I respectfully dissent.
Under our written policies in effect between 1989 and 1998, counsel appointed at any time to represent an indigent capital defendant on appeal had the specific additional obligation to undertake habeas corpus representation as well. (Supreme Ct. Policies Regarding Cases Arising From Judgments of Death (Policies), policy 3, former std. 1-1.) The Policies instruct that capital habeas corpus counsel must conduct a prompt investigation of potentially meritorious habeas corpus claims, then present such claims without substantial delay, i.e., within a reasonable time after the supporting facts were or should have been known. (Policies, stds. 1-1, 1-1.2.) If the petition’s claims are substantially delayed under this standard, they may be barred as untimely (id., std. 1-3) unless the petition establishes good cause “sufficient to justify” the delay (id., std. 1-2, italics added).1
The plurality now contradict these clear principles. On the one hand, they find that the petition in this case, filed in November 1994, was indeed *731“presented after a substantial delay.” (Plur. opn., ante, at p. 705.) That determination appears indisputable. The capital crimes occurred in 1981, and the death judgment was rendered in 1982. In January 1983, we appointed Dennis Riordan as petitioner’s counsel on appeal. We decided petitioner’s automatic appeal in 1990, and Riordan thereafter remained counsel of record until he was relieved in June 1994. As the plurality make clear, habeas corpus duties devolved upon Riordan no later than June 1989, when our Policies went into effect. Yet aside from filing a confidential request for investigation funds in October 1989, Riordan did nothing toward the investigation and filing of a petition. The current petition fails to show that the claims now proffered neither were nor should have been known in time to present them at some earlier point during Riordan’s tenure. (See Robbins, supra, 18 Cal.4th 770, 780.)
Nonetheless, the plurality hold that under our Policies, Riordan’s “abandonment” of his duties on habeas corpus is itself good cause for the substantial delay. The plurality thus hold that our Policies make the very creation of delay its own excuse. Under the plurality’s analysis, counsel appointed to pursue capital habeas corpus claims in timely fashion can prevent a finding of untimeliness simply by walking away from the duty of prompt action imposed by the appointment.
Try as they may, the plurality cannot avoid the mischievous conundrum created by this holding. If our Policies allow counsel to create timeliness by ignoring it, the requirement of timeliness itself loses all meaning.
The plurality acknowledge the importance of our timeliness Policies for capital habeas corpus petitions. These policies vindicate the strong societal interest that a valid criminal judgment become final and be carried out. They discourage abuse of the writ. And they serve our own institutional concerns with the management of our docket. The plurality also concede, as they must, that a condemned prisoner has no federal constitutional right to legal assistance, diligent or otherwise, in the investigation and presentation of a first petition for habeas corpus. Nonetheless, the plurality advance several reasons why we must excuse delay caused by counsel’s “abandonment,” and must consider on the merits all claims raised in a petition thus delayed. None of these reasons are persuasive.
First, the plurality reason that insofar as California has elected to afford appointed habeas corpus assistance to condemned prisoners, our timeliness Policies do not require us to tolerate the “abandonment” of duty by appointed counsel. In a related vein, the plurality insist that by “abandoning” *732the client, an appointed attorney leaves the client as though unrepresented by counsel. With respect to any such period of “abandonment,” the plurality assert, we must therefore apply our timeliness guidelines as we would if the petitioner had no lawyer to assist him. Moreover, the plurality suggest, we must not penalize a faultless prisoner, who reasonably believed appointed counsel was looking after his interests, by refusing to consider claims delayed as a result of counsel’s dereliction.
The ominous premise of each of these theories is one specifically rejected for federal constitutional purposes, i.e., that a capital prisoner has the right to the effective assistance of counsel on habeas corpus. (Coleman v. Thompson (1991) 501 U.S. 722, 752-754 [111 S.Ct. 2546, 115 L.Ed.2d 640]; Wainwright v. Torna (1982) 455 U.S. 586 [102 S.Ct. 1300, 71 L.Ed.2d 475] (per curiam).) Of course, the plurality concede the absence of a federal constitutional right, and they also refrain from finding any such entitlement under the California Constitution. Instead, they insinuate that because this court’s policy, and now state statutory law, call for the provision of appointed habeas corpus counsel to indigent capital prisoners, a minimal guaranty of competent assistance is implied.
Even if there were a kernel of abstract logic in this view, its application in derogation of our timeliness standards is misguided in the extreme. This is because, in the upside-down world of death penalty law, counsel more often helps than hurts his client by avoiding the prompt pursuit of postconviction remedies. Circumstances usually dictate that counsel’s best course in his condemned client’s interest is to do nothing until action becomes absolutely necessary.
