I concur that the guilt judgment must be affirmed. However, I respectfully dissent from the reversal of the special circumstance and penalty judgments.
I dispute, as a matter of law, the majority’s conclusion that defendant may challenge the 1971 Illinois murder conviction proved against him as a special circumstance (Pen. Code, § 190.2, subd. (a)(2)) on grounds that his appointed counsel’s voluntary absence during one trial day in the long-final Illinois case, and his consequent temporary representation by his codefendant’s attorney, rendered the Illinois conviction constitutionally invalid. Our cases have recognized certain fundamental defects which may render a prior conviction invalid as a basis for enhancing the sentence in a later case. However, no prior decision of this court has imposed upon California trial or appellate courts the burdensome duty to evaluate detailed claims of mere “trial error” in a prior unrelated criminal proceeding.
Recent United States Supreme Court authority makes clear that the federal Constitution does not require examination of the validity of a prior conviction beyond a claim that counsel was completely denied. Contrary to the majority’s suggestion, there is no indication from the high court that a different rule applies in capital cases.
*1142Moreover, the error the majority identify as a basis for striking defendant’s 1971 Illinois conviction is a particularly unwarranted ground for such relief, since the same challenge was rejected on defendant’s direct appeal from the Illinois judgment. I find no basis for allowing an accused to attack his prior conviction upon grounds previously considered and rejected on direct or collateral appellate review of the prior case. The majority’s contrary view violates compelling principles of finality and comity, and it may also place California trial courts in the unseemly position of second-guessing final appellate decisions from California or elsewhere.
Accordingly, I reject the analysis by which the majority strike defendant’s prior murder conviction as a special circumstance. My reasons follow in greater detail.
In Gideon v. Wainright (1963) 372 U.S. 335 [9 L.Ed.2d 799, 83 S.Ct. 792, 93 A.L.R.2d 733] (Gideon), the United States Supreme Court held that the state court conviction of an indigent is constitutionally invalid, whenever and wherever obtained, if counsel was neither appointed for the accused nor waived by him. This court soon acknowledged that if a conviction was invalid under Gideon, it could not be used as a statutory sentencing factor in a later case. (In re Woods (1966) 64 Cal.2d 3, 5-6 [48 Cal.Rptr. 689, 409 P.2d 913]; see also People v. Merriam (1967) 66 Cal.2d 390, 397 [58 Cal.Rptr. 1, 426 P.2d 161] [“One seeking to challenge prior convictions charged against him may do so only through a clear allegation . . . that. . . he neither was represented by counsel nor waived the right to be so represented.”], italics omitted.) Such use, we concluded, could be challenged either by a habeas corpus petition after sentencing in the later proceeding (In re Woods, supra, 64 Cal.2d at p. 5), or during the proceeding itself by a motion to strike the prior conviction (People v. Coffey (1967) 67 Cal.2d 204, 215 [60 Cal.Rptr. 457, 430 P.2d 15]).
Eventually, the United States Supreme Court itself recognized that a conviction rendered in violation of Gideon may not constitutionally influence the sentence imposed for a later offense. (United States v. Tucker (1972) 404 U.S. 443, 447-449 [30 L.Ed.2d 592, 596-597, 92 S.Ct. 589]; Burgett v. Texas (1967) 389 U.S. 109 [19 L.Ed.2d 319, 88 S.Ct. 258].) Any other result, the court indicated, would “erode” the principle of Gideon that every conviction obtained without affording an indigent accused any assistance of counsel is “void.” (Burgett v. Texas, supra, 389 U.S. at pp. 115-116 [19 L.Ed.2d 324-326]; see also United States v. Tucker, supra, 404 U.S. at p. 449 [30 L.Ed.2d at pp. 597-598].)
