Go to:	Case Menu	Briefs Menu

C. Custis v. United States.

When Custis v. United States, 511 U. S. 485 (1994) was being briefed and argued, the advocates involved understood that the case was about whether, not when, priors may be collaterally attacked. Defendant referred to the position of the Fourth Circuit and the government as a "prohibition on court review of nearly all constitutional violations." Brief for Petitioner in Custis v. United States, No. 93-5209, p. 25. The government argued, "At some point, a state court judgment becomes final and is no longer open to challenge on grounds that the defendant has not preserved." Brief for United States in Custis v. United States, No. 93-5209, p. 22 (emphasis added); see also Brief for Criminal Justice Legal Foundation as Amicus Curiae in Custis v. United States, No. 93-5209, pp. 2-3.

The Custis majority also saw the issue this way, as indicated by both the language of the opinion and the manner in which it addressed the constitutional precedents. On the constitutional question, the two principal precedents cited by defendant were Burgett v. Texas, 389 U. S. 109 (1967) and United States v. Tucker, 404 U. S. 443 (1972). See Custis, 511 U. S., at 493. If the case were about timing of challenges, these precedents would be quite different, while if it were about allowability they would be quite similar. In Burgett, defendant challenged the use of priors on Gideon⁽⁶⁾ grounds during the course of the trial, see 389 U. S., at 112, and the case came to this Court on direct appeal. See id., at 110. Tucker, on the other hand, involved a proceeding under 28 U. S. C. §2255, many years after the original conviction. See 404 U. S., at 445.

If the issue in Custis had been merely whether Custis could bring his challenge in the sentencing proceedings or had to delay them to a §2255 motion, the only relevance of Tucker would have been to illustrate the propriety of the latter course. Custis would not have needed to distinguish Tucker on any basis other than the differing procedural positions of the two cases. Instead, though, the Custis Court treated Burgett and Tucker as equivalent on the point at issue. The Court distinguished both cases on the ground that they involved Gideon claims and that Gideon was a "unique constitutional defect." Custis, 511 U. S., at 496.

Custis took a further step toward recognizing that constitutional claims differ in their impact on fundamental fairness and that this difference ought to result in differing degrees of reviewability. This concept is highly controversial, and the members of this Court have differed sharply on it. Compare Rose v. Lundy, 455 U. S. 509, 543 (1982) (Stevens, J., dissenting) ("claims of constitutional error are not fungible"), with Stone v. Powell, supra, 428 U. S., at 515 (Brennan, J., dissenting) ("there are no 'second class' constitutional rights for purposes of federal habeas jurisdiction"). This step would have been completely unnecessary if the question had merely been a choice between sentencing and §2255 as the vehicle for the challenge. Under well-settled law, there is nothing remotely unconstitutional about channeling constitutional challenges to one procedural vehicle rather than another. See, e.g., Yakus v. United States, 321 U. S. 414, 443-444 (1944); United States v. Hayman, 342 U. S. 205, 223-224, and n. 40 (1952) (upholding §2255's displacement of habeas corpus).

If Custis had been merely a choice-of-procedure case, there would have been no need to distinguish Gideon claims from others, particularly on the basis of Gideon's uniqueness and jurisdictional nature. See Custis, 511 U. S., at 496. Dividing constitutional claims in this manner without a need to do so would violate the principle that constitutional questions are not to be decided unless necessary to resolve the case, see Carnival Cruise Lines, Inc. v. Shute, 499 U. S. 585, 589-590 (1991) (quoting Burton v. United States, 196 U. S. 283, 295 (1908)), and then no broader or more novel than necessary. See Greater New Orleans Broadcasting Assn., Inc. v. United States, 527 U. S. 173, 184 (1999). Custis distinguished Gideon claims from others because it needed to. It distinguished Tucker on this basis, rather than the narrower and better-settled ground of Yakus and Hayman, because the question was allowability, not timing or procedure.

"Custis argues that regardless of whether §924(e) permits collateral challenges to prior convictions, the Constitution requires that they be allowed." 511 U. S., at 493 (emphasis added). Any argument that the Constitution requires one vehicle rather than another would have been frivolous and would have been brushed off with a single paragraph. The argument made by Custis and rejected by the Court was that the Constitution requires some vehicle for attacking priors on non-Gideon grounds. "Custis invites us to extend the right to attack collaterally prior convictions used for sentenced enhancement beyond . . . Gideon. We decline to do so." Id., at 496. Both by its terms and by its rationale, this holding applies to collateral attack generally, not to the choice of one collateral vehicle over another.

