In this successive federal habeas corpus petition following his 1975 conviction for first-degree murder, appellant Kenneth David Skelton argues that the court's jury instructions called for a higher degree of doubt for acquittal than the reasonable doubt standard of In re Winship, 397 U.S. 358, 364, 90 S. Ct. 1068, 1072, 25 L. Ed. 2d 368 (1970). If his conviction were on direct appeal, the Supreme Court's decision in Cage v. Louisiana, --- U.S. ----, 111 S. Ct. 328, 330, 112 L. Ed. 2d 339 (1990), which invoked Winship to invalidate jury instructions similar to those used at Skelton's trial, would be directly controlling. Because Skelton's petition is his third filed in federal court, because it urges retroactive application of Cage, and because there was no contemporaneous objection to the jury instructions, we face instead a panoply of rules hedging the grant of federal habeas relief. We affirm the district court's order dismissing Skelton's petition.

On December 17, 1974, after drinking alcohol and smoking marijuana, Skelton and two companions decided to hitchhike and rob the first person who offered them a ride. Shortly after Larry J. Joseph gave them a ride in his car, Skelton shot him in the back of the head, killing him.

A Louisiana court sentenced Skelton to death. The Louisiana Supreme Court affirmed Skelton's conviction but ordered the trial court to resentence him to life imprisonment without benefit of parole, probation, or suspension of sentence for 20 years. See State v. Skelton, 340 So. 2d 256 (La.1976). Skelton then filed his first federal petition for a writ of habeas corpus, which was dismissed for failure to exhaust state remedies.

In 1979 and 1980, Skelton filed two more, unsuccessful habeas petitions in state court alleging ineffective assistance of counsel and prosecutorial misconduct. Skelton sought federal habeas relief on these claims, prompting the federal district court to deny his petition. Skelton returned to state court in 1985 claiming he had not shot Joseph. After a hearing was held, this application was denied.

In 1987, Skelton for the first time filed a state post-conviction motion challenging the validity of the jury instructions used at his original trial. He asserted that the trial court failed to instruct the jury as to the proper definition of reasonable doubt, effectively redefining that term to require a higher degree of doubt for acquittal.2 The trial court instructed the jury, in pertinent part, as follows:

Ladies and gentlemen, the accused, Kenneth David Skelton, is presumed by law to be innocent until each element of the crime necessary to constitute his guilt is proven beyond a reasonable doubt. The consequence of this rule of law is that he is not required to prove his innocence but may rest upon the presumption in his favor until it is overthrown by positive, affirmative proof. The burden, therefore, is upon the State to establish to your satisfaction and beyond a reasonable doubt the guilt of the accused as to the crime charged in the indictment or any lesser one included in it. Should you entertain a reasonable doubt as to the grade of the offense committed, it is your duty to find the Defendant guilty only of that grade of which you are convinced beyond a reasonable doubt he is guilty. If you entertain any reasonable doubt as to any fact or element indispensably necessary to constitute the Defendant's guilt, it is your sworn duty to give him the benefit of that doubt and return a verdict of acquittal, and even where the evidence demonstrates a probability of guilt yet if it does not establish it beyond a reasonable doubt, you must acquit the Defendant.

This doubt must be a reasonable one; that is, one founded upon a real tangible, substantial basis, and not upon a mere caprice, fancy or conjecture. It must be such a doubt as would give rise to a grave uncertainty raised in your minds by reason of the unsatisfactory character of the evidence; one that would make you feel that you have [not] an abiding conviction to a moral certainty as to the Defendant's guilt. To put it differently, you must be satisfied of the Defendant's guilt by that degree of assurance which induces a man of sound mind to act without doubt upon the conclusions to which it leads.

If, after giving a fair and impartial consideration to all the facts in the case, you find the evidence unsatisfactory upon any one single point indispensably necessary to constitute the Defendant's guilt, this would give rise to a reasonable doubt such as would justify you in rendering a verdict of not guilty.

