate of appealability (COA) was eventually denied and the appeal dismissed on April 6, 2000.
Five months later, this Court decided Artuz v. Bennett, 531 U. S. 4 (2000), which clarified some aspects of the tolling provision of §2244(d)(2). Bennett did not, however, directly address whether a time-barred state petition is "properly filed" so as to toll the federal statute, an issue presently before this Court in Pace v. DiGuglielmo, No. 03-9627. The parties in this case continue to dispute whether the original dismissal was correct. See Brief in Opp. 4.
On July 30, 2001, almost three years after the original District Court decision, petitioner filed a motion under Rule 60(b) of the Federal Rules of Civil Procedure based on an intervening change in the law. See Brief for Petitioner 4. The District Court denied it, evidently regarding the Court of Appeals' action on the appeal as conclusive. See ibid. A COA was granted but then vacated by a panel of the Court of Appeals. Gonzalez v. Secretary for the Dept. of Corrections, 317 F. 3d 1308, 1314 (CA11 2003). The panel quashed the original COA because it improperly addressed the original decision rather than the Rule 60(b) motion. Id., at 1312. The panel declined to issue a new COA on that motion on the alternative grounds that (1) the successive petition rule applied, ibid., and (2) even if it did not, Gonzalez clearly did not meet the "extraordinary circumstances" test for Rule 60(b)(6). Id., at 1312-1313.
The Court of Appeals granted rehearing en banc, granted a COA, and affirmed the denial of the Rule 60(b) motion. Gonzalez v. Crosby, 366 F. 3d 1253, 1286 (CA11 2004). This Court granted certiorari on January 14, 2005, limited to the Rule 60(b) question.
The Federal Rules of Civil Procedure do not apply to habeas corpus when their application would be contrary to the statutes or rules specifically governing habeas, or when they are unsuited to the specialized needs of this unique procedure. In particular, Rule 60(b) cannot be used to evade the successive petition rule. Congress has decided when a second round of habeas litigation may be commenced and has established special procedural safeguards for quickly determining which cases qualify for a second round. The purpose of this reform is not to reduce the number of petitions granted on a second round, which was never the problem, but to preclude the second round and its delay altogether. Broad consideration of Rule 60(b) motions would defeat the purpose of the statute.
Rule 60(b) does not apply to habeas corpus in any circumstance where it defeats the state's legitimate expectation of finality that AEDPA was enacted to protect. This includes nearly all uses of Rule 60(b) to reopen the habeas case after the normal termination of appellate review.
In the present case, use of Rule 60(b) to reopen the case over a year after the normal termination would have been contrary to the spirit of the habeas statutes, and the motion was properly denied.
The principal dissent in the Court of Appeals in this case illustrates perfectly how so many opinions on the present question start off on the wrong path and never return to the correct one. The opinion begins, "Federal Rule of Civil Procedure 60(b) authorizes the district courts to relieve a party to a civil action from the force of a final judgment on the following grounds . . . ." Gonzalez v. Crosby, 366 F. 3d 1253, 1287 (CA11 2004) (en banc) (Tjoflat, J., dissenting) (emphasis added). The question nearly screams off the page: yes, but what about habeas actions? The dissent never addresses this question. Judge Tjoflat proceeds to a detailed forensic examination of each tree, without ever checking to see if he is in the right forest.
Judge Edmonson's brief dissent similarly misses the threshold question. He approaches the question as one of reconciling an earlier and later statute which both apply to the subject, without ever pausing to ask if the earlier one applies. See id., at 1286-1287. The principle of "giv[ing] both of the laws Congress has approved their fullest possible force," id., at 1286, is simply inapplicable when one of those laws does not, by its own force, apply to the subject at hand. See Rancho Palos Verdes v. Abrams, 544 U. S. __ (No. 03-1601, Mar. 22, 2005) (slip op., at 6, n. 2) (canon against implied repeal inapplicable to question of whether newly created right with its own remedy is also enforceable through older, more general remedy).
Although habeas corpus is nominally a civil proceeding, that label is "gross and inexact." Harris v. Nelson, 394 U. S. 286, 293-294 (1969). "Essentially, the proceeding is unique." Id., at 294. As originally promulgated, the Federal Rules of Civil Procedure ("Civil Rules") had "very limited application to habeas corpus proceedings." Id., at 295. Civil Rule 81(a)(2) simply continued the application of civil rules to habeas proceedings to the extent they had been applied before the promulgation of the rules, but not further. Id., at 294. To the extent the Civil Rules introduced procedural innovations, such as broad discovery, they did not apply to habeas. See id., at 295. The Harris Court also noted "the unsuitability of applying to habeas corpus provisions which were drafted without reference to its peculiar problems." Id., at 296. Discovery as it exists in federal civil litigation was unsuited, because it would "do violence to the efficient and effective administration of the Great Writ." Id., at 297.
Unlike the principle that governs when earlier and later statutes both apply to the same case, no direct conflict with habeas statutes or rules is needed to render a Civil Rule inapplicable to habeas. Harris did not find that the civil discovery rules were in direct conflict with anything in the habeas statutes. It was enough that "their specific provisions are ill-suited to the special problems and characteristics of such proceedings." Rule 11 of the Rules Governing Section 2254 Cases in the United States District Courts ("Habeas Rules") was enacted seven years later, see Pub. L. 94-426 §1, 90 Stat. 1334 (1976), and it was "intended to conform with the Supreme Court's approach in the Harris case." Advisory Committee's Notes on Habeas Rule 11, 28 U. S. C., p. 479 (2000 ed.). Any proper analysis of whether a Civil Rule applies to habeas through Habeas Rule 11 must therefore consider the principles of Harris and its progeny. Yet petitioner's brief treatment of the subject does not, see Brief for Petitioner 10-11, the dissents in the Court of Appeals in this case do not, the Sixth Circuit's en banc opinion in In re Abdur'Rahman, 392 F. 3d 174 (2004) does not, and the other circuit cases collected by petitioner do not. See Brief for Petitioner 10-11, n. 7. Circuit cases which simply ignore the principles established by this Court for determining whether and how to apply a Civil Rule to habeas corpus have little, if any, persuasive value.
