dissenting:
Because today’s decision is counter to the explicit commands of the Supreme Court in Penry v. Lynaugh, 492 U.S. 302, 109 S.Ct. 2934, 106 L.Ed.2d 256 (1989) (Penry I), and is inconsistent with the logic which underlies that decision, I respectfully dissent.
I.
The majority denies petitioner Mark Robertson habeas corpus relief because it believes the Texas Special Issues provided the jury an adequate vehicle for registering its moral response to Robertson’s evidence of childhood abuse in making its death penalty determination.1 In Penry I, however, the Supreme Court held that the special issues are insufficiently capacious to encompass the kind of evidence Robertson offers here, making today’s decision contrary to Supreme Court precedent. Moreover, even assuming arguendo that the Court left “unplumbed” the issue of whether evidence of childhood abuse alone is adequately considered within the special issues, Penry /’s reasoning dictates finding the special issues insufficient here. While Robertson did receive a “nullification instruction” that was not received by Penry, Penry v. Johnson, 532 U.S. 782, 803-04, 121 S.Ct. 1910, 150 L.Ed.2d 9 (2001) (Penry II), makes clear this instruction did not resolve the Penry I problems present here.
A. The Penry Decisions
In Penry I, the Supreme Court held that (1) “at the time Penry’s conviction became final, it was clear from [Lockett v. Ohio, 438 U.S. 586, 98 S.Ct. 2954, 57 L.Ed.2d 973 (1978),] and [Eddings v. Oklahoma, 455 U.S. 104, 102 S.Ct. 869, 71 L.Ed.2d 1 (1982),] that a State could not, consistent with the Eighth and Fourteenth Amendments, prevent the sentencer from considering and giving effect to evidence relevant to the defendant’s background or character or to the circumstances of the offense that mitigate against* imposing the death penalty,” 492 U.S. at 318, 109 S.Ct. 2934; (2) “[t]he rule Penry [sought] — that when such mitigating evidence [of his mental retardation and abused childhood] is presented, Texas juries must ... be given jury instructions that make it possible for them to give effect to that mitigating evidence in determining whether the death penalty should be imposed — is not a ‘new rule’ under Teague because it is dictated by Eddings and Lockett,” id. at 318-19, 109 S.Ct. 2934; (3) “[underlying Lockett and Eddings is the principle that punishment should be directly related to the personal culpability of the criminal defendant,” id. at 319, 109 S.Ct. 2934; (4) “[I]t is not enough simply to allow the defendant to present mitigating evidence to the sen-tencer. The sentencer must also be able to consider and give effect to that evidence *275in imposing sentence,” id.; (5) “[i]n order to ensure reliability in the determination that death is the appropriate punishment in a specific case, the jury must be able to consider and give effect to any mitigating evidence relevant to a defendant’s background and character or the circumstances of the crime,” id. at 328, 109 S.Ct. 2934; and (6) therefore, “in the absence of instructions informing the jury that it could consider and give effect to the mitigating evidence of Penry’s mental retardation and abused [childhood] background by declining to impose the death penalty, ... the jury was not provided with a vehicle for expressing its reasoned moral response to that evidence in rendering its sentencing decision,” id. at 328, 109 S.Ct. 2934. (Internal quotations and citations omitted).
Thus, the Supreme Court in Penry I agreed with Penry’s argument “that his mitigating evidence of mental retardation and childhood abuse has relevance to his moral culpability beyond the scope of the special issues, and that the jury was unable to express its reasoned moral response to that evidence in determining whether death was the appropriate punishment.” Id. at 322, 109 S.Ct. 2934. The Court explained in detail why it rejected the State’s contrary argument that the jury was able to consider and give effect to all of Penry’s mitigating evidence in answering the three special issues. Id.
