A Test for Richter’s Reach

As I discussed in this post, in its decision last year in Harrington v. Richter, the Supreme Court gave more teeth to the command of 28 U.S.C. § 2254(d) that federal habeas courts must defer to state-court adjudications on the merits.  More specifically, the Court held that deference must be shown even to an unexplained, summary ruling by a state court.

Earlier this month, the Court granted cert. in a new case that may (let us hope) impose a limitation on Richter.  In Cavazos v. Smith (No. 11-465), the Court will consider whether § 2254(d) is owed to a state-court decision that did not even acknowledge the petitioner’s federal constitutional claim.

Here’s what happened.

Williams was tried for murder.  During the course of deliberations, the jury foreman complained to the trial judge that “Juror No. 6” was holding the state to an unduly high burden of proof.  The trial judge conducted an investigation and ultimately discharged Juror No. 6 for bias against the government.  With an alternate in place, the jury then convicted Williams.

On appeal, Williams argued that the discharge of Juror No. 6 violated both a state statute and the Sixth Amendment.  Her conviction was affirmed, with the appellate court addressing the statutory argument, but not the constitutional argument.

Williams once again asserted the Sixth Amendment claim in a federal habeas petition.  She lost in the district court, but the Ninth Circuit reversed, holding that § 2254(d) deference was not warranted and thus applying a de novo standard of review.

The relevant language of § 2254(d) speaks of a “claim that was adjudicated on the merits in State court proceedings.”  It is beyond me how a claim that was not even mentioned by the state court could be said to have been “adjudicated” by the court.  It would be nice to think that the justices would have to view Cavazos v. Williams as an easy affirm, at least on the § 2254(d) question.  But the very fact that they decided to review the Ninth Circuit’s decision is not comforting.  And Richter itself demonstrates that several of the justices favor an aggressive reading of § 2254(d) that is not really required by the statutory text.

The cert. grant seems limited to the § 2254(d) question, but the underlying Sixth Amendment question is a fascinating and difficult one.  Williams is remarkably similar to a Second Circuit case I teach in Criminal Procedure, United States v. Thomas, 116 F.3d 606 (2d Cir. 1997).  I read the Ninth Circuit in Williams as essentially adopting the Second Circuit’s approach, which is highly discouraging of discharging holdout jurors in the middle of deliberations, even when a holdout arguably seems bent on nullifying the law.

The trial judge in Williams attempted to distinguish between a discharge based on a nullifying intent (as to which the record on Juror No. 6 was equivocal) and a discharge based on bias against the government.  I’m not sure the distinction really holds up.  The question should not be whether a juror is more sympathetic to one side or another — who isn’t? — but whether the juror will put sympathies aside and follow the law.  Without a clear, well-supported finding that a holdout intends to nullify the law, the holdout should not be discharged.