You are prohibited by law and your oath from going beyond the evidence to seek for doubts upon which to acquit the accused, but you must confine yourselves strictly to a dispassionate consideration of the testimony upon the trial. You must not resort to extraneous facts or circumstances in reaching your verdict, nor are you at liberty to adopt unreasonable theories or suppositions in considering the evidence in order to justify a verdict of conviction.

You are to be governed exclusively by the evidence and the law as heard by you in this Court, and if upon this you find the accused guilty beyond a reasonable doubt, your duty is to so announce by your verdict. If, on the other hand, any reasonable view of the evidence shows that the Defendant is not guilty, then you should acquit.

The State has to prove every material allegation in the indictment and should the State fail to establish every ingredient necessary to conviction beyond a reasonable doubt, your sworn duty would be to acquit.

You are the exclusive judges of the facts. You find from the evidence what facts have been proven, and what facts have not been proven. For this purpose, you are the sole judges of the credibility of the witnesses and the weight to be given their testimony. You are to give that degree of credence to the testimony of a witness as you are impressed by his or her veracity....

In determining the guilt or innocence of the accused, you are only to consider those facts testified to and brought out during the trial of the case....

At the time of Skelton's trial, Louisiana law presumed that jury instructions such as this, which had been widely used by state trial courts for years, conformed to the strictures of Winship. Not until 1982 did Louisiana's highest court suggest otherwise. In State v. McDaniel, 410 So. 2d 754, 756 (La.1982), the Louisiana Supreme Court held that although portions of a trial court's jury charge were correct, the judge's definition of "reasonable doubt" as "a doubt that would give rise to a great uncertainty" and "one that would make you feel morally uncertain as to the defendant's guilt" was reversible error--even considering the jury instruction as a whole--because it may have caused the jury to overstate the degree of uncertainty required for a reasonable doubt. McDaniel also invalidated a portion of the instructions that prohibited the jury from "going beyond the evidence to seek for doubts upon which to acquit the defendant" and admonished it to confine itself to "consideration of the evidence presented upon the trial." Such a passage, the court said, was "clearly wrong" in light of La.Code Crim.P. art. 804(A)(2), which states that a reasonable doubt may arise from the lack of evidence in the case. Id.

Subsequent Louisiana Supreme Court cases, however, cast serious doubt on McDaniel. Less than a month before McDaniel, State v. Taylor, 410 So. 2d 224, 225 (La.1982), upheld a jury instruction that was virtually identical in all material aspects to the definition of reasonable doubt struck down in McDaniel. Even after announcing McDaniel, the court refused to reconsider its decision in Taylor. A more recent case, State v. Messiah, 538 So. 2d 175, 184 n. 5 (La.1988), upheld a similar jury instruction despite McDaniel:

We are not prepared to say that the use of these phrases, in the context of the whole charge, would have caused a reasonable person of ordinary intelligence difficulty in understanding the definition of 'reasonable doubt' ... we too discourage the use of such phrases as 'grave uncertainty' and 'moral uncertainty' in the reasonable doubt jury charge, but we do not regard their use, in the context of the full charge given in this case, as reaching the level of legal or constitutional error.

See also State v. Cage, 554 So. 2d 39, 41 (La.1989) (upholding jury instruction using the same phrases), rev'd and rem. sub. nom, Cage v. Louisiana, --- U.S. ----, 111 S. Ct. 328, 112 L. Ed. 2d 339 (1990), on remand, 583 So. 2d 1125 (La.1991), cert. denied, --- U.S. ----, 112 S. Ct. 211, 116 L. Ed. 2d 170 (1991). Against this backdrop, Skelton's motion attacking the validity of the jury instructions used at his trial was denied by the state district court, and the Louisiana Supreme Court later denied review. State v. Skelton, 551 So. 2d 627 (La.1989).