The first special issue, which asked whether the defendant acted “deliberately and with the reasonable expectation that the death of the deceased ... would result,” impermissibly limited the jury’s function because the term “deliberately” had not been defined by the Texas Legislature, the Texas Court of Criminal Appeals, or the trial court’s instructions. Id. at 322, 109 S.Ct. 2934. Even if the jurors “understood ‘deliberately’ to mean something more than ... ‘intentionally’ committing murder, those jurors may still have been unable to give effect to Penry’s mitigating evidence in answering the first special issue.” Id. The reason was “deliberately” was not defined “in a way that would clearly direct the jury to consider fully Penry’s mitigating evidence as it bears on his personal culpability.” Id. at 323, 109 S.Ct. 2934. Consequently, the Court concluded, unless there are “jury instructions defining ‘deliberately’ in a way that would clearly direct the jury to consider fully Penry’s mitigating evidence as it bears on his personal culpability, we cannot be sure that the jury was able to give effect to the mitigating evidence of Penry’s mental retardation and history of abuse in answering the first special issue.” Id. at 323, 109 S.Ct. 2934. “Thus, we cannot be sure that the jury’s answer to the first special issue reflected a reasoned moral response to Penry’s mitigating evidence.” Id. (internal quotation omitted).
The second special issue, which asked “whether there is a probability that the defendant would commit criminal acts of violence that would constitute a continuing threat to society,” permitted the jury to consider and give effect to Penry’s mental retardation and childhood abuse as “relevant only as an aggravating factor[.]” Id. The second special issue was inadequate both because it only gave effect to Penry’s evidence as an aggravating factor, and because it did not allow the jury to give full effect to Penry’s mitigating evidence. Id. at 323, 109 S.Ct. 2934. Thus, the Court concluded that Penry’s evidence of mental retardation and childhood abuse was a “two-edged sword,” diminishing “his blameworthiness for his crime even as it indicates that there is a probability that he will be dangerous in the future.”2 Id. at 324,109 S.Ct. 2934.
*276As a result the majority held, “in the absence of instructions informing the jury that it could consider and give effect to the mitigating evidence of Penry’s mental retardation and abused background by declining to impose the death penalty, ... the jury was not provided with a vehicle for expressing its ‘reasoned moral response’ to that evidence in rendering its sentence.” Id. at 328,109 S.Ct. 2934.
In Penry II the Court again confronted the constitutionality of Penry’s death sentence, this time after re-sentencing in the wake of Penry I. Texas attempted to correct the defects the Court identified in Penry I with a supplemental instruction to the three special issues. This so-called “nullification instruction” said:
You are instructed that when you deliberate on the questions posed in the special issues, you are to consider mitigating circumstances, if any, supported by the evidence presented in both phases of the trial, whether presented by the state or the defendant. A mitigating circumstance may include, but is not limited to, any aspect of the defendant’s character and record or circumstances of the crime which you believe could make a death sentence inappropriate in this case. If you find that there are any mitigating circumstances in this case, you must decide how much weight they deserve, if any, and therefore, give effect and consideration to them in assessing the defendant’s personal culpability at the time you answer the special issue. If you determine, when giving effect to the mitigating evidence, if any, that a life sentence, as reflected by a negative finding to the issue under consideration, rather than a death sentence, is an appropriate response to the personal culpability of the defendant, a negative finding should be given to one of the special issues.
Penry II, 532 U.S. at 789-790, 121 S.Ct. 1910.
The Penry II Court explained that there were two interpretations of this instruction, neither of which resolved the concerns it identified in Penry I. First, the Court noted the instruction may have told jurors to consider Penry’s mitigating evidence within the special issues. But such an interpretation left “the jury in no better position than was the jury in Penry I,” because “none of the special issues is broad enough to provide a vehicle for the jury to give mitigating effect to the evidence of Penry’s mental retardation and childhood abuse.” Id. at 798, 121 S.Ct. 1910. Alternatively, the instruction may have asked the jury to answer “no” to a special issue if it believed Penry did not deserve the death penalty, regardless of its honest answer to the question. This interpretation was also constitutionally unsound because it would require jurors to violate their oath to render a “true verdict” to give effect to Penry’s evidence, putting jurors in a logical and ethical bind. Id. at 799-800, 121 S.Ct. 1910. Thus, the Penry II majority concluded, the Texas court ruling that the supplemental instruction cured the Penry I problems was objectively unreasonable.