Petitioner claims Browder v. Director, Dept. of Corrections of Ill., 434 U. S. 257, 263, n. 8 (1978) "assume[s] that rule 60(b) applies in habeas corpus cases," Brief for Petitioner 10, but in fact that footnote simply says that the Court was not deciding any Rule 60 question, because the party who might have made a Rule 60 motion disclaimed any reliance on that rule. Petitioner stands on firmer ground with his contention that the concurring opinion assumed Rule 60(b) applies to habeas, see Browder, supra, at 272-274 (Blackmun, J., concurring), but a concurring opinion is not a precedent even for its holdings, much less its assumptions.
Browder is really about whether Civil Rules 52(b) and 59 apply to habeas corpus, see 434 U. S., at 265, and its methodology for answering that question reinforces Harris. The issue was whether the 10-day limit on motions for a new trial or to amend the judgment applied. See id., at 261, and n. 5. The Browder Court held they did, not because as Civil Rules they necessarily applied, but rather because they were consistent with and suited to the policies behind habeas procedure.
"Rule 59 in particular is based on an 'interest in speedy disposition and finality.' [Citation.] Although some aspects of the Federal Rules of Civil Procedure may be inappropriate for habeas proceedings, see Harris v. Nelson, supra; Preiser [v. Rodriguez, 411 U. S. 475 (1973)], at 495-496, the requirement of a prompt motion for reconsideration is well suited to the 'special problems and character of such proceedings.' Harris v. Nelson, supra, at 296. Application of the strict time limits of Rules 52(b) and 59 to motions for reconsideration of rulings on habeas corpus petitions, then, is thoroughly consistent with the spirit of the habeas corpus statutes." Id., at 271 (emphasis added).
Harris was applied specifically to Rule 60(b) in Pitchess v. Davis, 421 U. S. 482 (1975) (per curiam). A habeas petitioner successfully obtained a new trial, but then sought to preclude a retrial. He used Rule 60(b) to ask the Federal District Court to change its judgment from a conditional to an unconditional writ. Id., at 484-485. Under the circumstances, the basis of this claim could not be exhausted in state court until the post-trial appeal. See id., at 488. The Court held that Rule 60(b) could not be used to evade the exhaustion rule. Civil Rule 81(a)(2) precluded use of the Civil Rules in a manner contrary to the habeas statutes. "Since the exhaustion requirement is statutorily codified, even if Rule 60(b) could be read to apply to this situation it could not alter the statutory command." Id., at 489. Because there was a direct conflict with the statute on the facts of the case, there was no need in Pitchess to address the more general question of whether and when application of Rule 60 would be "consistent with the spirit of the habeas statutes."
The absence of a direct conflict with the habeas rules and statutes is a necessary condition to apply a Civil Rule to habeas, but it is not sufficient. The rule in question must be consistent with the purposes and policies of those rules and statutes in a more general sense as well.
A. Civil Rule 60.
Rule 60 "attempts to strike a proper balance between the conflicting principles that litigation must be brought to an end and that justice should be done." 11 C. Wright, A. Miller, & M. Kane, Federal Practice and Procedure §2851, p. 227 (2d ed. 1995) (cited below as "Wright & Miller"). Rule 60 was drafted with civil litigation in mind and therefore strikes that balance in the way thought appropriate for civil litigation. Criminal cases also raise the problem of balance between finality and error correction. The factors to be considered differ, however, and so a different rule addressing those concerns has evolved for habeas corpus. That different rule is the successive petition rule.
Taken as a whole, criminal procedure tilts the finality/justice balance sharply in the defendant's favor. For the criminal defendant, exclusively among all litigants, a favorable jury verdict is absolutely final and unreviewable. No matter how clearly erroneous, an acquittal cannot be overturned. See, e.g., United States v. Scott, 437 U. S. 82, 91 (1978). Criminal defendants alone, of all litigants, can have the lower federal courts review claims that state courts have erred on federal questions. The prosecution cannot, and civil litigants cannot. (2) See, e.g., Atlantic Coast Line R. Co. v. Locomotive Engineers, 398 U. S. 281, 286 (1970). Even when that review is concluded, res judicata as such does not apply to the denial of habeas relief. See infra, at 11.
The claim that unsuccessful habeas petitioners should be afforded relief from judgment on the same basis as civil litigants must be evaluated in this context. The entire proceeding is an additional layer of review that no other litigant receives, and the judgment from which relief is sought is not res judicata.
The direct effect of a judgment denying habeas relief is merely to leave the status quo intact. There is nothing to relieve. The prisoner remains in prison under the original judgment of conviction, not the judgment denying habeas. The need for relief only arises from the indirect effects. A second petition is subject to the limits of 28 U. S. C. §2244(b). In addition, because the statute of limitations is not tolled during the pendency of the first federal petition, see Duncan v. Walker, 533 U. S. 167, 181-182 (2001), the second petition might be time-barred. Congress has considered those possibilities and made its decision as to the appropriate balance. Exceptions for newly discovered facts, see 28 U. S. C. §§2244(b)(2)(B)(i), (d)(1)(D), retroactive new rules, see §§2244(b)(2)(A), (d)(1)(C), state-created impediments, see §2244(d)(1)(B), and actual innocence, see §2244(b)(2)(B)(ii), are built into the statute where and to the extent Congress thought them appropriate.
In the Thompson case, both this Court and the Ninth Circuit noted possible extreme cases where the state would have no legitimate expectation of finality. A party who procured a judgment by fraud upon the court would not have a legitimate judgment, Calderon v. Thompson, 523 U. S. 538, 557 (1998), and hence no entitlement to the finality of the judgment. Where the petitioner cannot meet the exceptions to the statutory rule because the state's misconduct prevents him from doing so, that misconduct could conceivably estop the state from asserting the bar. See Thompson v. Calderon, 151 F. 3d 918, 921, n. 3 (CA9 1998). Such rare circumstances can be addressed when and if they arise. The present case is a routine request to relitigate based on new legal authority. Nothing in petitioner's Rule 60(b) argument would confine use of the rule to any narrow subset of habeas cases. To hold that Rule 60(b) is generally applicable to the large class of cases where at least one claim was denied on the basis of default or statute of limitations would be to create a second, broad set of exceptions in addition to the narrow ones Congress decided upon when the issue was squarely before it. That is precisely what Harris, Pitchess, Civil Rule 81(a)(2), and Habeas Rule 11 forbid.
B. The Successive Petition Rule.
The rule on consideration of a second petition after denial of the first has evolved from unlimited allowance at common law to prohibition with only narrow exceptions under 28 U. S. C. §2244(b). This evolution has been a response to other changes in the law of habeas corpus.