In early 1990, Skelton filed his third federal habeas petition, repeating some of his earlier claims and raising several new ones, including the alleged unconstitutionality of the jury instructions. While the petition was pending, the Supreme Court finally settled the lingering controversy over the Louisiana jury instructions. In Cage v. Louisiana, --- U.S. ----, 111 S. Ct. 328, 112 L. Ed. 2d 339 (1990), the Court held that a reasonable juror could have interpreted the reasonable doubt instruction to allow a finding of guilt based on a degree of proof below that required by the due process clause of the fourteenth amendment. Like Skelton, Cage had been convicted of first-degree murder in a Louisiana trial court that used an instruction defining reasonable doubt in terms of "grave uncertainty" about the character of the evidence and "moral certainty" of the defendant's guilt. Such phrases, the Court concluded, "suggest a higher degree of doubt than is required for acquittal under the reasonable doubt standard." Id. 111 S.Ct. at 329-30.

In light of Cage, the district court held that the jury instructions issued at Skelton's trial violated due process. However, because the court found the error to be harmless under Chapman v. California, 386 U.S. 18, 87 S. Ct. 824, 17 L. Ed. 2d 705 (1967), it concluded that it was unnecessary to decide whether Cage could be applied retroactively. Skelton has appealed the resulting denial of habeas relief.3

Because the federal writ of habeas corpus, although a significant safeguard of constitutional rights, dramatically challenges the states' interest in the stability and finality of criminal judgments, its availability has been tailored in three ways pertinent to this case.4 To invoke the writ, a petitioner must first have obtained a ruling from state court on the constitutional issue he asserts. A state procedural bar may arise if he did not preserve error in state court. Engle v. Isaac, 456 U.S. 107, 130, 102 S. Ct. 1558, 1572, 71 L. Ed. 2d 783 (1981). Second, he may not ordinarily file serial petitions in federal court because such repetition itself interferes with the finality of the state judgment and will constitute writ abuse. McCleskey v. Zant, --- U.S. ----, 111 S. Ct. 1454, 1470, 113 L. Ed. 2d 517 (1991). Either of these limits may be escaped, however, if the petitioner establishes "cause" for the particular failure (to preserve error or to present the issue in a first habeas petition) and actual "prejudice."5 "Cause" may exist if the petitioner invokes a rule for which there was no reasonable basis in existing law at the time his conviction became final. McCleskey v. Zant, --- U.S. ----, 111 S. Ct. 1454, 1470, 113 L. Ed. 2d 517; Reed v. Ross, 468 U.S. 1, 15, 104 S. Ct. 2901, 2910, 82 L. Ed. 2d 1 (1984). Finally, a petitioner may not seek retroactive application of a "new rule" of constitutional law to his case. Teague v. Lane, 489 U.S. 288, 109 S. Ct. 1060, 103 L. Ed. 2d 334 (1989).

Skelton's petition facially breaches all three limits on habeas corpus. Skelton did not preserve error in the state courts, and he was procedurally barred by the Louisiana courts from review on the merits of his jury instruction issue. He did not raise that issue until after his second exhausted federal habeas petition. And, he patently seeks to invoke a 1990 Supreme Court decision to cure a constitutional error that was made in his 1975 trial.

Skelton's counsel adroitly tries to avoid those stumbling blocks by clinging to their exceptions. He escapes procedural bar and writ abuse, he contends, because the Cage rule was not reasonably knowable to defense counsel in 1975: the jury instructions applied in his case had been followed for years in Louisiana without question. Thus, Skelton had "cause" not to raise the issue earlier in the state courts. On the other hand, that Cage was "new" or "not reasonably available" vis a vis procedural bar and writ abuse, he contends, is not to say that it was a "new rule" which, under Teague, may not be retroactively applied to his case. Cage was assertedly "not new" because it was "dictated by precedent," Teague, 109 S. Ct. at 1070. Alternatively, Cage falls within the exceptions to the Teague doctrine and applies retroactively despite its "newness."

