In adjudication, facts matter. Cases are often disputes over conflicting versions of the facts. In both civil and criminal cases, certain important, so-called ultimate facts are specified as determinative by the applicable substantive law. When there is uncertainty about these facts, and trials become necessary to resolve the dispute, burdens of proof structure the tribunal's factual assessments. In American civil cases, for example, the ultimate facts that define a cause of action or defense usually must be shown by the plaintiff to be true “by a preponderance of the evidence,” and in criminal cases, the ultimate facts must be shown by the prosecution to be true “beyond reasonable doubt.” The epistemic components of these requirements reflect the fact that they do not involve a surrender to some kind of pure proceduralism, in which the quest for accuracy is ignored in favor of whatever results from fair procedures. Instead, they reflect the necessity of judgment under uncertainty and the need to exercise that judgment in a way that makes the best use possible of our unavoidably fallible assessments of the facts. This, at any rate, is the premise on which the following account will build.

But what exactly does it mean to prove a civil case “by a preponderance of the evidence”? Or to prove a criminal case “beyond reasonable doubt”? Much appellate ink has been spilled, and many issues settled, on how to formulate these standards verbally, on which standard applies in which kinds of cases, and on the applicability of yet other intermediate standards to some classes of cases. Nevertheless, fundamental questions about what these standards mean remain deeply controversial. In this book I address a set of issues critically important to answering these questions. Specifically, I explore the relationships among three ideas that infuse modern scholarship regarding the burdens of proof. Clarification of these ideas and their relationships promises significant advances in our understanding of the proof process.

Before stating these ideas, a few preliminary comments are in order. The present discourse concerns proof of “adjudicative” facts, facts concerning the conduct of parties to litigation that trigger the applicability of substantive legal rules – such as the fact that the defendant's conduct caused the injury to the plaintiff or the fact that the accused had the intent to kill the deceased.