At common law, the denial of habeas relief had no preclusive effect at all, and the petitioner could apply to a different judge for de novo reconsideration. See McCleskey v. Zant, 499 U. S. 467, 479 (1991). This rule was needed because there was no appellate review of the denial. See ibid. The rule was not a burden because of the extremely narrow scope of issues which could be considered on habeas. "As applied to criminal cases, habeas corpus was a pretrial remedy. . . . After conviction, the writ was not available to attack judgments of courts of competent jurisdiction." Scheidegger, Habeas Corpus, Relitigation, and the Legislative Power, 98 Colum. L. Rev. 888, 928 (1998) (footnotes omitted). The rule that denial of habeas was not res judicata presented no threat whatever to the finality of convictions, because the judgment of conviction itself was res judicata. See id., at 930; Ex parte Watkins, 3 Pet. (28 U. S.) 193, 203, 209 (1830).
Both the unreviewability of habeas decisions and the narrow scope of habeas issues are long gone, and the wide-open allowance of successive petitions is gone with them. Denial of federal habeas relief is reviewable by appeal to the Court of Appeals, 28 U. S. C. §2253, rehearing en banc in that court, Fed. Rule App. Proc. 35, and certiorari in this Court. See 28 U. S. C. §1254. The scope of issues now includes almost all constitutional questions, the only exception being the Fourth Amendment exclusionary rule. See, e.g., Withrow v. Williams, 507 U. S. 680 (1993) (declining to extend rule of Stone v. Powell, 428 U. S. 465 (1976) to Miranda claim). Constitutional claims, furthermore, have been vastly expanded beyond the basic requirements of fundamental fairness. See Rose v. Lundy, 455 U. S. 509, 543-544 (1982) (Stevens, J., dissenting). In capital cases, enormous amounts of time and money go into litigating alleged noncompliance with a complex and constantly changing web of rules governing the discretionary sentencing decision, rules having nothing whatever to do with guilt or eligibility for the death sentence. There is no limit to the number of such challenges creative lawyers could bring if allowed to do so, and hence without a successive petition rule capital sentences could never be carried out.
From the early twentieth century through the Antiterrorism and Effective Death Penalty Act of 1996, the law of habeas corpus evolved in the direction of greater finality, although not without " 'some backing and filling.' " Cf. Teague v. Lane, 489 U. S. 288, 308 (1989) (plurality opinion) (quoting Fay v. Noia, 372 U. S. 391, 411-412 (1963)). Salinger v. Loisel, 265 U. S. 224 (1924) was the first recognition of the successive petition rule in this Court. While rejecting the government's argument that a prior denial was res judicata, see id., at 230, the Court held, "Among the matters which may be considered, and even given controlling weight, are . . . (b) a prior refusal to discharge on a like application." Id., at 231. This was a very broad discretionary rule with no firm criteria for application. The Court indicated it would have affirmed if the District Court had denied relief on this basis, but it proceeded to the merits because the District Court had done so. Id., at 232. Either resolution was proper.
When Congress enacted a new judiciary code, it included §2244 on finality of determination. The original section was the same as the present subdivision (a), except that it applied to both state and federal prisoners. See 28 U. S. C. §2244 (1964 ed.). In keeping with Salinger, the rule was discretionary. See S. Rep. No. 1559, 80th Cong., 2d Sess., 9 (1948).
Sanders v. United States, 373 U. S. 1 (1963) was the final chapter of "a trilogy of 'guideline' decisions" on habeas corpus. Id., at 23 (Harlan, J., dissenting). "The over-all effect of this trilogy . . . [was] to relegate to a back seat . . . the principle that there must be some end to litigation." Ibid. Sanders sharply limited the Salinger rule regarding successive applications. See id., at 15.
Congress soon acted to reduce successive petitions. The problem was not the number of cases in which relief was being granted but rather the burdens of the litigation. See H. R. Rep. No. 1892, 89th Cong., 2d Sess., 5 (1966); S. Rep. No. 1797, 89th Cong., 2d Sess., 2 (1966). In addition, Congress found "disconcerting . . . the delays in executing State court sentences in capital cases as a result of habeas corpus applications seeking review of State court action . . . ." H. R. Rep. No. 1892, supra, at 5. Various proposals were considered to deal with this problem. See id., at 5-6. One that survived in the final bill was "to add to section 2244 . . . provisions for a qualified application of res judicata." Id., at 8. Subdivision (b) was added to provide that "after an evidentiary hearing on the merits of a factual issue, or after a hearing on the merits of an issue of law . . . a subsequent application . . . need not be entertained" unless based on a new ground which was not deliberately withheld or otherwise abusive.
Despite the clear intent of Congress to change the law in the direction of greater finality, see S. Rep. No. 1797, supra, at 2, it was another 20 years before this Court reexamined Sanders. In Kuhlmann v. Wilson, 477 U. S. 436 (1986), a plurality adopted Judge Friendly's "colorable claim of innocence" requirement for successive petitions. Id., at 454. "A 'successive petition' raises grounds identical to those raised and rejected on the merits on a prior petition." Id., at 444, n. 6. Five years later, McCleskey v. Zant reexamined the abuse-of-the-writ aspect of Sanders and adopted the cause-and-prejudice test for second petitions with claims omitted from the first. See 499 U. S., at 490.
In 1996, Congress decided that Kuhlmann and McCleskey had not gone far enough in restricting repeated rounds of habeas litigation. Just as in 1966, the concern was not with excessive grants of relief, but rather with the burden and especially the delay from the multiple rounds of litigation. In one notorious case, the McCleskey rule had not been clear enough to prevent the issuance of a stay of execution to entertain a fifth federal challenge to a death sentence on an obviously defaulted claim. See Gomez v. United States Dist. Court for the Northern Dist. of Cal. (Harris), 503 U. S. 653, 653-654 (1992) (per curiam). Senator Hatch also cited the Andrews case, which took 18 years and 30 appeals. See 141 Cong. Rec. 15,062, col. 2 (1995). That case took almost three years on the second round of federal habeas. See Andrews v. Deland, 943 F. 2d 1162, 1168 (CA10 1991) (petition filed July 19, 1989), cert. denied, 502 U. S. 1110 (1992), rehearing denied, 503 U. S. 967 (March 30, 1992). This was in a case involving "no question of Andrews' participation in the crimes," 943 F. 2d, at 1186, an almost unbelievably horrific case of sadistic torture and multiple murder. See State v. Pierre, 572 P. 2d 1338, 1343-1344 (Utah 1977).
