Secondary Historical Sources: Problems and Issues

In the 1980s and 1990s, sociologists called for guidelines on the use of secondary historical sources in the social sciences. Over the next few years, several comparative-historical social scientists offered scattered advice on using secondary sources appropriately, but since 2000 or so, few have devoted attention to the problem. Some argue that current practices in the field are acceptable, and nobody has offered a coherent and thorough set of guidelines for the use of secondary sources. In contrast, historians themselves are generally skeptical of the secondary sources that their own field produces.

Two issues arise from the various bodies of advice written by historians and social scientists. First, secondary historical sources seem to contain at least five different kinds of facts: reported facts, empirical inferences, summarized facts, empirical generalizations, and arguments-as-facts. Second, social scientists seem not to have grappled with the problem that historians produce idiographic interpretations, in which single facts are deeply woven into webs of other facts, while social scientists very often seek to extract such facts to use as “data” for nomothetic purposes.

In the first stage of a project of developing comprehensive guidelines and standards for using secondary historical sources in social science, I will analyze how comparative-historical scholars in sociology and political science use secondary sources in their published works. I will sample several recent works in this field and will analyze them to learn how many of which kinds of facts are prevalent, how social scientists use them as data, and how they make theoretical inferences from these bodies of facts. I will report on the progress of this research.