B. Applicability to Robertson
From the Penry I opinion it is clear that the Court considered Penry’s abused childhood, as well as his mental retardation, to be independently relevant mitigating evidence that the jury should have been instructed that it could consider and give effect to in determining whether to impose the death penalty. In reversing Penry’s *277death sentence, the Court concluded that “his mitigating evidence of mental retardation and childhood abuse has relevance to his moral culpability beyond the scope of the special issues, and that the jury was unable to express its reasoned moral response to that evidence in determining whether death was the appropriate response.” Penry I, 492 U.S. at 322, 109 S.Ct. 2934 (emphasis added); see also id. at 312, 109 S.Ct. 2934 (listing as separate evidence of Penry’s possible reduced personal culpability “his mental retardation, arrested emotional development, and abused background”); id. at 323, 109 S.Ct. 2934 (“[Bjecause of his history of childhood abuse, that same juror [who concluded that Penry acted ‘deliberately,’] could also conclude that Penry was less morally culpable than defendants who have no such excuse!.]”)
As Robertson has presented evidence of childhood abuse here, Penry I, along with Penry II, should be outcome determinative. Under Penry I, the special issues given at Robertson’s trial, which were identical to those given in Penry’s trial, were an inadequate vehicle for allowing a jury to consider Robertson’s child abuse evidence in making a reasoned death penalty determination. And the nullification instruction does not change this result.3 Penry II, 532 U.S. at 803-04, 121 S.Ct. 1910.
The majority responds to this clear textual command from the Court that the special issues are constitutionally infirm where a defendant presents evidence of childhood abuse with two arguments. First, it argues that “[c]hildhood abuse alone is not systematically discussed by Penry /” because there the evidence of child abuse “was inseparable from the Court’s greater concern with Penry’s mental retardation and poor impulse control.” Second, the majority contends that it is neither “logically or empirically true that generic childhood abuse, of whatever duration, type, or severity, bears the same characteristics as mental retardation.”
As to the first argument, the Supreme Court never suggested that either mental retardation or childhood abuse evidence by itself could be constitutionally weighed and acted upon by a jury within the shackles and confines of the special issues. If, as the majority argues, the Court’s “greater concern” was with mental retardation, it is odd the Court did not choose to list just mental retardation or “lost impulse control” as the factor the special issues could not accommodate. Or, if the Court wanted to impart its belief that childhood abuse is evidence that only in conjunction with mental retardation requires a special instruction, it easily could have used the phrase “mental retardation with childhood abuse” or “mental retardation caused by childhood abuse,” rather than “mental retardation and abused childhood,” to describe the problematic evidence. Thus, I, unlike the majority, am willing to credit the Court with saying what it means: that the Texas special issues are not equipped to handle child abuse evidence.
But even assuming the majority’s cramped reading of Penry I is correct, and the Court did not explicitly hold that childhood abuse evidence is not adequately encompassed by the Texas special issues, the Court’s reasoning in Penry I, applied to childhood abuse, mandates the same result. Childhood abuse evidence alone raises the same constitutional problems as the joint mental retardation/ childhood abuse *278evidence in Penry I. The first issue of “deliberateness,” not further defined, does not allow a jury to reflect its conclusion that while a defendant purposely committed a murder, his culpability for that purposeful killing was reduced as a consequence of his abuse as a child. Penry I, 492 U.S. at 322-23, 109 S.Ct. 2934. And the second special issue, future dangerousness, raises an even more troubling scenario of a “two-edged sword,” where child abuse serves as an aggravating, rather than mitigating, factor. Id. at 323, 109 S.Ct. 2934. As Chief Justice Rehnquist noted in a different context in Santosky v. Kramer, 455 U.S. 745, 789, 102 S.Ct. 1388, 71 L.Ed.2d 599 (1982) (Rehnquist, J., dissenting), “[i]t requires no citation of authority to assert that children who are abused in their youth generally face extraordinary problems developing into responsible, productive citizens.”4 Given such common knowledge of the greater likelihood of recidivism among abused children, the Texas special issues leaves a jury with no room to register its conclusion that while a defendant is a future threat to society, his abuse as a child reduces his responsibility for that threat, making use of the death penalty inappropriate.5
Thus, even assuming we are operating without a Supreme Court decision on whether evidence of childhood abuse alone can be considered within the Texas special issues, the presence of the same concerns the Court found with the evidence in Pen-ry I mandates a finding that the special issues are constitutionally infirm here.
