concurring:
I concur in the disposition of this case, but write separately to express my disagreement with the majority’s handling of two issues.
I.
In deciding whether the trial court’s exclusion of proffered mitigating evidence was harmless, the majority cites Brecht v. Abrahamson, 507 U.S. 619, 113 S.Ct. 1710, 123 L.Ed.2d 353 (1993), a pre-AEDPA case holding that a federal habeas court need only determine whether such exclusion had a substantial and injurious effect or influence on determining the jury’s verdict. In my view, the Brecht standard of review is inapplicable to post-AEDPA cases such as this one. Instead, our proper function is to determine whether the Oklahoma Court of Criminal Appeals’ disposition of this issue “was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States.” 28 U.S.C. § 2254(d)(1). More specifically, we must determine whether the Court of Criminal Appeals reasonably applied the harmless error standard outlined in Chapman v. California, 386 U.S. 18, 24, 87 S.Ct. 824, 17 L.Ed.2d 705 (1967) (requiring government to prove a constitutional error was harmless beyond a reasonable doubt).
Curiously, the majority suggests “[t]he Brecht standard in this setting is more rigorous” than the new AEDPA standards of review. I question if we have a basis for this conclusion. Until the meaning of the new AEDPA standards of review are sufficiently fleshed out by the Supreme Court, we have no basis for concluding whether one or the other is “more rigorous.” In any event, I believe the majority’s practice is dangerous because it effectively invites federal habeas courts to “pick and choose” from pre and post-AEDPA standards of review.
Applying the standard of review set forth in § 2254(d)(1), I conclude the Court *1211of Criminal Appeals’ resolution of this issue was neither “contrary to” nor “an unreasonable application of’ Chapman. More specifically, I believe the Court of Criminal Appeals properly concluded “there [wa]s no reasonable probability that the error might have contributed to the imposition of [Bryson’s] death sentence.” Bryson v. State, 876 P.2d 240, 257 (Okla.Crim.App.1994).
II.
In addressing Bryson’s claim that he was entitled to jury instructions on the lesser-included offenses of second degree murder and first degree manslaughter, the majority correctly notes “[t]he state trial and appellate courts determined that the evidence did not support giving these instructions.” The majority goes on, however, to characterize the state courts’ decisions as “factual determinations” entitled to a “presumption of correctness” under 28 U.S.C. § 2254(e)(1). Although it is not outcome determinative in the case before us, the majority’s characterization of these decisions is, in my view, clearly flawed.
Federal habeas courts are required, under both pre and post-AEDPA law, to afford a presumption of correctness to any “determination of a factual issue made by a State court.” 28 U.S.C. § 2254(e)(1) (outlining post-AEDPA standards); see Thompson v. Keohane, 516 U.S. 99, 108-09, 116 S.Ct. 457, 133 L.Ed.2d 383 (1995) (discussing pre-AEDPA standards); Case v. Mondragon, 887 F.2d 1388, 1392-93 (10th Cir.1989) (same). For this purpose, “factual issues” are defined as “basic, primary, or historical facts: facts ‘in the sense of a recital of external events and the credibility of their narrators.’ ” Townsend v. Sain, 372 U.S. 293, 309 n. 6, 83 S.Ct. 745, 9 L.Ed.2d 770 (1963) (quoting Brown v. Allen, 344 U.S. 443, 506, 73 S.Ct. 397, 97 L.Ed. 469 (1953)). Typically, when resolving factual issues, the factfinder must either choose between or among two or more conflicting versoions of the facts or, where a fact is uncontested, make a finding based upon the unconverted evidence. As courts of review, applelate courts review fact findings by reviewing the record to determine whether the fact findings are supported by the record. This appellate function does not involye fact finding in the first instance, but rather a review of the record to determine whether the factfinder had an evidentiary basis for its rulings which would satisfy the legal standard in question. Notably, “mixed questions of fact and law, which require the application of a legal standard to the historical-fact determinations, are not facts in this sense” and are not entitled to the presumption. Id. at 309 n. 6, 83 S.Ct. 745; Case, 887 F.2d at 1393 (presumption of correctness does not attach “to legal conclusions or determinations on mixed questions of law and fact”). Thus, our characterization of a state court determination as one of fact or law has critical implications, for it controls the standard of review we apply.
The question here is how to characterize the state courts’ determinations that Bry-son was not entitled to any lesser included offense instructions. This task is easy because the Oklahoma Court of Criminal Appeals has already characterized such determinations as issues of law. Hooks v. State, 862 P.2d 1273, 1280 (Okla.Crim.App.1993) (“it is the duty of the trial court to determine as a matter of law whether the evidence presented at trial is sufficient to justify the submission to the jury of instructions on lesser included offenses”), cert. denied, 511 U.S. 1100, 114 S.Ct. 1870, 128 L.Ed.2d 490 (1994). I note this characterization is consistent with the one we have implicitly adopted in direct criminal appeals. See, e.g., United States v. Abeyta, 27 F.3d 470, 473 (10th Cir.1994) (reviewing as a mixed question of law and fact the trial court’s decision not to give a lesser included offense instruction). Further, I believe this characterization is entirely reasonable. In deciding whether the evidence presented at trial was sufficient to warrant the giving of a particular instruction, a court is not deciding basic, primary, or historical facts, nor is its pre*1212dominant function to make credibility findings.