To preclude more than one round of federal review in all but the rarest cases, Congress clamped down hard on "second or successive" habeas corpus applications in its revision of §2244(b). See Tyler v. Cain, 533 U. S. 656, 661-662 (2001). Subdivision (1) forbids without exception claims "presented in a prior application," i.e., what has traditionally been called a "successive" petition. Subdivision (2) applies to claims "not presented in a prior application," i.e., the "abuse of the writ" scenario. Only two narrow exceptions are allowed: (1) retroactive new rules; and (2) newly discovered facts and actual innocence. The references to the "merits" in former subdivision (b) are not present in the new subdivision.
Congress's intent to preclude the second round of litigation, not merely to enable the state to prevail in that round, is further implemented by the extraordinary procedural measures in subdivision (b)(3). Subdivision (b)(3)(A) requires leave of the Court of Appeals to even file the petition, and subdivision (b)(3)(B) requires that decision to be made by a three-judge panel. This is to preclude shopping for a single judge to authorize filing and grant a stay. Subdivision (b)(3)(D) requires a decision in 30 days, and (b)(3)(E) forbids rehearing or certiorari review of that decision. The clear purpose here is that, in nearly all cases, the attempt to begin a second or subsequent round of federal review will be over in 30 days. In most capital cases, the state should be able to set an execution date the month following final disposition of the first federal habeas petition.
The intent of Congress would be subverted if the limits on successive habeas petitions could be circumvented merely by invoking a different procedural device. Twice this Court has rebuffed such attempts. In Gomez, the habeas petitioner withheld his challenge to the use of cyanide gas until the eve of execution and then filed the claim as a civil rights action under 42 U. S. C. §1983. See 503 U. S., at 653. "This action is an obvious attempt to avoid the application of McCleskey v. Zant, 499 U. S. 467 (1991) to bar this successive claim for relief." Ibid. The Court held that Harris had made no showing of cause, ibid., impliedly holding that the McCleskey standard applied.
Calderon v. Thompson, supra, is similar. "Thompson filed a motion with the Court of Appeals to recall its mandate denying habeas relief." 523 U. S., at 546. He also filed a Rule 60(b) motion in the District Court. See id., at 547. An en banc panel of the Court of Appeals recalled the mandate. It "asserted it did not recall the mandate on the basis of Thompson's later motion for recall, but did so sua sponte on the basis of the claims and evidence presented in Thompson's first federal habeas petition." Id., at 548.
Regarding recalls of habeas mandates in response to a petitioner's motion, the Thompson Court said,
"In a §2254 case, a prisoner's motion to recall the mandate on the basis of the merits of the underlying decision can be regarded as a second or successive application for purposes of §2244(b). Otherwise, petitioners could evade the bar against relitigation of claims presented in a prior application, §2244(b)(1), or the bar against litigation of claims not presented in a prior application, §2244(b)(2). If the court grants such a motion, its action is subject to AEDPA irrespective of whether the motion is based on old claims (in which case §2244(b)(1) would apply) or new ones (in which case §2244(b)(2) would apply)." Id., at 553 (emphasis added).
This statement is dictum, since the Court went on to hold that the sua sponte recall on the original petition was not subject to §2244(b). See id., at 554. Even so, it is an important statement of principle, and one on which the Court appeared to be unanimous. See id., at 569, n. 1 (Souter, J., dissenting).
Gomez and Thompson are applications of the more general principle that when Congress has provided a specific procedure for the resolution of particular controversies and placed limitations on it, those limitations cannot be evaded simply by choosing a different and more general procedure. Even though a challenge to the fact or duration of custody by state officers may fall within the broad language of the civil rights remedy statute, 42 U. S. C. §1983, it cannot be used in lieu of habeas for such a challenge. See Spencer v. Kemna, 523 U. S. 1, 20 (1998) (Souter, J., concurring). Despite the broad wording of the All Writs Act, it cannot be used when another "statute specifically addresses the particular issue at hand . . . ." Pennsylvania Bureau of Correction v. United States Marshals Service, 474 U. S. 34, 43 (1985).
The successive petition rule is a "modified res judicata rule," well within the power of Congress to enact. See Felker v. Turpin, 518 U. S. 651, 664 (1996). Courts can weigh the competing interests of habeas petitioners and the state in the absence of a statute or in construing a vague statute, as this Court did in Sanders, Kuhlmann, and McCleskey, see Lonchar v. Thomas, 517 U. S. 314, 323 (1996), but the balance struck judicially is always subject to legislative revision. Congress has deliberately moved the mark, and the wisdom of its choice is not for courts to decide. See id., at 328. Neither is the statute a mere inconvenience to be evaded.
Congress meant to give habeas petitioners one round of federal review, i.e., decision by the district court, appeal to the court of appeals, and certiorari to this Court. The end of that first round was meant to be the end in all but the rarest cases, with a highly expedited process for determining whether a case was one of the rare ones. A holding that Rule 60(b) is generally available would destroy this system. Even if nearly all Rule 60(b) motions are denied, unless they are subject to the requirements of §2244(b) the litigation of them will become the second round of review that Congress meant to prevent.
A. Finality and AEDPA.
In his dissent in Abdur'Rahman v. Bell, 537 U. S. 88 (2002), Justice Stevens endorsed the approach of Judge Tjoflat, drawing the line between Rule 60(b) motions that challenge the integrity of the habeas proceeding itself and those that assert new or newly bolstered attacks on the underlying criminal judgment. See id., at 94-96 (quoting Mobley v. Head, 306 F. 3d 1096, 1100-1105 (CA11 2002) (dissenting opinion)). Judge Tjoflat proffers the same distinction in the present case, designating the former type a "true Rule 60(b) motion." Gonzalez v. Crosby, 366 F. 3d 1253, 1292 (CA11 2004) (dissenting opinion).
This proposal would draw the line
at a point which is utterly irrelevant to the purpose Congress had in mind when
it enacted AEDPA. The overarching theme of AEDPA is to enhance the finality
of criminal judgments. Congress was deeply dissatisfied with the way federal
courts had entertained one proceeding after another, and it resolved to limit
habeas petitioners to "one bite of the apple." See Brief for Criminal Justice
Legal Foundation as Amicus Curiae in Bell v. Thompson,
No. 04-514, pp. 10-13 ("CJLF Thompson Brief"), http://www.cjlf.org/
briefs/ThompG.pdf.