As Justice Kennard has pointed out, “[m]ost defendants who challenge criminal convictions collaterally by means of a petition for writ of habeas corpus are prisoners who are serving their sentence in a penal institution. The hope of early release gives them a strong incentive to expedite the filing of a habeas petition, and they usually have little or nothing to gain by delaying the process. But for inmates on death row seeking habeas relief, the situation is otherwise. They, unlike other prisoners, have not yet begun to ‘serve’ their sentence of death. Although a successful habeas petition by an incarcerated capital defendant may produce immediate benefits in the form of release from prison, retrial, or reduction of sentence, a court’s final rejection of all habeas issues generally removes the last judicial barrier to execution. Because courts may grant stays of execution during the pendency of habeas corpus proceedings, prisoners facing a death sentence may seek to prolong their lives by ensuring that such proceedings are never finally concluded. Thus, death row inmates have an incentive to delay assertion of *733habeas corpus claims that is not shared by other prisoners.” (Clark, supra, 5 Cal.4th 750, 806 (conc. & dis. opn. of Kennard, J.).)
This case starkly illustrates the point. Events here followed an all-too-familiar pattern among our older capital matters. As the plurality explain, our Policies became effective in June 1989. We affirmed petitioner’s death judgment on appeal in September 1990, and the United States Supreme Court denied certiorari in May 1991. Riordan, as petitioner’s counsel throughout this period, “did not file a petition for a writ of habeas corpus in this court. Thereafter, the People obtained an order from the superior court authorizing petitioner’s execution. ([Pen. Code,] § 1227.) Petitioner successfully moved in federal court for a stay of the proceedings and appointment of new counsel. Newly appointed counsel subsequently filed a petition for a writ of habeas corpus in federal court on December 20, 1993. (Sanders v. Vasquez, CV-F-92-5471-REC-P.) After the People moved successfully to dismiss the petition for failure to exhaust state remedies, petitioner filed a petition for a writ of habeas corpus with this court on November 7, 1994.” (Plur. opn., ante, at p. 702.)
Riordan, one of this state’s leading specialists in representing condemned prisoners, vehemently denies his inaction had a tactical purpose. Yet by doing nothing until an execution date was set, then stepping aside while new counsel obtained a federal stay and maneuvered in federal court, Riordan actually bought his death row client the most precious of commodities— time. He not only forestalled a scheduled execution, but also substantially postponed the moment at which any final rejection of petitioner’s claims would remove them as an obstacle to the pending death sentence.2
It is just such concerns, unique to capital cases, that have informed our timeliness Policies, which specifically define the obligation of appointed capital habeas corpus counsel to investigate and present all potentially meritorious claims without substantial delay. The plurality eviscerate the Policies’ purposes by concluding that counsel’s very failure to follow these *734clear rules is its own excuse for the resulting tardiness of a petition for habeas corpus.3
The plurality are careful to admonish that delay for its own sake is not a permissible tactic. They “wam[]” that we will report future cases of “abandonment” to the State Bar and may also seek to recoup unearned fees and expenses. (Plur. opn., ante, at p. 723.) These steps are highly appropriate, but they would be so whether or not we also accepted counsel’s dereliction as grounds to excuse untimeliness. The availability and use of these sanctions should alone be ample proof that we do not tolerate counsel’s omissions. On the other hand, by holding that “abandonment” excuses the delayed presentation of a petition for habeas corpus, the plurality seriously undermine the hope that the sanctions they describe will deter counsel dedicated to prolonging the lives of their condemned clients.
Despite all this, the plurality insist their result is necessary to resolve the “tension” between “society’s desire for finality of its criminal judgments and its insistence the person being punished is actually guilty of the crimes” (plur. opn., ante, at p. 704), and to preserve the writ of habeas corpus as “a final possibility for prisoners to prove their convictions were obtained unjustly” {ibid..). Such is not the case. Our existing capital habeas corpus rules already balance the competing interests, and they provide ample assurance that meritorious claims of innocence, or of fundamental injustice, will be grounds for relief on habeas corpus whenever presented.
*735In the first place, as the plurality indicate, our long-standing practice, now codified in statutory law, is to provide counsel to assist every indigent capital defendant, not only on direct appeal, but also in the pursuit of a collateral attack on the judgment. Besides compensating such counsel for their own legal services, we furnish funds to cover the miscellaneous costs and expenses of investigating potential claims. By doing these things, we seek to promote both fairness and expedition by maximizing the incarcerated defendant’s opportunity to promptly discover, develop, and present all sound bases which may exist for relief on habeas corpus.
But even after counsel has been appointed, our timeliness Policies impose no absolute deadline or limitations period for the presentation of claims. Instead, they begin to count delay only from the time a claim was or should have been discovered. And even when a claim is not presented with reasonable promptness after that, it is still timely if there is a sound and acceptable explanation to justify the delay. Any claim that is timely by these standards may be considered on its merits, unless the claim is simply not cognizable on habeas corpus or some other procedural bar applies. For example, under the rubric of ineffective assistance of trial counsel, a timely petition can assert almost any form of garden-variety trial error.
