INTRODUCTION

By about 1830, l'homme éclairé had given way to l'homme moyen. The same awareness of a dynamic and perhaps unstable mass society that super-annuated the reasonable man as the most characteristic object of probability theory simultaneously brought into existence a new one. There was, to be sure, some continuity. The application of probability to error theory, insurance, and gambling problems survived unscathed. More abstractly, what could be subjected to mathematics was yet assumed to have a certain latent rationality, notwithstanding the dispiriting unruliness of manifest events. The allure of aggregate figures for many social thinkers lay precisely in their insensitivity to political and economic crises. The reasonable man might still exist, but he did not and could not control public life. Statistics was valued as a way of searching for the larger order that, it was hoped, would prevail nonetheless. Statisticians exulted in their ability to find such an order, for chance disappeared in large numbers, and with it the discontinuities of revolution – “always the most precarious of all games of chance,” in Gustav Schmoller's words (quoted in Semmel, 1968, p. 197). The main object of probabilistic analysis became mean values. Its most important and conspicuous field of application was now statistics.