In contrast to their interest in interdisciplinary cooperation, quantitative historians have neglected the international dimension of their work. Many of the pioneers of quantitative methods in the United States (like L. Stone, C. Tilly, or W. O. Aydelotte) specialize in European history, but the bulk of quantitative work deals with American topics; or, if it ventures abroad, investigates other countries in their own national terms (Kousser, 1980). Where there is a broader awareness, scholars tend to draw upon British and, due to the acclaim of the Annales school, also on French examples; while Germany, and even more so Russia, Latin America, or the Far East, remain in the outer darkness of quantitative consciousness (Rowney and Graham, 1969, as can be contrasted with Aydelotte, 1972). Some fields like economic history (due to its strong international organization) are internationalized; but others like the “new political history” are somewhat provincial, considering that few countries developed comparable democratic institutions in the nineteenth century, thereby providing material for similar research (Temin, 1981; Bogue, 1980). In less cosmopolitan scholarly communities than the American, the situation is worse, if anything. French quantitative historians seem to have shown some interest in their British or Hispanic neighbors, but have done little international or comparative work (Bourdelais, this issue; for an exception see Graham, 1982). Similarly, German historical social scientists appear to borrow methods freely, but do not occupy themselves much with transnational concerns (Jarausch, 1978).

In his presidential address to the Social Science History Association Convention in November 1981, Robert William Fogel declared sanguinely that social scientific historians had won their battle for legitimacy within the historical profession in America, and that we should now stop feeling embattled, spend less effort proselytizing, and calmly go on with our substantive work. While his statistics on the occupational advancement of social scientific historians do indicate a degree of acceptance, and while his advice to worry less and pay attention to business will be followed (as that is what nearly all of us were doing anyway), I am less optimistic than Fogel, read the employment trends differently, and see more signs of a reaction against quantitative social scientific history—or what I like to refer to as QUASSH—than he does (Kousser, 1980). Perhaps Professor Fogel and I differ only temperamentally. As a former Marxist, he still retains a bit of faith in the inevitable triumph of progressive forces; as a former Methodist, I am unable to shake off the pessimism that is the psychological residue of the doctrine of original sin. In any case, whereas Fogel seems to think that most recent criticisms of QUASSH are so obviously flawed as to require no answer, I fear that some people, especially those with substantial investments in “history-as-it-used-to-be-done,” may still be susceptible to false messiahs or, perhaps more precisely, false Jeremiahs.

It was fashionable some years ago for historians to speak of the quantitative revolution in their subject and to look forward, either with hope or with foreboding, to the day when history, like geography, sociology, and other social sciences before it, would be dominated by practitioners skilled in quantitative methods. New questions would be asked, new methods used, new sources exploited, and new discoveries made. In the process, those historians without training in quantitative methods would be swept from the battlegrounds of the subject. They would be forced to retreat either into the antiquarianism within which many quantitative historians, privately, located them or they would be forced to undergo a painful process of retooling in a struggle to avoid technological obsolescence. It was a reasonable expectation in those days that a statistical training would become as essential to a historian as a training in foreign languages or experience in burrowing for sources in the record office: As Emmanuel Le Roy Ladurie put it in 1968, by the 1980s “the historian will be a programmer or he will be nothing” (Le Roy Ladurie, 1968). As a result, quantitative historians gathering in the early 1970s thought that the quantitative revolution would be largely complete as the world entered the 1980s.

How do theory-oriented history and quantitative history relate to each other? According to Morgan Kousser, theory and methodology—meaning statistical and mathematical methodology for quantitative work—“are like the Emperor and his new clothes; without the one, you have an empty suit; without the other, a nude old man” (Kousser, 1977). I disagree. There is, on the one hand, quantitative historical work without much theory. On the other hand, there exists good theoretical history of a basically nonquantitative character. Whereas quantitative history can only gain by becoming more theoretical, theory orientation does not always lead to quantification.

Due to the central position it assigns to the longue durée, the Annales school—impetus for and crossroads of historical debate—used quantitative methods from the beginning and granted them their patents of nobility. For some decades now, it has been inconceivable for good historical research in France not to devote an important part of its efforts to the measuring of phenomena. True, the Annales school has never spoken with a single voice; the changes that have occurred in its membership and in neighboring disciplines, as well as new historical sensibilities, have modified its original characteristics. The study directed by A Burguière of the Ecole des Hautes Etudes en Sciences Sociales will soon report on the content and evolution of the themes, the network of contributors, and the intellectual climate that shaped the first generation of the Annales. Therefore this article limits itself to a rapid survey of publications, followed by an examination of the distinctive traits of the quantitative dimension of French historical work, the directions for and reasons behind new research. The picture must remain somewhat general because it focuses on only three major points: the preeminence of serial history over quantitative history; the underrepresentation of international comparative works, despite the example of F. Braudel; and the development of individual-level data banks, an important source for social and serial history.

In the late 1950s and early 1960s, historians in various parts of the Soviet Union began to explore the use of mathematical methods and computer technology in historical studies. This development occurred not only in Moscow, Leningrad, Novosibirsk, and Kiev, but also in Tallinn. A small group of historians at the Institute of History of the Estonian Academy of Sciences, in close collaboration with their colleagues in Moscow, made an effort to apply mathematical methods to the study of history. Thanks to the comradely support of the research centers of the Soviet Union, it was possible for the historians of this small republic to progress rapidly. As a result, their work has become known not only in the Soviet Union but abroad as well.

This work does not attempt to review the historiography produced by Latin American historians who, with different degrees of success, have applied quantitative methods to the study of the past. It is rather a reflection on, as well as an assessment of, the different schools of thought and the competing theories and methods of quantification as they are usually debated by Latin American historians.