In People v. Sumstine (1984) 36 Cal.3d 909 [206 Cal.Rptr. 707, 687 P.2d 904] we extended California’s motion-to-strike procedure, as established in *1143People v. Coffey, supra, 67 Cal.2d 204, to claims that a prior conviction was based on a guilty plea entered in ignorance of the constitutional trial rights thereby waived. (See Boykin v. Alabama (1969) 395 U.S. 238 [23 L.Ed.2d 274, 89 S.Ct. 1709]; In re Tahl (1969) 1 Cal.3d 122 [81 Cal.Rptr. 577, 460 P.2d 449].) And, in Curl v. Superior Court (1990) 51 Cal.3d 1292 [276 Cal.Rptr. 49, 801 P.2d 292], we held that a prior murder conviction, alleged as a special circumstance in a capital case, could be challenged on Boykin/ Tahl grounds in the capital proceeding. Despite dictum in Sumstine and other decisions to the effect that a prior conviction may be challenged on “any” constitutional grounds, we have never explicitly applied the principles of Coffey and Sumstine except to claims of complete denial of counsel and involuntary plea.1
The difficulties of allowing more extensive challenges are obvious. The record and evidence necessary to determine whether a prior case involved a complete denial of counsel or an involuntary plea are often not unduly complex. But the judicial burden of examining a recidivist’s prior record for less fundamental defects is great. The danger arises that each charged prior would thus become the basis for its own mini-appeal, in which every arguable misstep in the prior case could be asserted as “constitutional” error.
Carried to its logical end, such a rule would require, as here, the full examination of trial records in cases from any state or federal jurisdiction, long past and long final for all other purposes. Claims of “constitutionally” ineffective assistance might further require the taking of new evidence at far-flung locations, decades after witnesses have died and memories faded. The current trial would have to be postponed while the accused was given the chance to assemble evidence against his prior convictions. The delays and difficulties inherent in such a system would seriously undermine the orderly administration of justice and would jeopardize the numerous provisions which enhance punishment for recidivism.
Any inference that such burdensome and time-consuming procedures are required by the federal Constitution was recently repudiated by the United *1144States Supreme Court. In Custis v. United States (1994)_U.S. _ [128 L.Ed.2d 517, 114 S.Ct. 1732], one convicted of a federal firearms offense faced a maximum sentence of life without parole on the basis of charges that he had suffered three previous state felony convictions. He claimed the prior convictions were constitutionally invalid because they stemmed from ineffective assistance of counsel and involuntary pleas. Both the district court and the court of appeals ruled that these challenges were not cognizable in a federal sentencing proceeding. By a vote of six to three, the Supreme Court affirmed.
At length, the Custis majority dismissed defendant Custis’s contention that the federal recidivism statute contemplates collateral attacks on the validity of prior convictions used for sentencing. The majority then rejected the notion that any and all such challenges must be allowed as a matter of constitutional right. The majority acknowledged the decisions in Burgett v. Texas, supra, 389 U.S. 109, and United States v. Tucker, supra, 404 U.S. 443, holding that prior convictions are subject to collateral attack for Gideon violations. But the majority declined defendant Custis’s invitation to extend the grounds for such challenges.
The Custis majority said, “We think that since the decision in Johnson v. Zerbst [(1938) 304 U.S. 458 (82 L.Ed. 1461, 58 S.Ct. 1019, 146 A.L.R. 357)] [recognizing right to assistance of appointed counsel in federal prosecutions] more than half a century ago, and running through our decisions in Burgett and Tucker, there has been a theme that failure to appoint counsel for an indigent defendant was a unique constitutional defect. Custis attacks his previous convictions claiming the denial of the effective assistance of counsel, that his guilty plea was not knowing and intelligent, and that he had not been adequately advised of his rights in opting for a ‘stipulated facts’ trial. None of these alleged constitutional violations rises to the level of a jurisdictional defect resulting from the failure to appoint counsel at all. [Citation.]
“Ease of administration also supports the distinction. As revealed in a number of the cases cited in this opinion, failure to appoint counsel at all will generally appear from the judgment roll itself, or from an accompanying minute order. But determination of claims of ineffective assistance of counsel, and failure to assure that a guilty plea was voluntary, would require sentencing courts to rummage through frequently nonexistent or difficult to obtain state court transcripts or records that may date from another era, and may come from any one of the 50 States.
