Miller-El v. Cockrell

A COA is a jurisdictional prerequisite to an appeal from the denial of federal habeas relief. Under 28 U.S.C. § 2253(c)(2), the applicant must make a substantial showing of the denial of a constitutional right. That standard is met when reasonable jurists could disagree with the district court's resolution of the claim or could conclude that the issue deserves encouragement to proceed further.

The COA decision requires an overview of the claim and a general assessment of its merits, including the district court's application of AEDPA. But it does not permit a court of appeals to decide the appeal's ultimate merits first and then deny a COA because the petitioner would lose. Doing so effectively decides an appeal without jurisdiction.

The COA standard is more demanding than merely showing good faith or a nonfrivolous claim, and issuance is not automatic. But the prisoner need not show that the appeal will succeed. A claim may be sufficiently debatable to warrant a COA even if, after full briefing and review, every reasonable jurist ultimately agrees that habeas relief should be denied.

Section 2254(d)(2) permits habeas relief when a state-court decision rested on an unreasonable factual determination, while § 2254(e)(1) presumes state factual findings correct unless rebutted by clear and convincing evidence. The Fifth Circuit erred by treating these provisions as requiring Miller-El to prove an objectively unreasonable decision by clear and convincing evidence.

The clear-and-convincing standard in § 2254(e)(1) concerns particular state-court factual determinations. The objective-unreasonableness requirement in § 2254(d)(2) concerns the conditions for granting habeas relief. They are independent provisions, and neither authorizes a court to turn the threshold COA inquiry into a final merits adjudication.

AEDPA deference remains relevant because the question is whether reasonable jurists could debate the district court's AEDPA-based rejection of the constitutional claim. At the COA stage, however, the court need not definitively decide whether the state court's factual finding was rebutted by clear and convincing evidence or whether its decision was objectively unreasonable.

Batson's first step was no longer disputed: Miller-El had raised an inference that the prosecution used peremptory strikes on racial grounds. The State had also satisfied Batson's second step by offering facially race-neutral explanations. The relevant question was therefore Batson's third step—whether the explanations were credible or instead were pretexts for purposeful discrimination.

The strike statistics alone supported debate. Prosecutors removed 10 of 11 eligible Black prospective jurors, or 91 percent, while using peremptory strikes against only 13 percent of eligible non-Black jurors. Ten of the State's fourteen peremptory strikes were used against Black venire members, a disparity unlikely to result from chance.

The State's stated reasons for striking Black jurors appeared to apply as well to some white jurors who were allowed to serve. Prosecutors cited hesitation about capital punishment, doubts about executing a potentially rehabilitable defendant, and relatives' criminal histories. Yet white seated jurors expressed similar concerns about the death penalty and rehabilitation or had relatives with criminal records. This comparative evidence could support an inference that the supposedly neutral criteria were selectively applied.

The prosecutors' voir dire methods also displayed substantial racial disparities. Black prospective jurors were much more likely than white jurors to receive a graphic description of execution before being asked about the death penalty. And white jurors were overwhelmingly more likely to be told that the statutory minimum murder sentence was five years before being asked what minimum sentence they could impose. The record supported an inference that prosecutors designed their questions to elicit answers from Black venire members that could later justify exclusion.

The prosecution's use of jury shuffles added to the evidence. Prosecutors sought shuffles when many Black prospective jurors were near the front of the panel, and they objected to a defense shuffle only after the result moved Black jurors forward. Although jury shuffling was not itself a Batson violation, it bore on the credibility of the State's claim that race played no role in jury selection.

Finally, the record contained evidence that the Dallas County District Attorney's Office had historically trained and encouraged prosecutors to exclude minority jurors, including a written policy expressly directing prosecutors not to accept Black or other minority jurors. Evidence suggested that the discriminatory culture persisted into the period relevant to Miller-El's prosecution. While this history did not by itself establish a Batson violation, it was relevant circumstantial evidence casting doubt on the asserted neutrality of the strikes in this case.

The state court's Batson hearing occurred more than two years after voir dire, when the court had not originally been required under Swain to assess the credibility of the prosecutors' particular explanations. Its failure to address important evidence, including the jury shuffles and the office's history, did not eliminate that evidence's significance. The Court did not decide the ultimate Batson claim, but held that the record plainly made the district court's rejection of it debatable.