The problem with this argument is that it states conclusions petitioner must reach without explaining the reason for distinguishing between "new rules" for procedural bar and writ abuse and the Teague "newness" principle. We cannot categorically say that a habeas petitioner could never steer among these shoals and prove that he had "cause" for failing to preserve an issue, in state court or an earlier habeas petition because it was legally novel, but the issue was "not new" under Teague so that it could apply retroactively to his case. It seems dubious, however, and Skelton has not really tried,6 to draw such distinctions among degrees of "newness" with any intellectual consistency.7

It seems more sensible that in nearly all instances, the "newness" of a rule has a consistent meaning for purposes of showing "cause" to avoid procedural bar and writ abuse on one hand and the Teague doctrine on the other. Thus, if a constitutional rule is "new" under Teague, it may not be retroactively applied. Of course, it is but a pyrrhic victory that a petitioner might then have escaped procedural bar or writ abuse because the rule was "new" and he had "cause" for his earlier default. Conversely, if a rule is "not new" under Teague, ordinarily it was "not new" from the standpoint of furnishing "cause" for avoiding a procedural bar or writ abuse, and the petitioner could not overcome these two stumbling blocks to habeas relief. Compare, Cuevas v. Collins, ("The question is not whether subsequent legal developments have made counsel's task easier, but whether at the time of the default the claim was 'available' at all?" 932 F.2d 1078, 1082 (5th Cir.1991), citing Smith v. Murray, 477 U.S. 527, 106 S. Ct. 2661, 2667, 91 L. Ed. 2d 434 (1986).

This discussion suggests the primacy of deciding the "newness" question in a federal habeas case that presents writ abuse or procedural bar issues in tandem with a Teague question. See Smith v. Black, 904 F.2d 950, 982 (5th Cir.1990) ("newness" question is to be answered first). As an alternative, a district court might avoid difficult analysis of "newness" if it concludes, as did the court here, that even if the "new" rule invoked by petitioner was retroactively applicable, and even if he had "cause" for not earlier raising it in state or federal court, the particular rule did not prejudice his case.8

All this said, and in view of the potentially large number of Louisiana convictions that may have been cast in doubt by Cage, we must next consider whether Cage fashioned a "new rule"--hence a non-retroactive rule--under Teague.

We conclude that it is. Although Teague did not purport to define the exact contours of "newness," the Court declared that "a case announces a new rule if the result was not dictated by precedent existing at the time the defendant's conviction became final." 109 S. Ct. at 1070 (emphasis in original). This is emphatically such a case. The jury instructions used at Skelton's trial were hardly the mean-spirited ramblings of a renegade trial judge. As the Louisiana cases cited above reveal, that state's highest court had more or less consistently upheld virtually identical jury instructions from constitutional attack. Only in 1990, when the U.S. Supreme Court set the matter straight in Cage, did it finally become clear that the Louisiana Supreme Court had misinterpreted the Winship standard as to reasonable doubt.

In retrospect, it may be tempting to criticize Louisiana Supreme Court decisions like Taylor and Messiah for upholding jury instructions that the Supreme Court has now declared to violate a constitutional precedent that is more than two decades old. Accordingly, petitioner insists that Yates v. Aiken, 484 U.S. 211, 108 S. Ct. 534, 98 L. Ed. 2d 546 (1988), stands for the proposition that "if a new decision rendered on a question of constitutional criminal procedure is merely the application of an earlier ruling that existed at the time of the defendant's conviction ... the defendant [can] take advantage of the new rule on collateral review." Skelton's argument boils down to this: Cage did not create a "new" rule of criminal procedure; it merely reversed a state court that had misapplied the well-settled rule announced in Winship.