At the end of the opinion, after reaching this conclusion, the Court noted that a prisoner still in custody on a prior conviction could attack that prior on habeas. See id., at 497. The Court reserved the question of whether a §2255 motion would lie to reopen an enhanced sentence after such a prior had actually been vacated in another proceeding. See ibid. Nothing in this passage says or implies that the §2255 motion itself could be the vehicle for collaterally attacking the prior. The qualification that the prisoner must still be in custody under the prior very strongly implies the contrary.

The Custis Court did not mention the possibility of habeas attack on the state prior until after it had rejected Custis's argument. Nothing in its rationale depends on the circumstance of such an attack remaining available. The question reserved in this paragraph is whether a sentence can be reopened after the prior has otherwise been vacated, not whether a motion to vacate the present sentence can be used to collaterally attack the prior. The latter question is indistinguishable from the argument rejected in Custis.

D. Law and Finality.

In a state-prisoner habeas proceeding, the law authorizes relief "only on the ground that [the petitioner] is in custody in violation of the Constitution or laws or treaties of the United States." 28 U. S. C. §2254(a). In federal prisoner cases, relief is authorized if "the sentence was imposed in violation of the Constitution or laws of the United States . . . or is otherwise subject to collateral attack . . . . " 28 U. S. C. §2255. In the context of the present case, that means the same scope as §2254. See Kaufman v. United States, 394 U. S. 217, 221-223 (1969) (§§2254 and 2255 have same scope for constitutional claims); United States v. Frady, 456 U. S. 152, 166, n. 17 (1982) (procedure default rule; finality interest same as state cases).

The question here is whether the government violates the Constitution by treating a final judgment as truly final and refusing to entertain a new attack on it. In other words, is res judicata unconstitutional? Even if Custis had not already answered the question, the answer would still be no.

The procedural default cases establish that a judgment may be regarded as legitimate despite an underlying constitutional defect. The habeas petitioner "is in custody pursuant to a judgment." Coleman v. Thompson, 501 U. S. 722, 730 (1991) (emphasis in original). If the state courts have legitimately refused to entertain an attack on that judgment because the claim has been defaulted, the judgment "rests on an adequate and independent state ground." Ibid. Despite the defect, the judgment is a "legitimate reason[ ] for holding the prisoner." Ibid. When the state has provided a means for reviewing judgments and the defendant has not used it, his custody is not in violation of the Constitution within the meaning of the habeas statute. See Brown v. Allen, 344 U. S. 443, 485 (1953).

Judgments on prior convictions should be treated similarly. When the jurisdiction rendering the judgment provides corrective process and the defendant does not invoke it within the time required, the judgment should stand as the conclusive resolution of the question of the defendant's guilt. For most constitutional claims, the end of custody will be the outer limit for bringing the claim, although the rendering jurisdiction may choose to entertain later claims. See Johnson v. Mississippi, 486 U. S. 578, 586-587 (1988) (sentence reopened after rendering jurisdiction set aside prior judgment).

In short, general principles of finality support treating final judgments as conclusive. An exception to the general principles would be warranted if there were compelling, countervailing considerations of fundamental justice. Cf. Schlup v. Delo, 513 U. S. 298, 321 (1995) (actual innocence as exception to procedural default). There are none here.

E. Basic Justice.

Faced with a claim of a procedural defect in a prior judgment, there are three possible responses: (1) entertain the claim and, if sustained, disregard the prior crime regardless of actual guilt; (2) refuse to entertain the claim, recognizing there is some possibility, however remote, defendant might be innocent of the prior crime; or (3) retry the question of guilt of the prior crime. The third alternative presents obvious difficulties of resources, stale evidence, and often great inconvenience to far away witnesses.⁽⁷⁾ If the choice is between the first two alternatives, the choice must depend in part on the degree to which a procedural defect undermines our confidence in the accuracy of the determination of guilt. Cf. Teague v. Lane, 489 U. S. 288, 313 (1989) (plurality opinion) (accuracy enhancing value as basis of distinguishing retroactive from nonretroactive rules); Strickler v. Greene, 527 U. S. 263, 289-290 (1999) (confidence in verdict as criterion for denying relief for nondisclosure of evidence).