Nothing in the language or purpose of AEDPA limits its enhanced finality to cases where the federal court reaches the petitioner's underlying claim, as opposed to those decisions based on procedural default or the statute of limitations. Just the opposite, Congress repealed references to "the merits" in the earlier statute. See supra, at 13. The successive petition limit is "a modified res judicata rule," Felker v. Turpin, 518 U. S. 651, 664 (1996), and res judicata applies just as much to a decision that a claim is permanently barred as it does to a decision that a claim was meritless from the beginning. See Angel v. Bullington, 330 U. S. 183, 190 (1947); see also Plaut v. Spendthrift Farm, Inc., 514 U. S. 211 (1995) (dismissal on statute of limitations was res judicata, cannot be reopened even when Congress directed reopening). Cases involving decisions that federal relief cannot be granted yet, as opposed to decisions that federal relief cannot be granted at all, are inapposite. Cf. Slack v. McDaniel, 529 U. S. 473, 487 (2000) (unexhausted); Stewart v. Martinez-Villareal, 523 U. S. 637, 644-645 (1998) (premature).
Under Judge Tjoflat's proposed separation, decisions on procedural grounds would never be truly final. The travesty of the Robert Alton Harris execution that Congress specifically sought to ensure would never happen again, see supra, at 13; CJLF Thompson Brief 10-11, could happen again. Among claims that might be considered "seeking relief from the federal court's final order entered in a habeas proceeding," Abdur'Rahman, 537 U. S., at 94 (dissent), include claims that the court erred in deciding it could not reach the underlying claim because of (1) the statute of limitations, (2) procedural default, (3) failure to develop facts in state court, or (4) nonretroactivity of the claimed rule of law. In other words, use of virtually any of the tools that Congress and this Court have provided to limit second-guessing of the state court by the federal court would transform a judgment in the state's favor into a mere tentative ruling, reopenable at any time on a Rule 60(b) motion. While the First Circuit expresses confidence that "the district courts will be able to sift wheat from chaff without undue difficulty," Rodwell v. Pepe, 324 F. 3d 66, 71 (2003), Congress does not share this confidence. It enacted AEDPA precisely because it was exasperated with the federal courts' insufficient respect for the need for finality in criminal judgments.
Amicus Abdur'Rahman contends that Rule 60(b) motions are only filed in a "miniscule fraction" of all habeas cases. See Brief for Abu-Ali Abdur'Rahman as Amicus Curiae 4-5 ("Abdur'Rahman Brief"). However, the methodology of this survey does not capture all 60(b) motions, see id., at 5, n. 6, and probably misses most of them. More importantly, the survey does not separate the cases Congress was most interested in--capital cases--from the others. In capital cases, 60(b) motions are not rare. In Amicus CJLF's survey of executed capital judgments in the Ninth Circuit, there were three such motions denied by the District Court and appealed to the Court of Appeals out of nineteen cases, nearly one-sixth of the total. When combined with motions to recall the mandate (or to further stay it and reconsider after this Court's denial of certiorari) plus civil rights actions, there was some substantial attempt to evade Congress's successive petition limit in nearly every case. See CJLF Thompson Brief, Appendix A.
Amicus Abdur'Rahman also quibbles between attacking the finality of the underlying conviction and attacking the finality of the habeas judgment. Abdur'Rahman Brief 4, n. 5. Any ongoing proceeding that forces the state to marshal its resources to protect its judgment, see Teague v. Lane, 489 U. S. 288, 310 (1989), is an attack on the finality of the judgment. In capital cases, any ongoing proceeding that lengthens the stay of execution, delaying and denying justice, is an attack on the finality of the judgment. Abdur'Rahman's distinction has no difference.
B. Applying Harris to Rule 60(b).
Harris v. Nelson, 394 U. S. 286, 298 (1969), held that, even though the discovery provisions of the Civil Rules did not directly apply to habeas corpus, courts could nonetheless order discovery adopting such procedures as are suited to habeas. Habeas Rule 6 subsequently codified the specific holding of Harris for discovery, while Habeas Rule 11 codified its general approach to adoption of Civil Rules.
The same approach is in order for Civil Rule 60. Looking at how that rule is applied in civil cases, is each application consistent with the goal of Congress to enhance the finality of criminal proceedings? The simplest case is presented by Rule 60(a), correction of clerical mistakes. "The State can have little interest . . . in preserving a [judgment] not in accordance with the actual decision rendered by the court." Calderon v. Thompson, 523 U. S. 538, 557 (1998). Rule 60(b) presents the more difficult questions.
Preliminarily, two overall requirements of Rule 60(b) are worth noting. First, the movant must show an arguably meritorious claim. See 16 J. Moore, et al., Moore's Federal Practice §60.24, pp. 60-77 to 60-79 (3d ed. 2000); cf. Slack v. McDaniel, 529 U. S. 473, 478 (2000) (COA on procedural point requires at least debatable merit on both procedural claim of error and underlying constitutional claim). Second, all six clauses of Rule 60(b) require a motion within a reasonable time, and the first three have an additional, strict, one-year cut-off. See Moore's, supra, §60.65, at 60-198 to 60-203.
1. Mistake, inadvertence, surprise, and excusable neglect.
An entry of judgment against a pro se prisoner who is unable to respond due to the circumstances of his incarceration is an example of a situation warranting relief under this clause. See id., §60.41[1][b], at 60-87 to 60-88. As applied to habeas, this would typically happen when the respondent is at least partly at fault, and the one-year limit prevents indefinite delay of true finality. The situation will not arise in capital cases, where §2254 petitioners always have the right to counsel. See 21 U. S. C. §848(q)(4)(B). Errors of counsel are generally not cognizable under this clause, see Moore's, supra, §60.41[2], at 60-95 to 60-96, and entertaining a Rule 60(b) motion on grounds of ineffective assistance of habeas counsel would circumvent the rule of Coleman v. Thompson, 501 U. S. 722, 753-754 (1991).