Whatever the superficial attractiveness of this theory, and even assuming Yates somehow supports it, Skelton's argument is unpersuasive because it rests on a fundamental misconception of the purpose of collateral review. As Justice Harlan warned, and as post-Yates retroactivity doctrine decisions from Teague forward command, federal habeas corpus should not ordinarily be used to undo the good-faith judgments of state trial courts--even when, as here, those judgments are inconsistent with the Court's more recent pronouncements on prior precedents.9 In this regard, we note that the form jury instructions used in Skelton's case remained a matter of public record long after Winship was decided. There were, of course, constitutional challenges to them based on Winship during at least part of this same period. But with the exception of the Louisiana Supreme Court's 1982 decision in McDaniel--a case all but ignored in later decisions--those challenges failed, until Cage was decided in 1990. In the interim, trial courts were certainly entitled to rely in good faith on the constitutional precedents binding on them, and to interpret Winship in that light.

A post-Teague case, Butler v. McKellar, 494 U.S. 407, 110 S. Ct. 1212, 108 L. Ed. 2d 347 (1990), drives home this point. In Butler, the Court refused to apply retroactively the Roberson decision, which held that the fifth amendment bars police-initiated interrogation following a suspect's request for counsel in the context of a second investigation. See Arizona v. Roberson, 486 U.S. 675, 108 S. Ct. 2093, 100 L. Ed. 2d 704 (1988). Butler had contended that Roberson could not be a "new rule" under Teague because Roberson fell within the "logical compass" of Edwards v. Arizona, 451 U.S. 477, 101 S. Ct. 1880, 68 L. Ed. 2d 378 (1981) (applying Miranda 's right to remain silent to a situation where detectives renewed an interrogation of the accused about a series of offenses after he had requested counsel earlier in the initial interrogation). In rejecting this claim, the Butler Court defined "newness" expansively to "validate reasonable, good faith interpretations of existing precedents made by state courts even though they are shown to be contrary to later decisions." 110 S. Ct. at 1217.10 Petitioner has shown no reason why Butler, which mustered a majority of the Court (in contrast to the Teague plurality), is not controlling here.

Given that Cage creates a "new rule," we must consider whether it falls within either of the Teague exceptions. Petitioner concedes that the first exception, that a "new" rule should be applied retroactively if it places "certain kinds of primary, private individual conduct beyond the power of the criminal law-making authority to proscribe," see Teague, 109 S. Ct. at 1075, is inapplicable here. Skelton does, however, seek to avail himself of the second Teague exception by arguing that the jury instruction implicated one of those "bedrock procedural elements that must be found to vitiate the fairness of a particular conviction." Id. 109 S.Ct. at 1076.

Decisions following Teague have indicated that the relatively rare "watershed rules of criminal procedure" covered by the second exception must implicate both the fundamental fairness and accuracy of the criminal proceeding. See Saffle v. Parks, 494 U.S. 484, 110 S. Ct. 1257, 1263, 108 L. Ed. 2d 415 (1990). As this court explained in Sawyer v. Butler, 881 F.2d 1273, 1294 (5th Cir.1989) (en banc), aff'd, sub. nom., Sawyer v. Smith, 110 S. Ct. 2822, 111 L. Ed. 2d 193 (1990), the second Teague exception is limited to those rules designed to redress constitutional violations that "so distort the judicial process as to leave one with the impression that there has been no judicial determination at all, or else skew the actual evidence crucial to the trier of fact's disposition of the case." See also Smith, 904 F.2d at 986.

Petitioner asks us to classify Cage as a "watershed case" within the meaning of Teague. Accordingly, Skelton cites Winship directly:

The reasonable-doubt standard plays a vital role in the American scheme of criminal procedure. It is a prime instrument for reducing the risk of convictions resting on factual error. The standard provides concrete substance for the presumption of innocence--that bedrock 'axiomatic and elementary' principle whose enforcement lies at the foundation of the administration of our criminal law.

