Gabriel Tarde is thought to have “lost” his debates with Durkheim by insisting that sociology ought to occupy itself with observable interpersonal processes. Given contemporary interest in such processes—much abetted by the computer—Tarde's reputation is being rehabilitated. Terry Clark (1969) was first to notice that Tarde (1898) had anticipated Lazarzfeld's two-step flow of communication. Tarde's work has bearing on social networks, interpersonal influence, diffusion of innovation and the aggregation of public opinion.

During the oral exam on my doctoral thesis—later to become Part 1 of Personal Influence—Robert Merton asked me to name the scholar who debated Durkheim on the nature of sociology. It was the one question to which I had no answer. This failure is all the more embarrassing now, fifty years later, inasmuch as intellectual historians such as Terry Clark (1969) and Serge Moscovici (1985) pay homage to the French social psychologist, Gabriel Tarde, for having anticipated the “two-step flow of communication” and other propositions in the classic Columbia voting studies by Lazarsfeld et al. (1944) and Berelson et al. (1954) and in Personal Influence (Katz and Lazarsfeld 1955). At least in some measure, it accounts for my many years of ardent advocacy of Tarde's all but forgotten work on opinion and communication. But penance aside, the rediscovery of this forefather—not just by me—has amply justified the effort. His renewed presence can enliven almost every aspect of current work on political communication, on diffusion of innovation, on social network theory, on public opinion, on collective behavior and on the deliberative democracy of the “public sphere.”

Gabriel Tarde achieved renown in turn-of-the-century France. To his professional training in the law, he added criminology, statistics and social psychology and moved from the provincial courtroom of his aristocratic forebears to the Collège de France. “Tarde held virtually every leader position open to a French social scientist out-side the university system,” says Terry Clark (1969).