The circuits are divided on whether a mistake of law by the district court is a ground for relief under Rule 60(b)(1), and those that do allow it generally construe "reasonable time" to mean within the time to appeal. See Moore's, supra, §60.41[41], at 60-99 to 60-113. Allowing the district court to correct a patent error before appeal that would surely be grounds for reversal on appeal makes good sense in terms of efficiency. See id., at 60-106. Once the appeal is under way, the appeal itself is the efficient and proper way to correct a legal error. Once the appeal is over, the intent of Congress to create "a modified res judicata rule," see supra, at 15, forbids use of this ground to obtain an otherwise forbidden second round of litigation.
2. Newly discovered evidence.
Congress expressly considered the question of newly discovered evidence and provided when it would be grounds for a successive petition and to restart the statute of limitations clock. See 28 U. S. C. §§2244(b)(2)(B), (d)(1)(D). Rule 60(b)(2) cannot be used to strike the balance at a different point.
3. Fraud on the court.
If the state procures a judgment by fraud, it has no legitimate interest in the finality of its illegitimate judgment. See Calderon v. Thompson, 523 U. S., at 557.
4. Void judgment.
Rule 60(b)(4) is unnecessary in many cases, see Moore's, supra, §60.44[b], at 60-140 to 60-141, and so it is for the petitioner in habeas corpus. If the first petition was denied by a court lacking jurisdiction, the second is simply not successive, and no relief from the first is required. The State, on the other hand, may need this relief. See Rumsfeld v. Padilla, 542 U. S. __, 124 S. Ct. 2711, 2715-2717, 159 L. Ed. 2d 513, 524-527 (2004) (release ordered by court lacking subject-matter jurisdiction).
5. Satisfaction, reversed prior judgment, prospective application.
The "satisfaction" prong of this clause has no application to a denial of habeas relief. There is nothing to satisfy. Reversed prior judgment could be a legitimate use for Rule 60(b)(5), or it could be unnecessary, depending on how this Court resolves Johnson v. United States, No. 03-9685.
Unjust prospective application applies only to habeas judgments granting relief. Denial of injunctive relief does not have prospective application. See Moore's, supra, §60.47[1][e], at 60-160 to 60-161. On the other hand, where the custodian has been ordered to release the prisoner or give him a new trial, it would be unjust to require prospective application if new authority comes down before the trial negating the basis of the judgment. Cf. Agostini v. Felton, 521 U. S. 203, 215 (1997) (60(b)(5) relief from injunction due to change in law). For example, in Ritter v. Smith, 811 F. 2d 1398, 1400-1401 (CA11 1987), the Court of Appeals had directed that a conditional writ of habeas corpus issue, requiring a new sentencing hearing, based on a holding that the sentencing statute was unconstitutional. Before the resentencing was conducted, this Court held in another case that the statute in question was constitutional. Ritter held that the state was entitled to 60(b) relief, although mistakenly placing the case under 60(b)(6). See id., at 1403. As Agostini now makes clear, the correct clause is (b)(5).
6. Catch-all.
The difficult problem, and the great danger, is the "catch-all" provision of clause 6: "any other reason justifying relief from the operation of the judgment." What justifies relief is a value-laden judgment, and the potential for evasion of AEDPA by judges who strongly disagree with Congress's value choices is great.
This Court has already established that the broad language of Rule 60(b)(6) may not be used to evade the limits on the other clauses. See Liljeberg v. Health Services Acquisition Corp., 486 U. S. 847, 863, and n. 11 (1988). "Extraordinary circumstances" are required. See id., at 864; Ackermann v. United States, 340 U. S. 193, 202 (1950).
The majority position is that a change in decisional law does not amount to "extraordinary circumstances" even in the civil cases to which Rule 60(b)(6) directly applies. See Moore's, supra, §60.48[5], at 60-180 to 60-183; Hess v. Cockrell, 281 F. 3d 212, 216, and n. 18 (CA5 2002) (habeas case relying on civil precedents). Certainly it is not a ground for relief as applied to habeas corpus. Congress has expressly considered when such a change will be a ground for relief: new rules of constitutional law made retroactive on habeas by this Court. See 28 U. S. C. §2244(b)(2)(A); Tyler v. Cain, 533 U. S. 656, 662 (2001). To reopen a criminal case after the normal end of one full round of habeas review based on any lesser change in case law would be to strike the balance at a different point than the one chosen by Congress.
Amicus Abdur'Rahman presents several examples of cases where application of Rule 60(b) to habeas is purportedly necessary to prevent injustice and consonant with the spirit of the habeas statutes. In some, application of Rule 60(b) is, indeed, appropriate. In others, other remedies are available and appropriate. In still others, application of Rule 60(b) is contrary to the purpose of AEDPA, gives too little finality to the judgment, and is improper.
Abdur'Rahman's first example is himself. This case was thoroughly briefed when it was before this Court, and the argument need not be repeated here. See Brief Amicus Curiae for Criminal Justice Legal Foundation in Abdur'Rahman v. Bell, No. 01-9094, http://www.cjlf.org/pdf/Abdur'R.pdf. For the reasons explained in our prior brief, habeas was correctly denied in that case. Abdur'Rahman personally bound and blindfolded the victims, rendering them helpless for slaughter. See id., at 2. The miscarriage of justice in Abdur'Rahman's case is the fact that the judgment remains unexecuted 19 years after the crimes.
Most of the other examples involve the Second Circuit's changing interpretation of the AEDPA statute of limitations, 28 U. S. C. §2244(d), to pre-AEDPA cases. In Peterson v. Demskie, 107 F. 3d 92, 93 (1997), the Second Circuit indicated that in such cases the petition needed to be filed within a "reasonable time" after enactment of AEDPA on April 24, 1996, but that prisoners who had already waited several years at that point need not be given the full additional year. Then in Ross v. Artuz, 150 F. 3d 97, 102-103 (1998), the Second Circuit decided that these petitioners would have the full year, until April 23, 1997. Either interpretation is reasonable, and either one protects the diligent petitioner who files his federal petition promptly upon exhaustion of his state remedies. At least since the adoption of Habeas Rule 9(b) in 1976, prisoners have been on notice that delayed petitions might be dismissed if they prejudiced the state, and no one who waited years after exhaustion of state remedies could legitimately claim to be unfairly surprised.