II.
The majority fails to reach this result today because it relies on what I believe is an erroneous line of precedent first established in our en banc decision in Graham v. Collins, 950 F.2d 1009, 1029 (5th Cir. 1992) (en banc). Under the Graham test whether a defendant’s mitigating evidence is not adequately encompassed within the Texas special issues, turns on whether that evidence meets four stringent criteria: vol-untariness, permanence, severity, and attribution. The majority describes these requirements for “constitutionally relevant mitigating evidence” “readily apparent from the Court’s opinion in Penry I.” I believe, however, that it is “readily apparent” that this test bears no connection to the reasoning underlying the Penry I decision.
The Supreme Court’s most fundamental holding regarding mitigating evidence at the capital sentencing phase is that “a sentencer may not be precluded from considering, and may not refuse to consider, any relevant mitigating evidence.” Penry I, 492 U.S. at 318, 109 S.Ct. 2934. This evidence includes “any aspects of a defendant’s character or record and any of the *279circumstances of the offense that a defendant proffers as a basis for a sentence less than death.” Id. at 317, 109 S.Ct. 2934 (quoting Lockett v. Ohio, 438 U.S. 586, 604, 98 S.Ct. 2954, 57 L.Ed.2d 973 (1978) (plurality opinion)). Where our en banc decision in Graham and its progeny go wrong is by failing to ask the fundamental Penry I question of how, if at all, the Texas special issues allow a jury to give meaningful consideration to a particular piece of mitigating evidence. For the Graham test to be correct under Penry I, the Texas special issues must allow a jury to adequately weigh a piece of mitigating evidence wherever that evidence does not fall within the Graham criteria. But this is not the case. For example, how do the special issues allow a jury to weigh evidence of a defendant’s childhood abuse where the crime is not directly attributable to that abuse? Under the Texas system a jury could believe that this history of abuse made a defendant more likely to commit future crimes, but, absent a meaningful mitigating evidence instruction, would have no outlet to express its determination that the defendant is not morally culpable enough, as a consequence of his history of abuse, for the death penalty.6
The majority makes two responses to the argument that the Graham line of cases misapplies Penry I. First, the majority notes that Supreme Court decisions in Graham v. Collins, 506 U.S. 461, 113 5.Ct. 892, 122 L.Ed.2d 260 (1993), and Johnson v. Texas, 509 U.S. 350, 113 S.Ct. 2658, 125 L.Ed.2d 290 (1993), established the principle that Penry I is an exception to Jurek v. Texas, 428 U.S. 262, 96 S.Ct. 2950, 49 L.Ed.2d 929 (1976), which found that the Texas special issues system was not unconstitutional on its face, rather than vice versa. How this rather unremarkable statement alters the Penry I analysis escapes me. In both Graham and Johnson the question was whether youth as a mitigating factor was adequately accounted for within the Texas special issue system. Significantly, in determining whether the special issues were sufficiently capacious to encompass evidence of youth, the Court did not pick up the four-pronged Fifth Circuit test the majority advocates here. Rather, it answered the question Penry I mandated it answer: whether the Texas special issues allowed a jury to express its belief that because the capital defendant was young when he committed his crime, he is not morally deserving of the death penalty. Graham, 506 U.S. at 475, 113 S.Ct. 892 (explaining essence of Penry I is whether relevant mitigating evidence is placed beyond the “effective reach of the sentencer”); Johnson, 509 U.S. at 367, 113 S.Ct. 2658 (“The question presented here is whether the Texas special issues allowed adequate consideration of petitioner’s youth.”) And the answer there was yes, because unlike with evidence of childhood abuse, the impact of youth at the time of committing a crime can be reflected in a “no” to future dangerousness, to reflect the lower risks of recidivism with age. Graham, 506 U.S. at 475, *280113 S.Ct. 892 (“[I]t is evident that Graham’s evidence-unlike Penry’s had mitigating relevance to the second special issue concerning his likely future dangerousness.”) Johnson, 509 U.S. at 368, 113 S.Ct. 2658 (“We believe there is ample room in the assessment of future dangerousness for a juror to take account of the difficulties of youth as a mitigating force in the sentencing determination.”)
