Williams v. Taylor
The Supreme Court ruled that a Virginia death-row inmate's trial lawyers were constitutionally ineffective for failing to investigate and present extensive evidence of his abusive childhood and mental impairments at sentencing, and that the Virginia Supreme Court was wrong to deny him relief.
The decision also set the ground rules for a brand-new federal habeas law, explaining how much deference federal courts must give state courts' legal decisions -- rejecting a lower-court test that would have made federal review almost meaningless.
“It is past question that the rule set forth in Strickland qualifies as "clearly established Federal law, as determined by the Supreme Court of the United States."”
The Court's conclusion that the Strickland ineffective-assistance test is clearly established federal law.
How it got here: A state habeas judge recommended a new sentencing hearing; the Virginia Supreme Court reversed; a federal district court granted habeas relief; the Fourth Circuit reversed that grant, and Williams asked the Supreme Court to step in.
The Case in Depth
What happened
Terry Williams was convicted of capital murder in Virginia and sentenced to death. At sentencing, his lawyers presented only brief, weak testimony describing him as a "nice boy." They never investigated or introduced records showing Williams had suffered severe childhood abuse, neglect, and possible mental impairment, evidence a state habeas court later found should have been presented to the jury.
The question before the Court
Did a Virginia death-row inmate's constitutional right to effective counsel get violated when his lawyers barely investigated or presented mitigating evidence at sentencing?
The Court's answer
Yes -- the Court ruled that Terry Williams' lawyers were constitutionally ineffective because they failed to investigate and present extensive evidence of his abusive childhood, neglect, and possible mental impairment at his death-penalty sentencing, and that this failure created a reasonable probability the jury would have spared his life. The Virginia Supreme Court's decision denying him relief was both legally wrong and unreasonable under the federal habeas law's new standard.
Beyond Williams' individual case, the Court also settled how much deference federal judges must give state courts when reviewing constitutional claims under the newly amended habeas statute. It rejected an overly deferential lower-court test and held that federal courts must independently decide whether a state court's decision was legally wrong, and may grant relief when that error also counts as unreasonable.
Curious how the Court got there? See the step-by-step legal reasoning →
Why it matters
The ruling gave federal judges reviewing state prisoners' constitutional claims a workable standard under the 1996 habeas reform law, shaping how thousands of future habeas cases -- especially death penalty appeals -- would be decided. It also reinforced that defense lawyers in capital cases have a real duty to dig up and present mitigating evidence about a defendant's background before a jury decides on execution.
What changes now
The case returns to the lower federal courts for further proceedings consistent with the Court's ruling, which effectively means Williams' habeas relief should be reinstated and Virginia may need to hold a new sentencing hearing. The decision is a final merits ruling, not a temporary order, and its interpretation of the federal habeas deference standard became the framework federal courts use nationwide when reviewing state prisoners' constitutional claims.
What this does not decide
The Court did not decide whether Williams' lawyers were also ineffective during the guilt phase of his trial -- that was not disputed. It also left open, without deciding, exactly how to classify cases involving unreasonable extensions of legal principles to new factual contexts under the habeas statute.
Concurrences and dissents
Concurrence in part — Justice O'Connor
Justice O'Connor agreed Williams should get relief but rejected Justice Stevens' reading of the habeas statute. She argued the 'contrary to' and 'unreasonable application' clauses must be given separate, independent meanings: the first covers state decisions that reach an opposite legal conclusion or misapply precedent to materially identical facts, while the second covers objectively unreasonable applications of the correct legal rule to new facts.
Dissent in part — Justice Rehnquist
“It was not, therefore, unreasonable for the Virginia Supreme Court to decide that a jury would not have been swayed by evidence demonstrating that petitioner had a terrible childhood and a low IQ.”Rehnquist's argument that the state court's decision to deny relief was reasonable given overwhelming evidence of future danger.
Chief Justice Rehnquist agreed with O'Connor's interpretation of the habeas statute but disagreed that Williams deserved relief. He argued the Virginia Supreme Court did not actually rely on Lockhart and reasonably concluded, given overwhelming evidence of Williams' future dangerousness, that the jury would not have been swayed by evidence of his troubled childhood and low IQ.
How the Court got there
The legal reasoning, step by step
- The Court first had to define how much deference the 1996 federal habeas law owes to state courts. It rejected the Fourth Circuit's rule that relief could issue only if all reasonable judges would agree the state court was unreasonable, calling that an unworkable and overly deferential reading of the statute.
- The Court explained that 'clearly established Federal law' means the rule dictated by Supreme Court precedent at the time the state conviction became final -- essentially the same idea already used in the Court's retroactivity doctrine from Teague v. Lane, a 1989 case about when new legal rules apply to old convictions.
- The Court then addressed the statute's two triggers for relief: a state decision that is 'contrary to' clearly established law (meaning a federal court disagrees on a pure legal question) or one that 'unreasonably applies' that law to the facts, concluding federal courts must independently decide if a state ruling is wrong, and grant relief when that wrongness is also unreasonable.
- Applying that framework, the Court identified Strickland v. Washington's two-part test -- deficient lawyering plus a reasonable probability the outcome would have differed -- as the clearly established rule governing Williams' ineffective-assistance claim.
- The Court found the Virginia Supreme Court's decision was 'contrary to' Strickland because it wrongly treated a separate case, Lockhart v. Fretwell, as adding a further fundamental-fairness requirement on top of Strickland's ordinary prejudice test, when Lockhart only applies in unusual circumstances not present here.
- The Court also found the Virginia Supreme Court unreasonably applied Strickland by never weighing the totality of the mitigating evidence -- the abuse, neglect, and possible mental impairment -- against the case for a death sentence, concluding there was a reasonable probability the outcome would have been different had that evidence been presented.
Doctrinal impact
Cases affected by this decision
Reaffirms Strickland v. Washington (466 U.S. 668)
The Court applies Strickland's two-part deficient-performance-plus-prejudice test as the governing standard.
Limits Lockhart v. Fretwell (506 U.S. 364)
The Court clarifies Lockhart's fairness exception applies only in unusual cases, not as a general modification of Strickland.
Reaffirms Teague v. Lane (489 U.S. 288)
The Court relies on Teague's framework to define what counts as clearly established federal law.