Deadly Confusion: Juror Instructions in Capital Cases

Abstract

A fatal mistake. A defendant is sentenced to die because the jury was misinformed about the law. The justice system should be designed to prevent such a tragic error. Yet our interviews with jurors who served in South Carolina capital cases indicate that this nightmare is a reality. Although our data are limited to South Carolina, the question whether jurors are adequately instructed in capital cases is of national concern. For example, the issue whether jurors should be more fully informed about the alternative to a death sentence has arisen in other states. And the question whether jurors understand the burdens of proof in capital cases can arise in any death penalty state. As with many death penalty issues, it is tempting to view the question of juror instructions solely as a question for resolution by the Supreme Court as a matter of federal constitutional law. This narrow perspective may be reinforced by the Supreme Court's grant of certiorari in State v. Simmons to decide whether a jury should be informed when a life sentence means life without the possibility of parole. However important Supreme Court death penalty decisions are, the initial responsibility for instructing jurors rests with trial judges. Our data and analysis should inform trial judges about the real impact of the instructions they choose to give and not to give. Even if the Constitution does not mandate full and clear instructions, trial judges and reviewing courts should provide them in the sound exercise of their discretion. After describing the data and the law in Part I, Part II shows that jurors' false expectations about alternatives to the death sentence probably influence their sentencing decisions. Part III establishes that jurors do not understand the burdens of proof governing the sentencing phase of murder trials. Part IV shows that confusion works against the defendant because the jurors' strong initial inclination is to sentence to death.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 121 citations exist upstream.