Brecht v. Abrahamson
The Supreme Court ruled that federal courts reviewing state convictions on habeas corpus should use a more forgiving harmless-error test than the one state courts use on direct appeal, when the error is the ordinary kind that occurs during trial.
Applying that more forgiving test to a Wisconsin murder case, the Court found that the prosecutor's improper questions about the defendant's post-arrest silence did not affect the jury's guilty verdict, so the conviction stands.
“had substantial and injurious effect or influence in determining the jury's verdict.”
The new, more lenient standard the Court adopted for habeas review of trial errors.
How it got here: Wisconsin's Supreme Court reinstated Brecht's conviction; a federal district court granted habeas relief; the Seventh Circuit reversed, and the Supreme Court took the case to resolve a circuit split.
The Case in Depth
What happened
Todd Brecht shot and killed his brother-in-law, a local district attorney who had taken him in after prison and disapproved of his drinking and sexuality. At trial Brecht claimed the shooting was an accident. The prosecutor pointed out, including after Brecht received Miranda warnings, that he had never told anyone about the shooting being an accident until he testified. Wisconsin courts found this violated Doyle v. Ohio but called it harmless.
The question before the Court
If a state prosecutor wrongly uses a defendant's silence after his arrest to attack his testimony, how forgiving can a federal habeas court be about that mistake?
The Court's answer
Federal habeas courts should use the more lenient Kotteakos test, not the strict Chapman test. Under Kotteakos, a state prisoner must show that a trial-type constitutional error actually had a "substantial and injurious effect or influence" on the jury's verdict, rather than requiring the state to prove beyond a reasonable doubt that the error was harmless. The Court reasoned this fits habeas's narrower purpose of correcting only grievous wrongs, respects state courts' primary role, and avoids disturbing final convictions absent real impact.
Applying that standard to Brecht's case, the Court found the improper references to his post-arrest silence were brief, largely cumulative of properly admitted evidence of his earlier silence, and outweighed by strong physical evidence contradicting his story. So the error did not substantially influence the verdict, and Brecht was not entitled to habeas relief.
Curious how the Court got there? See the step-by-step legal reasoning →
Why it matters
The decision makes it considerably harder for state prisoners to win federal habeas relief based on ordinary trial mistakes, even constitutional ones, because they must now show the error actually swayed the jury rather than merely show it wasn't proven harmless beyond a reasonable doubt. This affects thousands of habeas petitions filed by state inmates every year.
What changes now
This is a final merits decision resolving Brecht's habeas petition against him; his conviction stands. Going forward, federal courts hearing state prisoners' habeas claims involving ordinary trial errors must apply the more lenient Kotteakos 'substantial and injurious effect' standard instead of the stricter Chapman standard used on direct appeal, reshaping harmless-error review across federal habeas litigation.
What this does not decide
The Court left open that in an unusual case involving a deliberate, especially egregious trial error, or one combined with a pattern of prosecutorial misconduct, habeas relief might still be warranted even without substantial influence on the verdict. It did not address structural errors requiring automatic reversal.
Concurrences and dissents
Concurrence — Justice Stevens
Justice Stevens agreed with adopting the Kotteakos standard but wrote separately to stress that it remains demanding: the burden stays on the prosecution, courts must review the whole record de novo, and judges cannot simply ask whether they think the defendant would have been convicted anyway. He emphasized that the difference between Kotteakos and Chapman matters less than the quality of judgment applied.
Dissent — Justice White
“Because neither the federal habeas corpus statute nor our own precedents can support such illogically disparate treatment, I dissent.”White's objection that the ruling treats identical constitutional violations differently depending on the stage of review.
Justice White argued that because Chapman established a federal constitutional floor, any conviction upheld despite non-harmless constitutional error is unconstitutional custody entitling the prisoner to habeas relief regardless of forum. He objected that the majority created an illogical system where the same violation gets treated differently depending on whether direct review happened to catch it, and would have remanded for application of Chapman.
Dissent — Justice Blackmun
Justice Blackmun joined Justice White's dissent in full, agreeing that the majority's decision could not be reconciled with the reasoning of Stone v. Powell.
Dissent — Justice O'Connor
Justice O'Connor argued the Court should have exercised more restraint before lowering the harmless-error standard for a rule so closely tied to the accuracy of verdicts, distinguishing this case from prior decisions limiting habeas review of purely prophylactic rules. She warned the new standard would not reduce judicial burdens and risked preserving convictions tainted by errors that actually affected reliability.
Dissent in part — Justice Souter
Justice Souter joined most of Justice White's dissent but declined to join the footnote and Part III, and clarified he was not expressing a view on the merits of Stone v. Powell.
How the Court got there
The legal reasoning, step by step
- The Court first classified the constitutional problem as 'trial error' -- a mistake that happens during the presentation of evidence and can be measured against the rest of the trial record -- rather than a 'structural defect' that taints the whole proceeding and requires automatic reversal.
- Because Doyle error is trial error, some harmless-error test applies; the question was which one. On direct appeal, states must use the demanding Chapman test, requiring proof beyond a reasonable doubt that the error didn't affect the verdict.
- The Court reasoned that habeas review serves different purposes than direct appeal -- protecting finality of convictions, respecting state courts' primary role in enforcing criminal law, and reserving relief for those 'grievously wronged' -- so a less searching standard is appropriate on collateral review.
- The Court adopted the Kotteakos test from a 1946 nonconstitutional case, under which the petitioner must show the error had a 'substantial and injurious effect or influence' on the jury's verdict, rather than requiring the state to disprove harm beyond a reasonable doubt.
- Applying Kotteakos itself rather than remanding, the Court weighed the brief, infrequent references to Brecht's post-arrest silence against extensive proper evidence, including his lawful pre-arrest silence and strong physical evidence contradicting his accident story, and concluded the error did not substantially sway the verdict.
Doctrinal impact
Cases affected by this decision
Limits Chapman v. California (386 U.S. 18)
Confines the strict 'harmless beyond a reasonable doubt' standard to direct review, no longer applying it to trial errors raised in federal habeas.
Reaffirms Kotteakos v. United States (328 U.S. 750)
Extends this nonconstitutional harmless-error test to constitutional trial errors raised on federal habeas review.
Reaffirms Arizona v. Fulminante (499 U.S. 279)
Relies on its trial-error versus structural-defect framework to classify the Doyle violation as trial error.