The Supreme Court decided Fry v. Pliler, No. 06-5247, today. The Court was unanimous on the main legal question they had taken the case to resolve. When a state court finds no federal constitutional error, and therefore does not do any harmless error analysis, a federal habeas court which does find error still applies the habeas harmless error rule of Brecht v. Abrahamson, 507 U.S. 619 (1993), not the direct appeal standard of Chapman v. California, 386 U.S. 18 (1967). The Chapman rule is more favorable to the defendant. In announcing the Brecht rule, the Supreme Court gave several reasons, only one of which was that in the case before it the state court had already done a harmless error analysis. The Eighth Circuit had erred in finding that Brecht was limited to that situation, and the other circuits had decided to the contrary.
Four Justices agreed with that conclusion but were troubled by the same thing that bothered me about this case. The error in question was a claimed violation of the rule of Chambers v. Mississippi, 410 U.S. 284, 302 (1973) regarding exclusion of evidence. As subsequently limited by United States v. Scheffer, 523 U. S. 303, 315 (1998), such exclusion is not constitutional error unless it significantly undermined the fundamental elements of the defendant’s defense. The finding of Chambers error under the proper standard would seem to be inconsistent with a holding it was not prejudicial under Brecht. See Stevens opinion at 3; Breyer opinion at 2; CJLF Brief at 5.
Justices Stevens, Souter, and Ginsburg would give the defendant the benefit of the dubious holding that there was Chambers error and reverse the inconsistent finding that it was harmless. Justice Breyer would simply vacate and remand for the Ninth Circuit to resolve its inconsistent holdings. The majority just says that question was not presented.