“The interest in promoting the finality of judgments provides additional support for our constitutional conclusion. As we have explained, ‘[ijnroads on the concept of finality tend to undermine confidence in the integrity of *1145our procedures’ and inevitably delay and impair the orderly administration of justice. [Citations.] We . . . noted in Parke v. Raley, 506 U.S. 20, 113 S.Ct. 517, . . . (1992), that principles of finality associated with habeas corpus actions apply with at least equal force when a defendant seeks to attack a previous conviction used for sentencing. By challenging the previous conviction, the defendant is asking a district court ‘to deprive [the] [state court judgment] of [its] normal force and effect in a proceeding that ha[s] an independent purpose other than to overturn the prior judgment].’ [Citation.]” (Custis v. United States, supra,_U.S. at pp._-_[128 L.Ed.2d at pp. 528-529, 114 S.Ct. at pp. 1738-1739], italics added, last paragraph text brackets in Custis.)
Though Custis dealt with federal sentencing and the relation of federal to state courts, its constitutional analysis is clear and persuasive in all contexts.2 Apparently chastened by the force of Custis’s reasoning, the majority here wisely avoid adopting a blanket rule that all “constitutional” challenges against prior convictions are permissible in all criminal cases under California law.
Nonetheless, the majority suggest that when a prior conviction is charged for purposes of death eligibility in a capital case, an attack upon the prior conviction’s validity should not be limited to Gideon error, but may include certain other “fundamental” constitutional defects, such as defendant’s claim that he was deprived of counsel at a “critical stage” of his 1971 murder trial. The majority submit that because of the heightened reliability concerns in a capital trial, a capital defendant must be allowed to raise any fundamental constitutional error in a prior conviction alleged against him as a basis for the death penalty. But the United States Supreme Court decision on which the majority rely for this principle is inapposite.
In Johnson v. Mississippi (1988) 486 U.S. 578 [100 L.Ed.2d 575, 108 S.Ct. 1981] the defendant was sentenced to death for a 1981 murder. The Mississippi death judgment was predicated in part on evidence that he had sustained a 1963 New York felony conviction for assault. During the trial *1146and direct appeal in the Mississippi case, defendant Johnson never attempted to challenge the validity of the New York conviction. The Mississippi Supreme Court affirmed the death judgment.
Though Johnson did not attack his New York conviction in the original Mississippi proceedings, he had previously made several attempts in New York, without assistance of counsel, to appeal that conviction on grounds that his confession was coerced. All such efforts had been rejected as untimely. After the Mississippi death judgment was rendered, however, Johnson’s attorneys commenced a new attack in New York’s courts against the 1963 conviction. Johnson’s new motion asserted that he had never been informed of his right to appeal in the 1963 case. Ultimately, his claim prevailed, and his 1963 conviction was reversed by the New York Court of Appeals.
Johnson then moved in the Mississippi Supreme Court for postconviction relief from the death judgment on grounds that the now invalid New York conviction could not serve as an aggravating circumstance. The motion was denied.
On certiorari, the United States Supreme Court noted that the issue was whether the Eighth Amendment’s reliability requirements for capital cases would be violated by “allowing [Johnson’s] death sentence to stand although based in part on a reversed conviction.” (Johnson v. Mississippi, supra, 486 U.S. 578, 585 [100 L.Ed.2d 578, 584], italics added.) As the high court explained, the Mississippi prosecutor had used the 1963 New York conviction as the sole proof that Johnson committed the criminal conduct underlying the conviction. “Since that conviction has been reversed,” said the court, “unless and until [Johnson] should be retried, he must be presumed innocent of that charge. Indeed, even without such a presumption, the reversal of the conviction deprives the prosecutor’s sole piece of documentary evidence of any relevance to Mississippi’s sentencing decision.” (Ibid., italics added.)
Thus, Johnson establishes only the unremarkable principle that a prior conviction which no longer exists for whatever reason cannot constitute evidence of criminality in support of a death judgment. Nothing in Johnson’s analysis supposes that a conviction which remains in full force and effect nonetheless becomes irrelevant in a later capital case if then found to be tainted with constitutional error. And Johnson nowhere suggests that a state court in which capital proceedings are pending must entertain the defendant’s motion to preclude use of a prior conviction, otherwise fully extant, on such grounds.