But when claims have been unduly delayed, the state’s countervailing interests in a final and enforceable judgment, and in avoiding both habeas corpus litigation and retrial of the underlying charges, assume much greater importance. Those interests justify substantial limits on a petitioner’s right to raise collateral challenges that were postponed solely because of counsel’s inaction.
Even here, however, we have recognized that the state’s interests are not strong enough to prevent habeas corpus relief against a “fundamental miscarriage of justice.” The petitioner may establish such a fundamental miscarriage at any time by showing (1) he is actually innocent, (2) his trial included constitutional flaws so basic that they produced an irrational guilt verdict, (3) his penalty trial was so factually distorted that it produced an irrational death verdict, or (4) he was sentenced to death under an invalid law. (Clark, supra, 5 Cal.4th 750, 797-798.)
This logical, fair, and well-established balance is upset, rather than vindicated, by the plurality’s new and misguided theory, which threatens the bedrock underlying the timeliness rules themselves. I remain unpersuaded that counsel’s “abandonment” of the duties imposed by our Policies is an excuse for the resulting delay in presenting a petition for habeas corpus. I would hold that petitioner has failed to show good cause for the petition’s *736substantial delay, and that its claims are therefore untimely. Having done so, I would discharge the order to show cause.4
In In re Robbins (1998) 18 Cal.4th 770 [77 Cal.Rptr.2d 153, 959 P.2d 311] (Robbins), we provided one example of circumstances that might “justify” substantial delay in presenting claims on habeas corpus. As we noted, in order to avoid the piecemeal presentation of claims, counsel might have good cause to postpone presentation of some already developed claims while a bona fide investigation of still other potentially meritorious claims was ongoing. (Id., *731atpp. 805-806; see also In re Clark (1993) 5 Cal.4th 750, 767-770, 111 [21 Cal.Rptr.2d 509, 855 P.2d 729] (Clark).)
Of course, as noted by Justice Kennard in Clark, supra, 5 Cal.4th 750, a delayed judicial finding of actual merit in one or more of the petition’s claims might postpone the moment of petitioner’s “release from prison, retrial, or reduction of sentence.” (Id., at p. 806 (cone. & dis. opn. of Kennard, J.).) But where the alternative to such possible success is certain execution, tactical incentives still weigh strongly toward delay.
Moreover, Riordan’s inaction has provided his client with another important tactical benefit. By thus ensuring that we will not impose our own procedural bar of untimeliness against the petition’s claims, but will reject those claims, if at all, solely on the merits, Riordan has also deprived the Director of Corrections of a possible procedural defense against identical claims raised in a later federal petition. (See, e.g., Coleman v. Thompson, supra, 501 U.S. 722, 729-730 [111 S.Ct. 2546, 2554] [rule barring federal habeas corpus consideration of claims rejected by state court on independent and adequate state procedural grounds].)
In finding that appointed counsel’s entire failure to pursue efforts toward a first habeas corpus petition should excuse the petition’s substantial delay, the plurality make much of certain statements in Clark, supra, 5 Cal.4th 750. There we suggested that counsel’s incompetent failure to include all available claims in a first petition may be grounds for our consideration of an otherwise barred second petition. We reasoned in part that “[r]egardless of whether a constitutional right to counsel exists, a petitioner who is represented by counsel when a petition for writ of habeas corpus is filed has a right to assume that counsel is competent and is presenting all potentially meritorious claims.” {Id., at p. 780, italics in original.) However, Clark has no application beyond its narrow context. Clark meant, at most, that we would not bar relief on a meritorious claim simply because counsel’s inadequacy had prevented its presentation in a prior petition. Nothing in Clark suggests that meritless claims in a second petition must be denied solely on the merits simply because counsel’s incompetence led to their exclusion from a prior petition. More fundamentally, Clark's comments did not extend to the distinct issue of substantial delay, and in particular, the indefinite delay caused by counsel’s entire failure to undertake habeas corpus representation. There is ample ground to distinguish the two situations. Even if, as Clark indicated, a capital prisoner has a “right to assume” the completeness of a habeas corpus petition actually filed in his behalf, this does not imply a similar right to assume the competence of an attorney who has done nothing. Moreover, we can afford to be more lenient when piecemeal presentation is the only procedural difficulty with a meritorious claim. It is prolonged and inexcusable delay in the presentation of claims that represents the greatest threat to the fair and orderly administration of our laws governing capital punishment.
While I dissent from the plurality’s holding that habeas counsel’s “abandonment” of duty constitutes good cause for the delayed filing of a later habeas corpus petition, I heartily endorse those passages of the plurality opinion that seek, once again, to explain the limitations on counsel’s obligation to investigate and present capital habeas corpus claims. As the plurality opinion makes clear, counsel neither must nor should conduct an “unfocused” investigation (plur. opn., ante, at p. 707); may cease efforts without filing a petition if reasonable investigation discloses no potentially meritorious claims (ibid.); and should, in the interest of “ ‘effective . . . advocacy,’ ” present only those claims that are potentially meritorious, rather than all possible claims (id., at p. 713, fn. 8).