Matos v. Portuondo, 33 F. Supp. 2d 317 (SDNY 1999) is an example of a legitimate use of Rule 60(b) to correct a mistake of law by the District Court that would surely have resulted in reversal on appeal. Ross came down five days after the District Court dismissed the petition under Peterson, and the petitioner filed his 60(b) motion 16 days after dismissal, within the time to appeal. This is simple judicial efficiency. There was no point in going through the motions of an appeal when the result was obvious. Granting relief in this situation does not impair any legitimate expectation of finality, and it is proper under Rule 60(b)(1), not (b)(6). See supra, at 20.
Robles v. Senkowski, No. 97 Civ. 2798, 1999 U. S. Dist. LEXIS 11565 (SDNY, July 30, 1999) is an example where another remedy was available and should have been used. The appeal was already pending when Ross was decided, but the Court of Appeals dismissed it because petitioner had neither paid the filing fee nor moved to proceed in forma pauperis. See id., at *2-*3. The available, correct procedure was to cure and reinstate the appeal, at which point the Court of Appeals would have reversed and remanded under Ross. See Gaca v. United States, 411 U. S. 618 (1973) (per curiam); Reyes v. Keane, 90 F. 3d 676, 677-678, 681 (CA2 1996) (reinstating appeal of pro se habeas petitioner in similar circumstances).
Liberatore v. McGuiness, No. 96 Civ. 6943, 1998 U. S. Dist. LEXIS 22842 (SDNY, Oct. 7, 1998), is an example where Rule 60(b)(6) was improperly used to defeat the finality of the judgment. Liberatore was convicted of violent crimes in 1987. See id., at *1. He received review of his claims in a published decision of the Appellate Division. See ibid. (citing People v. Liberatore, 561 N. Y. S. 2d 832 (1990)). He then waited five years after exhaustion of state remedies to file a federal habeas petition. The District Court dismissed the habeas petition as untimely on September 20, 1996. See id., at *2. If this dismissal was erroneous, there was a remedy by appeal, but petitioner did not appeal. See ibid. Sometime after Ross, petitioner filed his Rule 60(b) motion, which makes it well after the strict one-year cut-off for 60(b)(1). See id., at *3-*4. Contrary to the District Court's decision in this case, there were no "extraordinary circumstances" justifying the drastic action of reopening a final judgment. A simple change in case law is not "extraordinary" by any definition. It happens all the time. In civil cases, it is the nature of res judicata that a judgment once final decides the matter between the parties, even if a subsequent development in case law indicates that the decision was erroneous. Even in civil cases, Rule 60(b)(6) is not an open-ended invitation to reopen any judgment any time a precedent on which it is based is overruled. "Indeed, the common law could not safely develop if the latest evolution in doctrine became the standard for measuring previously resolved claims. The finality of judgments protects against this kind of retroactive lawmaking." See Biggins v. Hazen Paper Co., 111 F. 3d 205, 212 (CA1 1997).
In Liberatore, there was no compelling need to extend the temporal reach of Ross to rescue a dilatory petitioner who waited years to file his federal habeas petition and did not bother to appeal its dismissal. The state had won its case at trial, successfully defended its judgment in the state courts, and successfully defended it again under the then-prevailing circuit rule on federal habeas. Congress has decided that in such circumstances further reconsideration will be strictly limited to a very small set of exceptionally compelling cases. See 28 U. S. C. § 2244(b). There are a great many cases that would be considered incorrectly decided under later case law that do not qualify for these strict criteria. Congress has decided that finality in criminal law is important enough that it outweighs the need for error correction in all but the cases specified. By deciding that a simple change in procedural law "justifies" relief under Rule 60(b)(6), the District Court substituted its value judgment for that of Congress.
Petitioner proffers the example of claims of mentally retarded inmates on death row under Atkins v. Virginia, 536 U. S. 304 (2002) as a reason why Rule 60(b)(6) motions must be allowed long after the normal end of the first habeas petition. See Brief for Petitioner 42-44. However, the "functional equivalent" approach of petitioner and the cases he relies on is neither necessary nor sufficient to address the Atkins cases.
Under this approach, a "true" or "proper" 60(b)(6) motion addresses a defect in the habeas process itself, as distinguished from a newly asserted or reasserted ground for attacking the state judgment. The latter is the functional equivalent of a second or successive habeas petition. See Brief for Petitioner 19-20. A petitioner claiming to be mildly retarded and claiming that this condition categorically excluded him from the death penalty would have been denied relief before Atkins in a perfectly correct habeas proceeding. Every district and circuit court was duty-bound to apply the not-yet-overruled Supreme Court precedent on point, Penry v. Lynaugh, 492 U. S. 302, 334 (1989). A post-Atkins Rule 60(b)(6) motion by such a petitioner does "address directly the underlying state court proceedings," Brief for Petitioner 19, and this is the functional equivalent of a second or successive habeas petition under petitioner's own authorities. The proposed rule is therefore not sufficient to redress the Atkins claims.
The proposed rule is also not necessary for these claims. There has been no state-court rebellion against Atkins. State courts and legislatures are fashioning procedures and hearing the claims. See, e.g., In re Hawthorne, 35 Cal. 4th 40, 43-44, 105 P. 3d 552, 554 (2005) (allowing fourth habeas petition and transferring case for evidentiary hearing). Executive clemency is also available for any death row inmate who either was under 18 at the time of the crime or claims to be mentally retarded. See, e.g., Statement of Gov. Rick Perry on U. S. Supreme Court Ruling on Death Penalty (March 1, 2005), http://www.governor.state.tx.us/divisions/press/pressreleases/PressRelease.2005-03-01.3948 (directing parole board evaluation in response to Roper v. Simmons, 543 U. S. __ (No. 03-633, Mar. 1, 2005)). If and when a retarded or juvenile murderer actually faces execution and is denied any remedy, this Court can decide whether the Constitution requires one. The remedy need not be Rule 60(b) or, for that matter, federal habeas corpus.
IV. There is no constitutional
doubt in this case.