Indeed, the holding of Johnson is entirely consistent with the discussion of state remedies more recently set forth in Custis v. United States, supra,_ *1147U.S. [128 L.Ed.2d 591, 528-529, 114 S.Ct. 1732, 1739]. Both cases simply recognize that if the accused succeeds in overturning a prior conviction for all purposes, the extinct conviction, as such, cannot influence the sentencing decision in a later case.
The limited nature of Johnson's holding is further indicated by the court’s response to the state’s separate contention that no federal question was presented in that case. The state asserted that by failing to challenge the validity of his 1963 conviction during the trial and direct appeal in the capital case, defendant Johnson had failed to make a timely claim under Mississippi law, and this was an “independent state ground” for Mississippi’s refusal to grant postconviction relief.
In rejecting this procedural argument, Johnson merely invoked the familiar rule that federal jurisdiction will not be defeated by a state procedural bar unless the state itself has applied the bar “consistently or regularly.” (Johnson v. Mississippi, supra, 486 U.S. at p. 587 [100 L.Ed.2d at pp. 585-586].) The Mississippi Supreme Court, Johnson noted, had ruled in both capital and noncapital cases that a sentencing proceeding was never the appropriate forum to litigate the validity of a prior conviction, and that the prior must be attacked in the jurisdiction which rendered it. Hence, Johnson reasoned, Mississippi courts had not “consistently or regularly” applied the bar of timeliness which the state now sought to assert. (Id. at pp. 587-589 [100 L.Ed.2d at pp. 585-587].)
By proceeding in this narrow fashion, Johnson avoided expressing any opinion about whether the views of the Mississippi Supreme Court were constitutionally correct. Hence, Johnson is no authority for the instant majority’s view that capital defendants have a constitutional right, above and beyond that of noncapital defendants, to challenge the constitutional validity of prior convictions which have not otherwise been extinguished.
I would find no such right applicable to this case. Counsel was appointed for defendant in his 1971 Illinois murder case, and he exercised his right to a trial on the issue of his guilt. He presumably received the assistance of counsel with respect to all motions, evidence, instructions, and argument. The Illinois jury then found him guilty of murder beyond a reasonable doubt. The conviction remains in existence to this day, having never been reversed, vacated, or expunged. Because final, it carries a high presumption of accuracy and validity. Under these circumstances, I cannot conclude that the Eighth Amendment nonetheless obliges California courts, at this late date, to examine the trial record of the prior final conviction for constitutional error, “fundamental” or otherwise, before allowing its use as a special circumstance.
*1148There is a further reason why the flaw the majority find in the 1971 conviction should not preclude use of that judgment in this proceeding. As the majority concede, the same claim of error was raised and rejected on direct appeal in the 1971 case. (People v. Horton (1976) 43 Ill.App.3d 150 [1 Ill.Dec. 762, 356 N.E.2d 1044].) The majority’s acceptance of defendant’s claim is nothing less than a direct refusal to effectuate the long-final appellate determination of a sister jurisdiction.
In re Harris (1993) 5 Cal.4th 813 [21 Cal.Rptr.2d 373, 855 P.2d 391] recently affirmed that a petition for habeas corpus may not raise claims already considered and rejected on appeal unless they (1) are based on an intervening new rule of law (id. at pp. 841-841), (2) legitimately assert “jurisdictional” defects in the judgment under attack (id. at pp. 836-841), or (3) allege constitutional error, “both clear and fundamental, [which] strikes at the heart of the trial process” (id. at p. 834, citing Arizona v. Fulminante (1991) 499 U.S. 279, 309 [113 L.Ed.2d 302, 331, 111 S.Ct. 1246] [only errors amounting to a structural defect in the trial mechanism deserve automatic reversal]). The general rule against reconsideration of appellate claims stems from the presumption that an appeal, where afforded, is usually an adequate remedy and that judgments must eventually become final. (In re Harris, supra, 5 Cal.4th 813, 824-834.) Nothing in Harris indicates that more lenient terms should apply to habeas corpus petitions attacking judgments of death.