Precisely how the state courts’ determination of this legal issue in Bryson’s case can now be transformed into a “factual determination” for purposes of federal ha-beas review is not explained by the majority. Instead, the majority cites four habeas cases from this circuit: Hooks v. Ward, 184 F.3d 1206 (10th Cir.1999); Boyd v. Ward, 179 F.3d 904 (10th Cir.1999); Newsted v. Gibson, 158 F.3d 1085 (10th Cir.1998), cert. denied, - U.S. -, 119 S.Ct. 1509, 143 L.Ed.2d 661 (1999), and Lujan v. Tansy, 2 F.3d 1031 (10th Cir.1993), cert. denied, 510 U.S. 1120, 114 S.Ct. 1074, 127 L.Ed.2d 392 (1994).
Addressing these cases in reverse order, I turn first to Lujan, a pre-AEDPA case brought by a New Mexico state prisoner convicted of first degree murder. The petitioner argued the trial court erred in refusing to give diminished capacity instructions (which would have allowed the jury to determine whether the petitioner “had the ability to form the deliberate intention to take away the life of another”). The Lujan panel began its analysis of this claim by noting the New Mexico Supreme Court had’ rejected the claim on the grounds there was “ ‘evidence in the record that the [petitioner] was able to form a deliberate intention, with no evidence to the contrary.’ ” 2 F.3d at 1035 (quoting State v. Lujan, 94 N.M. 232, 608 P.2d 1114, 1115 (1980)). With no explanation or citation, the Lujan panel characterized the New Mexico Supreme Court’s conclusion as a “factual finding” entitled to a presumption of correctness under 28 U.S.C. § 2254(d) (the pre-AEDPA provision that afforded deference to state court factual findings). Id. at 1035. In my view, Lujan is not controlling here. Aside from the fact that the Lujan panel’s characterization appears to ignore New Mexico state law, see, e.g., State v. Vallejos, 122 N.M. 318, 924 P.2d 727, 734 (1996) (determination of whether there is some evidence to establish defense and require jury instructions is a question of law), it is not controlling here, where we are characterizing what the Oklahoma state courts did when they applied Oklahoma state law.
In Newsted, another pre-AEDPA case, an Oklahoma state prisoner convicted of first degree murder sought federal habeas relief on the ground that his trial counsel was ineffective for failing to request a heat ■of passion manslaughter instruction. In analyzing this claim, the Newsted panel began by noting the Oklahoma Court of Criminal Appeals had summarily rejected the claim. The Newsted panel then stated, without explanation or citation, that the Court of Criminal Appeals’ decision “necessarily encompassed factual findings about the evidence in th[e] case.” 158 F.3d at 1091. It is not clear what the Newsted panel meant by this latter language. To the extent it was only meant to indicate the Court of Criminal Appeals made findings of historical fact about what evidence was properly admitted at the petitioner’s trial, I can agree with that proposition. However, to the extent it was intended to suggest the Court of Criminal Appeals made factual findings based upon the evidence properly admitted at trial, I cannot agree. Because of the vague language employed, as well as the lack of supporting citation, I believe it is appropriate to narrowly interpret the above-quoted language in Newsted.
Having distinguished Newsted and Lu-jan, I turn to Boyd and Hooks, two post-AEDPA habeas cases from Oklahoma. In each case, the petitioner argued he was deprived of his right to lesser included offense instructions (in Boyd, the challenge was asserted via an ineffective assistance of trial counsel “gateway”). With no citation or analysis, the panels in both cases appear to have concluded the state courts’ decisions that the evidence presented at trial was not sufficient to warrant the giving of lesser included instructions were “factual determinations” entitled to a presumption of correctness under § 2254(e)(1). Hooks, 184 F.3d at 1230-31; Boyd, 179 F.3d 904, at 917; but see White v. Scott, 141 F.3d 1187, 1998 WL 165162 *1213(10th Cir.1998) (table) (concluding “the [state] trial judge’s determination not to give a self-defense instruction” was a “mixed question of law and fact”).
Assuming, arguendo, the panels in Hooks and Boyd intended to reach these conclusions, I acknowledge we are bound by them. Fortunately, resolution of this issue does not affect the outcome of Bry-son’s ease since he is not entitled to habeas relief regardless of how the Oklahoma courts’ decision on the lesser included offense issue is characterized. However, since this issue has potential ramifications for future habeas cases, it is an issue that should be addressed at some point by the en banc court.