Petitioner's claims in the present case illustrate why attacks on guilty pleas seldom undermine confidence in the defendant's actual guilt of the prior offense. He contends that his only involvement in the 1978 robbery was being in the getaway car. J. A. 52. Habeas petitioners as a group have long been notorious for saying whatever they need to say. See United States v. Hayman, supra, 342 U. S., at 217, n. 25. Daniels claims he did not know this was insufficient for guilt of robbery, his lawyer did not tell him, J. A. 53, and the transcript of the plea hearing is missing. J. A. 38, n. 3.

This story defies belief. This case does not involve some esoteric aspect of mens rea that scholars debate. This is a gut-level matter of culpability, or lack of it. Even the most uneducated person would recognize that basing guilt of a felony on merely being a passenger in a car is fundamentally wrong and would at the very least ask his lawyer about it. The likelihood that a person who was actually innocent under these circumstances would go ahead and plead guilty, and that his lawyer would allow him to, seems extremely remote.

Defendant's second attack does even less to undermine confidence in actual guilt. He claims ineffective assistance in failure to make a suppression motion. J. A. 48. Justice Powell has questioned whether such claims establish "prejudice" even in an attack on the present conviction, see Kimmelman v. Morrison, 477 U. S. 365, 394-398 (1986) (opinion concurring in the judgment), an issue which has yet to be resolved. If the case for entertaining such claims in an attack on the present conviction is weak, the case for entertaining them in attacks on priors is nearly nonexistent. Such a claim does not undermine confidence in the correctness of the present judgment at all. Defendant is a habitual criminal regardless of whether a prior attorney missed an opportunity to obtain an undeserved windfall for him.

Petitioner notes, correctly, that defendants may forego attacking a conviction at the proper time because of a plea bargain. See Brief for Petitioner 21. Indeed, this is probably the single most prevalent reason for foregoing collateral attacks on convictions. The vast majority of convictions are guilty pleas. See U. S. Dept. of Justice, Bureau of Justice Statistics, Sourcebook of Criminal Justice Statistics 1999, Table 5.51, p. 454 (2000). Far from being an argument to entertain challenges to such priors, this is a reason not to entertain them.

A typical plea bargain involves a plea to a lesser offense or fewer offenses than the defendant actually committed or a stipulation to a sentence less than he would likely receive upon full proceedings. A defendant who does not receive such consideration has no tangible incentive to plead guilty. The obvious reason why the defendant does not attack the plea-bargain judgment is that he knows he has gotten off easy. If he overturns the judgment, he will likely be tried, convicted, and sentenced for the full measure of the crimes he actually committed.

The plea bargain is a mutual agreement of the defendant and the state to bring an end to the case. The state receives a termination of the case with a savings of its resources, and the defendant receives a termination of the possibility of the full sentence for the offenses originally charged. Finality concerns have special force in guilty plea cases, United States v. Timmreck, 441 U. S. 780, 784 (1979), in part because the danger of convicting an innocent person is vastly diminished. See ibid. In addition, the defendant has chosen to hasten the finality of the case, and he has typically received consideration for it.

The flood of extra attacks on sentences that petitioner envisions, see Brief for Petitioner 23-24, is unlikely to happen. It is not true that "all prisoners would be well-advised to mount every imaginable challenge." Cf. id., at 24. Most would be subject to retrial on more charges, more serious charges, or heavier sentences if they succeeded in overturning their present judgments, and that consequence would typically outweigh the remote, contingent possibility that the conviction might someday be used for enhancement.

A person who has been wrongfully convicted has a simple and fully effective way to avoid the use of that conviction for enhancement. All he has to do is obey the law. If he does not commit a new offense, the prior will never be used for enhancement.

In summary, a fully expired prior conviction, which is attacked on grounds similar to those in the present case, and which the defendant never challenged when he had a procedure available, is highly unlikely to involve actual innocence of the prior conviction. It has no bearing whatever on guilt of the present offense and the interest in preventing injustice is greatly diminished when issues go only to sentence and not to guilt. The arguments that justify an exception to res judicata in other contexts are insufficient here. The general rule should apply. The final judgment should be regarded as conclusive and not reopened.

CONCLUSION

The decision of the Court of Appeals for the Ninth Circuit should be affirmed.

November, 2000

Respectfully submitted,

Kent S. Scheidegger

	Beginning of this file	Case Menu	Briefs Menu

6. Gideon v. Wainwright, 372 U. S. 335 (1963).

7. Notwithstanding these difficulties, amicus suggests that, if the Court should decide to allow these attacks, defendant should be required to show that he is innocent of the prior offense by the preponderance of the evidence.  
 
December 2000