ABSTRACT
Historians are right to be uneasy about the way in which social scientists approach historical evidence. On the one hand, when historical truth is not an end in itself, historical data may be processed and selected as evidence to meet the needs of a perilously commanding theory.1 On the other hand, unscholarly bias (otherwise simply called judgement) in an evidentiary quest for support of a favoured explanation is as prevalent among historians as among social scientists. Such caveats matter because this chapter and the next two present in outline analytical histories of three RMAs, and also explain how each RMA worked strategically. These discourses must be rather more constructed than is usual for historical presentation, because the episodes whose stories are narrated are each defined by the contested concept of RMA. To illustrate: historians are by no means agreed on the relationship between the causes, effects, and feedback loops that explain the course of the First World War. How much more difficult is it to secure a convincing grip on the historical narrative of the RMA of the First World War? The same point applies to each case study.
