USA v. LEMUREL E. WILLIAMS, No. 15-1194
Lemurel Williams was convicted of being a felon in possession of a gun. We agree with Williams’s that a new trial is needed because the totality of the circumstances regarding the jury’s verdict was impermissibly coercive.
After three hours of deliberating, the jury returned a guilty verdict, which was read aloud in court. At defense counsel’s request, the jury was polled—that is, jurors were individually asked, “Was this and is this your verdict with regard to the defendant, Lemurel E. Williams?” Juror 1 responded “no.”
Ten minutes later, the jury sent the judge a note. The top of the note read, “We apologize, we misunderstood the question that was presented to each juror.” The bottom of the note read, “We have the verdict.” Williams argues that he was denied a fair trial because Juror 1 was coerced into joining the guilty verdict.
Any criminal defendant . . . being tried by a jury is entitled to the uncoerced verdict of that body.” Lowenfield v. Phelps, 484 U.S. 231, 241 (1988). Coercion occurs when jurors “surrender their honest opinions for the mere purpose of returning a verdict.” United States v. Blitch, 622 F.3d 658, 668 (7th Cir. 2010). To evaluate potential coercion we look at the totality of the circumstances.Lowenfield, 484 U.S. at 250; Blitch, 622 F.3d at 669-70. We focus on the situation facing the juror, not the intent of the party or the judge whose actions created that situation. Blitch, 622 F.3d at 668.
We cannot know for certain whether Juror 1 favored conviction all along and merely misunderstood the poll question, in which case she was not coerced. She was not asked to explain how she misunderstood the question. We have no quarrel with that, given the sometimes unclear line between asking jurors whether the verdict form contains a mistake and improperly asking about the deliberative processes.See generally Fed. R. Evid. 606(b).
We begin with the polls. Long ago, the Supreme Court held that it is always reversible error to ask a divided jury to reveal its numerical division (e.g., ten jurors on one side and two on the other). Brasfield v. United States, 272 U.S. 448 (1926). But polling a jury that is believed to be unanimous is different—a poll rarely reveals division; its purpose and usual effect is to confirm unanimity. See, e.g., Lyell v. Renico, 470 F.3d 1177, 1184 (6th Cir. 2006) (emphasizing the difference between a post-verdict poll and an inquiry made to a jury known to be divided).
Once Juror 1 rejected the verdict, the verdict could not stand; polling the remaining jurors was pointless. See, e.g., Lyell, 470 F.3d at 1183 . We are confident that the judge did not intend to pressure anyone (it appears he did not hear Juror 1’s response), but the judge’s intentions are not at issue. Blitch, 622 F.3d at 668. We must look at the totality of the circumstances, including the fact that the instruction came at a very sensitive moment— immediately after Juror 1 was identified as the lone dissenter. SeeBlitch, 622 F.3d at 670
We reiterate that we are confident the judge did not intend to coerce anyone. But the focus is on the juror and we hold that the totality of the circumstances—the combination of the polls, the fact of a lone dissenter, the instruction, the content and timing of the jury’s note, and the form of the judge’s direct question—was impermissibly coercive, even if none of those circumstances standing alone would have been. So the judgment cannot stand.
We REVERSE the judgment of the district court and REMAND for a new trial.