The value judgment implicit in petitioner's "constitutional doubt" argument is a belief that every state prisoner should get a federal court ruling on the merits of his underlying claim, and the courts should bend every rule to see that he gets it. That belief was the guiding light of this Court's habeas corpus jurisprudence at one time under the regime of Fay v. Noia, 372 U. S. 391 (1963), but that day is long gone. Fay and its approach to habeas were steadily eroded by a long series of decisions, see e.g., Wainwright v. Sykes, 433 U. S. 72, 90 (1977), until it was formally overruled in Coleman v. Thompson, 501 U. S. 722, 750 (1991). Coleman stated expressly what had long been implied: Fay had underestimated the importance of finality. See ibid. Congress not only endorsed this revised policy determination, but it decided this Court had not gone far enough in the direction of finality. The successive petition rule enacted by Congress is considerably more stringent than this Court's pre-AEDPA cases, see Calderon v. Thompson, 523 U. S. 538, 558 (1998), and the new statute of limitations is more stringent that the pre-AEDPA Habeas Rule 9(b), as construed in Lonchar v. Thomas, 517 U. S. 314, 328 (1996). Petitioner's call to return to Fay's imperative of deciding the merits of every underlying claim is an anachronism. Indeed, now that Congress has spoken, this Court could not return to Fay even if it were inclined to do so. Absent a constitutional mandate, the decision of Congress on a question of policy is final.
Apparently aware that the intent of AEDPA is against him, petitioner turns to the last refuge for interpretive lost causes: the doctrine of constitutional doubt. See Reno v. Flores, 507 U. S. 292, 314, n. 9 (1993); Brief for Petitioner 32-39. In habeas cases, this inevitably takes the form of an appeal to the Suspension Clause, U. S. Const., Art. I, §9, cl. 2. Like the monster in a series of bad horror movies, this argument keeps coming back no matter how many times it appears to have been killed. It is high time for this Court to bury it for good.
There are two simple reasons why the Suspension Clause creates no doubt whatsoever of the constitutionality of AEDPA. First, the Suspension Clause creates no obligation for Congress to extend federal habeas to state prisoners. Second, the Suspension Clause creates no obligation to make habeas corpus available as a means of collateral attack on the criminal judgments of courts of general jurisdiction. Either is sufficient to eliminate any doubt.
On federal habeas corpus ad subjiciendum for state prisoners, the historical record is crystal clear. The First Congress flatly prohibited it. See Judiciary Act of 1789, §14, 1 Stat. 73, 81-82. Although at least one federal court chafed at the restriction, see Ex parte Cabrera, 4 F. Cas. 964, 965-966 (No. 2278) (CCD Pa. 1805), no doubts were expressed as to its constitutionality. See Scheidegger, Habeas Corpus, Relitigation, and the Legislative Power, 98 Colum. L. Rev. 888, 932 (1998); see also Ex parte Dorr, 44 U. S. (3 How.) 103, 105 (1845) (following statute, no constitutional doubts raised or expressed).
On the use of habeas corpus as a collateral attack on the judgment of a court of general jurisdiction, the record is equally clear. The writ was so clearly unavailable for that purpose at the time of the Founding that it was 41 years before anyone even tried. See Scheidegger, 98 Colum. L. Rev., at 929, n. 280. When someone finally did, the attempt was definitively rejected. See id., at 929-930 (discussing Ex parte Watkins, 28 U. S. (3 Pet.) 193 (1830)).
In Felker v. Turpin, 518 U. S. 651, 663 (1996), this Court effectively acknowledged that the constitutional doubt argument was baseless on the original understanding of the Suspension Clause. The Felker Court did not resolve whether the Suspension Clause could have somehow "evolved" to make a constitutional mandate of something the founding generation flatly forbade. Instead Felker held that the limitation on successive petitions was consistent with even a hypothetical evolved Suspension Clause. See id., at 664. That hypothetical should be addressed and rejected in the present case. Whatever argument might be made for constitutional evolution in such vague and expansive phrases as "unreasonable search" or "cruel and unusual punishment," see, e.g., Roper v. Simmons, 543 U. S.__ (No. 03-633, Mar. 1, 2005) (slip op., at 6), there is no credible argument for such amendment by interpretation here. The Suspension Clause used a legal term of art, habeas corpus, that referred to a specific procedure for a specific purpose: getting people out of jail who are held there illegally before trial or without trial. The fact that this procedure later evolved into a mode of collateral attack on final convictions is no ground for similarly expanding the constitutional mandate, nor is it any ground for mandating federal review of state convictions, when Article III clearly empowers Congress to set the jurisdictional bounds of the "inferior" courts it creates. See Scheidegger, 98 Colum. L. Rev., at 918-919.
The First Congress had the power to forbid federal habeas for state prisoners altogether. No amendment has rescinded that power. The First Congress created a writ for federal prisoners that did not include collateral attack on judgments of courts of general jurisdiction. No amendment has rescinded that power. The greater power to forbid collateral review includes the lesser power to offer it on limited terms, so long as the limitations are not based on suspect classifications or other unconstitutional discrimination. See Scheidegger, 98 Colum. L. Rev., at 953-957. There is no doubt about the constitutionality of AEDPA under any reasonable interpretation, and hence the doctrine of constitutional doubt has no application to this case.
IV. Use of Rule 60(b) in
the present case would be
contrary to the spirit of AEDPA.
Applying the foregoing principles is straightforward. Petitioner's claims were reviewed and rejected by the state courts. The federal courts rejected his petition as untimely in a decision that was a reasonable interpretation of the statute and may very well have been correct. He had his one bite at the federal-court apple, which does not necessarily include determination of the merits of the underlying claim. The state's reasonable expectation of finality attached when the Court of Appeals denied a certificate of appealability and petitioner sought no further review of that decision. Cf. Calderon v. Thompson, 523 U. S. 538, 556 (1998).
To allow the use of Rule 60(b) to reopen the habeas case based on a mere change in procedural case law would be contrary to the purpose of AEDPA to strictly limit such reopenings. Petitioner has not alleged a miscarriage of justice in the sense that he is actually innocent of the offenses for which he is in prison. Cf. id., at 558; 28 U. S. C. §2244(b)(2)(B)(ii). Absent such a claim, and credible evidence to support it, the federal habeas case should be over when it is over. Cf. Y. Berra, quoted in J. Bartlett, Familiar Quotations 903 (15th ed. 1980).
The decision of the Court of Appeals for the Eleventh Circuit should be affirmed.
April, 2005
Respectfully submitted,
Kent S. Scheidegger
Attorney for Amicus Curiae
Criminal Justice Legal Foundation
1. This brief was written entirely by counsel for amicus,
as listed on the cover, and not by counsel for any party. No outside contributions
were made to the preparation or submission of this brief.
Both parties have given written consent to the filing of this brief.
2.
On March 21, 2005, Congress made an extraordinary exception for one specific
case, Pub. L. 109-3, but the foregoing statement remains true for all others.