I shall propose in this article 1 a new theory of stare decisis—a term I use loosely to mean the practice of courts in deciding new cases in accordance with precedents—that draws upon the insights of communications theory as well as upon some previous work of my own on the decision-making process in tort law. The attempt to apply communications theory to the law is not new or—given that judicial decision-making is a species of verbal behaviour—unexpected. Previous efforts to apply communications theory to problems of judicial decision-making have foundered, however, on a lack of clear conception as to what that theory means and can tell us about the judicial process, and it is with an attempt at clarification of the relevant concepts that I begin.
Communications theory is not a unified body of thought. It has three quite distinct branches. The first, 'syntactics', is concerned with the logical arrangement, transmission, and receipt of signals or signs. It is the domain of the electrical engineer; its concern is with the transmission of signals, whatever their meaning. The second is 'semantics', which is concerned with the meaning of the signals to people. The third is 'pragmatics', which is the study of the impact of signal transmission on human behaviour (Cherry 1957). This tripartite division is not wholly satisfactory; we shall return to that point.
The key concepts of syntactics, for our purposes, are 'information', 'redundancy', and 'feedback', of which the first two are best discussed together. For the telegraphic engineer, information is the content of a signal that could not have been predicted by the receiver; it is a probability concept. The more probable the transmission of a given sign, the less information its