Perhaps realizing that allusions to Graham and Johnson would not rescue its four-headed hydra, the majority stretches even further in its second defense of its test, ascribing significance to the Supreme Court’s failure to grant certiorari in earlier cases challenging it. It should not bear repeating that denial of certiorari does not shed any light on the views of the Court on the merits of the dispute in which the petition was denied. Holloway v. McElroy, 632 F.2d 605, 636 n. 50 (5th Cir.1980). But more startling is the majority’s cavalier expectation that the Supreme Court repeatedly consider Texas death penalty cases to sort out the applicable legal standards, rather than recognizing our respon-
Thus, the important principle to derive from Graham and Johnson is that in those cases the Court reaffirmed the basic test of Penny I for when the Texas special issues transgress the Eighth Amendment (i.e., when they do not allow a jury to consider a particular piece of mitigating evidence in making the death penalty determination). And this bears little resemblance to our en banc Graham test. sibility to get those cases right using previous Supreme Court decisions. Rather than counting on the Court’s denials of certiorari, the majority would have been well served to attempt to apply the principles of Penny I here, lest the Court decide to again correct an “unreasonable” application of its existing precedent. Penny II, 532 U.S. at 803-04,121 S.Ct. 1910.
III.
Today’s decision is deeply troubling. Petitioner Robertson introduced mitigating evidence of childhood abuse in the penalty phase of his trial, but that evidence could not be considered by the jury in making its moral, reasoned death determination because of the limited special issues inquiry. Yet rather than require Texas to re-sentence Robertson in line with constitutional mínimums, as mandated by Penry I and Penny II, the majority applies a test that bears little relation to Supreme Court precedent to find Robertson’s sentence constitutionally sound. And rather than attempt to apply the Court’s clear principles, the majority resorts to inapposite precedent and counting cert, denials to defend its actions. Because I believe this is wrong, I cannot join the majority.
. There is no dispute that evidence of substance abuse is adequately encompassed within the special issues. Harris v. Cockrell, 313 F.3d 238, 242 (5th Cir.2002).
. The third special issue, which asked "whether the conduct of the defendant in *276killing the deceased was unreasonable in response to provocation, if any, by the deceased,” was not relevant there (or here) because provocation was (and is) not in issue.
. As Judge DeMoss' dissent explains the nullification instruction here and in Penry II were similar, although not identical. But neither the majority, nor Texas asserts that the differences affect our analysis here.
. Or perhaps it does require a citation. The majority suggests that childhood abuse of lesser duration or severity may not have the same effect within the Texas special issues system as the very serious childhood abuse Penry suffered. I take this to mean that the majority contends that in cases of less serious childhood abuse the evidence is not necessarily aggravating because there is a lesser risk of recidivism than among the seriously abused because the effects of less severe childhood abuse are treatable. Motley v. Collins, 18 F.3d 1223, 1235 (5th Cir.1994). As an empirical matter this is not necessarily correct. A study done by the State University of New York (SUNY) — Albany showed that the fact of child maltreatment, rather than its form, was the greatest predictor (and presumably cause) of later delinquency. Office of Juvenile Justice and Delinquency Prevention, U.S. Dep’t of Justice, Juvenile Justice Bulletin, In the Wake of Childhood Maltreatment (Aug. 1997).
. And under the Court’s decision in Penry II, the presence of the nullification instruction does not change this outcome. Penry II, 532 U.S. at 803-04, 121 S.Ct. 1910.
. Of course the majority may respond that one is not less morally culpable for a crime because of a history of child abuse unless the crime can be proven by expert testimony to be attributable to that abuse. While I disagree with this conclusion, the opinion of the court on this matter is, frankly, irrelevant. It is the province of the jury in Texas, as sentencer, to weigh mitigating evidence and draw inferences of culpability from that evidence. Pen-ry /, 492 U.S. at 328, 109 S.Ct. 2934 ("... the jury must be able to consider and give effect to any mitigating evidence relevant to a defendant’s background and character or the circumstances of the crime.”); id. at 318, 109 S.Ct. 2934 (explaining that under Lockett and Eddings the sentencer must be allowed consider all mitigating evidence). Our Graham test substitutes this court's cramped judgment of what is mitigating for that of the jury.