Study participants were 170 jury-eligible citizens. In each of a series of 2-hour sessions, a group of 12 to 16 individuals listened to an audiotaped description of the evidence presented in both the guilt and sentencing phases of the trial of James Free. The jurors then heard an audiotape of the instructions on the law that were to be applied to the facts of the case. One group of six death-qualified jurors from each session was randomly selected to deliberate, and the remaining jurors individually answered a set of written questions designed to elicit their understanding of what these instructions told them about how jurors should apply the law. In an effort to correct the sources of misunderstanding in the pattern instructions, the same kinds of linguistic problems that are likely to interfere with comprehension of any jury instructions were addressed: confusing or incoherent discourse organization, needlessly complex syntax, semantic obstacles, and text that invites incorrect inferences due to misleading wording or omitted information. In addition, however, the study clarified the three major legal points that seemed to confuse respondents in a previous study: unenumerated mitigators, differences of opinion on mitigators, and weighing. The study concludes that a fundamental step in improving the ability of legal instructions to convey legal standards not only accurately but effectively is to rewrite the instruction by applying linguistic principles that identify which features of language facilitate comprehension of a text and that interfere with it. For jury instructions to communicate effectively, they need not only legal accuracy and clear exposition, but also to address the way juror meanings may differ from legal meanings. 51 footnotes