The majority do not suggest that the flaw they find in defendant’s 1971 conviction would be cognizable on habeas corpus under any of the Harris exceptions. Specifically, the majority make no claim, nor could they, that the temporary absence of defendant’s counsel from the 1971 trial, during which the codefendant’s attorney undertook to represent both of the accused in resolving a jury deadlock, rendered the 1971 proceeding so pervasively unfair as to constitute a “structural [defect] in the . . . trial mechanism” (Arizona v. Fulminante, supra, 499 U.S. 279, 309 [113 L.Ed.2d 302, 331]) or a blow “at the heart of the trial process” (In re Harris, supra, 5 Cal.4th 813, 834).
Instead, the majority invoke the principle that procedural limitations on habeas corpus do not necessarily govern motions to preclude use of a prior conviction in a later sentencing proceeding. Moreover, the majority suggest, a capital defendant in particular must be allowed to challenge a prior conviction, even on grounds previously rejected, when “an error in an appellate court’s prior review of an alleged constitutional violation appears on the face of the judgment itself, and the claimed violation is tantamount to a complete denial of representation at a critical trial stage.” (Maj. opn., ante, p. 1138.)
*1149The majority’s first premise is drawn from People v. Sumstine, supra, 36 Cal.3d 909. There we rejected the argument that a motion to strike a prior conviction, where otherwise appropriate, should be limited by considerations of timeliness which apply to petitions for habeas corpus. We noted that the purpose of a motion to strike is not to vacate the prior conviction entirely, but merely to challenge the “present effect” of the conviction. “As the state is the party proposing to assert the effect of the prior conviction at the current trial,” we reasoned, “the state should also be prepared to face challenges to it.” (Id. at p. 921, fn. omitted.)
Whatever the force of that rationale on the issue of timeliness, it hardly explains why a motion to strike may relitigate claims already specifically rejected in the prior case. No reason appears why the state, when seeking to use the prior conviction in a later trial, must face challenges to that conviction which have already been considered and dismissed. On the contrary, as in habeas corpus, the fact that the defendant’s claim was raised and expressly rejected on appeal should constitute a particularly strong bar to subsequent litigation.
Moreover, intervening law casts grave doubt on Sumstine’s general proposition that challenges to the validity of prior convictions deserve more lenient procedural treatment than habeas corpus petitions. As noted above, the majority in Custis v. United States, supra,_U.S._[128 L.Ed.2d 517, 114 S.Ct. 1732] has made clear that “principles of finality associated with habeas corpus actions apply with at least equal force when a defendant seeks to attack a previous conviction used for sentencing. By challenging the previous conviction, the defendant is asking a district court ‘to deprive [the] [state court judgment] of [its] normal force and effect in a proceeding that ha[s] an independent purpose other than to overturn the prior júdgmen[t].’ [Citation.]” (Id. at p. _ [128 L.Ed.2d at p. 529, 114 S.Ct. at p. 1739], brackets in Custis.) Similar considerations apply when California courts are asked, in unrelated proceedings, to deny such effect to final appellate judgments rendered here or elsewhere.
The logical weight of such conclusions again forces the majority into retreat. As a “fallback,” they suggest that the second-guessing of a prior appellate judgment may nonetheless be appropriate when the challenge is raised in a capital case, the nature of the claim is “tantamount to a complete denial of representation at a critical trial stage,” and the prior appellate resolution of the issue is wrong “on the face of the judgment.” (Maj. opn., ante, p. 1138.)
Even were I to accept these dire characterizations of the 1971 conviction, which I do not, I would find the majority’s position unpersuasive. The *1150majority cite no authority for their narrow rule, and it seems but an ipse dixit conveniently designed to allow consideration of the claim on which the majority base their reversal.
