Statistical sources in France before World War I

Abstract

Today, quantitative data are doubtless quite rightly occupying an increasingly large position in economic history. We are no longer in the period of vague descriptions and collections of isolated facts that did not lead to any valid explanation. The influence of economists involved historians in the handling of figures, essential measures for those who wish to understand structures and detect movements. It is nonetheless true that there are serious differences between the approach of economists and that of historians. Economists apply reasoning to practically only the present time or to a relatively short period. Their models and patterns are difficult to apply to periods when the structures were markedly different. They also use regular statistical series covering a considerable number of facts, and above all series that may not be perfect (perfection is illusory here) but provide serious guarantees. Historians are less privileged. They dissect economic systems that are very different to our own and whose structures have not yet been closely studied. They possess only sparse statistics whose reliability seems extremely doubtful. Economists are not usually faced with the problems of source and critique that are the daily lot of historians. This article is aimed less at providing complete results than stimulating certain research on detail. Our statistical knowledge and our knowledge of statistical data are still too fragmentary and imperfect for it to be possible to envisagean absolutely definitive overall study. The reader will find here only the components of a general problematic. The main reasons for this attitude are explained below.