In Jurek v. Texas, 428 U.S. 262 (1976), the Court announced that Texas was one of the lucky ones that had guessed correctly. Their system, in which the penalty was determined by answers to specific questions about deliberateness and future dangerousness, was constitutional because it narrowed the class of murderers eligible for the death penalty and allowed consideration of mitigating circumstances. Conspicuously absent from the opinion was any requirement that the jury be instructed to consider all mitigating circumstances the defendant might offer. The special issues quite obviously did not do that, nor did the finite list of mitigating circumstances in the Florida system approved the same day in Proffitt v. Florida, 428 U.S. 242 (1976).

Two years later, the Court issued its disastrous fiat in Lockett v. Ohio, 438 U.S. 586 (1978). The mandate was now expanded to consider any mitigating factor the defendant might proffer. The Court neither overruled Jurek nor explained how its new decision fit with what it said before. Eleven years went by, with the Texas courts continuing to use the system the Supreme Court had expressly approved in a decision it had not overruled. Then in Penry v. Lynaugh, 492 U.S. 302 (1989), the Court said the system was not constitutional, at least as applied to Penry and his claimed (but hotly disputed) mitigating circumstance of retardation. Yet the Court still did not overrule Jurek, and three years later the Court reaffirmed Jurek and disclaimed any interpretation of Penry under which "the Texas special issues system in almost every case would have to be supplemented by a further instruction." Johnson v. Texas, 509 U.S. 350, 372 (1993). It would have seemed from Johnson that Jurek was still the rule and Penry was the exception. The Fifth Circuit and the Texas Court of Criminal Appeals struggled to reconcile this tangle of decisions by limiting Penry claims to narrow categories of evidence.

The Texas Legislature, which meets only every other year, amended the statute in 1991 to add the supplemental issue that a broad reading of Penry (the one rejected in Johnson) would seem to require. However, there were substantial numbers of cases tried before Penry in which only the original special issues instruction was given or in the interval between Penry and the amendment in which judges came up with various fixes to accommodate Penry while still conforming as nearly as possible to Texas law.

The Present Cases

Smith v. Texas is a case in which the trial judge tried to fix the problem. The defendant objected on the ground that the judge had no power under Texas law to make the fix, but he did not object to the supplemental instruction itself as being insufficient to fix the Penry problem on federal constitutional grounds, despite an express invitation from the judge to suggest changes to the instruction. The Texas Court of Criminal Appeals first held that the fix was good; there was no federal constitutional error. This decision was summarily reversed and remanded by the Supreme Court. On remand, the Texas court applied its equivalent of the plain error rule for objections not raised at trial, cf. United States v. Olano, 507 U.S. 725 (1993), and held that the prejudice to the defendant, if any, did not rise to the level required for reversal under that rule. Unfortunately, the opinion is not well written and not does not cleanly separate what is harmless error from what is not error in the court's view.

This procedure brought howls of indignation from two amici supporting Smith, a group of former federal judges represented by Erwin Chemerinsky and the Constitution Project. They claim the Texas court is evading the Supreme Court's mandate and call up the ghosts of Southern evasion of civil rights decisions in the bad old days. CJLF's brief supports the propriety of applying a plain-error-type rule on remand in this situation. We note, among other things, that the Supreme Court directed the federal courts of appeals to do exactly that in United States v. Booker, 543 U. S. 220, 268 (2005).

Abdul-Kabir v. Quarterman and Brewer v. Quarterman deal more directly with how mitigating evidence can be considered under the old system and whether the state court decision was a reasonable application of Supreme Court precedent at the time. The briefs are available at Supreme Court Times. Much may depend in this case on how willing the high court is to admit that its jurisprudence in this area has been confusing and inconsistent. They did admit that in Lockyer v. Andrade, 538 U.S. 63 (2003), and the result was a win for the state. Congress has limited habeas relief to cases where the state court decision "was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States...." When the Supreme Court caselaw is a tangled mess, very little is clearly established.