On its own merits, moreover, the majority’s rule is not convincing. As previously noted, nothing in the special nature' of capital cases requires us to adopt procedures that would be unworkable and mischievous in other contexts. With the possible exception of Gideon claims, the notion that final appellate judgments should ever be reexamined on their merits in later sentencing matters violates the considerations set forth in Custis v. United States, supra,_U.S__[128 L.Ed.2d 517, 114 S.Ct. 1732]. In my view, for the reasons stated in Custis, any such procedure constitutes an unwarranted refusal to accord such final judgments their normal force and effect upon defendant’s sentence for later crimes. Hence, I cannot join in the majority’s effort.3
The majority may feel that we are entitled to take such action in this appeal, as the highest court of California reviewing a California death judgment. But the majority overlook the implications of their ruling for future capital cases in this state. Hereafter, whenever a motion to strike filed in a California capital trial meets the majority’s criteria for reconsideration of a prior appellate determination, the trial court will face the awkward and unseemly task of passing judgment on the reasoning and result of an appellate court.
Because both California and foreign appeals will presumably be subject to review under the majority’s rule, a future California trial court may actually be obliged to decide, on the basis of the same record we previously considered, whether this court committed a “fundamental” constitutional error in a prior case. I do not envy any trial court put in such a position.
Finally, the majority suggest that allowing motions to strike in these limited circumstances will conserve judicial resources by eliminating the need for post-trial habeas corpus petitions which would ultimately prevail in either state or federal court. The majority miss the point. If a prior conviction cannot be attacked during the trial and appeal by means of a motion to strike, *1151such an attack should be no more cognizable by means of a later habeas corpus petition. The overarching principle, which I believe both state and federal courts must apply after Custis, is that the Constitution contemplates no basis for a sentencing attack on a prior conviction, except where the defendant asserts a Gideon-style claim that counsel was completely withheld in the prior case.
I would therefore affirm the judgment in its entirety.
Lucas, C. J., concurred.
In People v. Wharton (1991) 53 Cal.3d 522 [280 Cal.Rptr. 631, 809 P.2d 290], we rejected, on the merits, defendant’s claim that his prior murder conviction, alleged as a special circumstance in a capital case, was invalid because the trial court in the prior case had failed, sua sponte, to order a hearing on his competency to plead guilty, or to allow withdrawal of his plea on grounds of incompetency. (Id. at pp. 583-585.) Like Boykin/Tahl challenges, such a claim goes to the voluntariness, in the constitutional sense, of a waiver of trial. In People v. Coleman (1969) 71 Cal.2d 1159 [80 Cal.Rptr. 920, 459 P.2d 248], after reversing both guilt and penalty judgments in a capital case on other grounds, we briefly noted that on retrial, defendant could employ the procedures of People v. Coffey, supra, 67 Cal.2d 204, to challenge use of his prior Virginia burglary conviction on grounds that he received ineffective assistance in the Virginia case. The nature of the ineffective-assistance claims, however, was not specified. (People v. Coleman, supra, 71 Cal.2d at p. 1169.)
After holding unequivocally that there is no constitutional right to challenge the validity of a prior conviction in a federal sentencing proceeding, except with respect to Gideon claims, the Custis majority remarked that because the defendant in that case was still “in custody” on his state convictions at the time of his federal sentence, he retained the right to attack those convictions in state habeas corpus proceedings. If successful in vacating the state-court judgments, the Custis majority surmised, the defendant might then be permitted to reopen the federal sentencing proceeding on the premise that his federal sentence had been influenced by nonexistent priors. (_U.S. at p._[128 L.Ed.2d at pp. 528-529, 114 S.Ct. at p. 1739].) Contrary to implications in the instant majority opinion, nothing in these comments suggests that the Custis majority relied on the availability of state remedies in that case as a basis for its constitutional decision.
The majority suggest that issues raised and rejected in a prior appeal are nonetheless reviewable on a subsequent motion to strike in a capital case where an error, affecting “fundamental” constitutional rights, is “apparent from the [prior] appellate opinion itself’ (maj. opn., ante, p. 1138) or “on the face of the [prior] judgment itself’ (ibid.). In reality, however, the majority’s test carries them well beyond the face of the prior opinion and judgment. Indeed, as they concede, they must review the transcripts of the 1971 proceeding to discover the relevant trial facts and to reach their conclusion that those facts were stated and applied incorrectly by the Illinois appellate court.