Labeling and Analyzing Historical Phenomena: Some Preliminary Challenges

A serious obstacle to the search for a more scientific history is that humans label themselves and their actions. These labels can be extremely sticky and often obscure the categories which might be most useful for seeking regularities. Another, related, problem is a focus on dramatic events that seem to be relatively rare and are commonly recognized as landmarks, e.g. political and industrial revolutions. Having formed several of these major events into a class, scientifically-minded historians have then often searched for a very small set of discrete variables that could predict the occurrence or non-occurrence of these very special events. By contrast, I would argue that we are likely to be better off by looking at more general processes that may include but are not limited to these dramatic events, and looking for clusters of variables which interact with each other; the hoped-for result would usually be not to explain the categorical presence or absence of some process (e.g., “economic development”) but to group many cases into families, seeking to explain both within-group and between-group variation by means of systematic comparison