397 U.S. at 363, 90 S. Ct. at 1068. We cannot accept petitioner's argument. Winship was held retroactively applicable, Ivan v. City of New York, 407 U.S. 203, 92 S. Ct. 1951, 32 L. Ed. 2d 659 (1972), but Ivan is distinguishable from Cage. Winship had been adjudged delinquent in a proceeding governed by the preponderance of the evidence standard. Here, the problem is not a complete absence of the reasonable doubt standard or a deliberate reliance on an incorrect standard of proof, compare Jackson v. Virginia, 443 U.S. 307, 320 n. 14, 99 S. Ct. 2781, 2790, n. 14, 61 L. Ed. 2d 560 (1979), but rather whether the jury instruction inadvertently diluted the reasonable doubt standard to some degree.11 Both the gravity of the procedural error and the determination whether it caused an inaccurate verdict are much more ambiguous in Cage and in this case than in Winship.12

Moreover, the Court has observed the scope of Teague exceptions "must be consistent with the recognition that '[a]pplication of constitutional rules not in existence at the time a conviction became final seriously undermines the principle of finality which is essential to the operation of our criminal justice system.' Teague, 489 U.S., at 309, 109 S.Ct., at 1074 ... [T]he 'cost imposed upon the State[s] by retroactive application of new rules of constitutional law on habeas corpus thus generally far outweigh the benefits of this application.' " Sawyer v. Smith, --- U.S. ----, 110 S. Ct. 2822, 2831, 111 L. Ed. 2d 193 (1990).

In Saffle v. Parks, 494 U.S. 484, 110 S. Ct. 1257, 1264, 108 L. Ed. 2d 415 (1990), the Court stated that:

Although the precise contours of this [second] exception may be difficult to discern, we have usually cited Gideon v. Wainwright [372 U.S. 335, 83 S. Ct. 792 (1963) ] [citation omitted], holding that the defendant has the right to be represented by counsel in all criminal trials for serious offenses, to illustrate the type of rule coming within the exception.

Finally, in Teague the Court believed it "unlikely that many such components of basic due process have yet to emerge." Teague, 489 U.S. at 313, 109 S.Ct. at 1076. These passages highlight the critical difference between Winship and Cage. The failure to give a reasonable doubt instruction would seriously diminish the likelihood of obtaining an accurate conviction. Errors in an instruction long used, seldom challenged, and promulgated with good intentions, however, should not lightly be held to have violated fundamental constitutional norms in a way that requires retroactive habeas relief. See also Carriger v. Lewis, 948 F.2d 588 (9th Cir., Nov. 4, 1991) (Beck v. Alabama, 447 U.S. 625, 100 S. Ct. 2382, 65 L. Ed. 2d 392 (1980), set forth a rule that enhances the reliability of capital sentencing but is "new" under Teague and is not retroactively applicable under Teague 's second exception).

Although we are convinced that Teague does not permit retroactive application of Cage to Skelton's case, we agree with the district court that even if Cage applied, Skelton cannot demonstrate the prejudice required to overcome his abuse of writ and procedural default. "Prejudice" for such purposes has been interpreted, in the precise context of a jury charge issue, to mean showing

not merely that the errors at his trial created a possibility of prejudice, but that they worked to his actual and substantial disadvantage, infecting his entire trial with error of constitutional dimensions.

United States v. Frady, 456 U.S. 152, 171, 102 S. Ct. 1584, 1596, 71 L. Ed. 2d 816 (1982). No such prejudice exists here. There is not the least likelihood that a different reasonable doubt instruction would have resulted in his exoneration of murder, and if any doubt about the extent of his guilt existed, it was cured by the Louisiana Supreme Court's reducing his sentence, on other grounds, from the death penalty to a life sentence. Skelton's co-defendants vividly testified that he planned Joseph's robbery and deliberately, unnecessarily shot him at point blank range in the back of the head. Any assertion of prejudice is disingenuous.

The Cage error that was made in instructing Skelton's jury was not retroactively applicable to afford him habeas relief, because the error was a "new rule" under the standard of Teague and did not fall within its second exception for watershed rules of criminal procedure. We do not reach the other grounds for affirmance raised by the state. The judgment of the district court dismissing Skelton's petition for habeas corpus relief is AFFIRMED.