The use of counterfactuals has long distinguished economic historians from their more normal historian counterparts. There is, however, a problem, which I will tell you about here.
The philosopher Nelson Goodman coined the term “cotenability” in Fact, Fiction, and Forecast (1955). A counterfactual could only be meaningful, he argued, if nothing in it had been shaped by the thing it was designed to study. The relevant conditions of a counterfactual must be “cotenable” with its premise. In other words, you must “avoid resting a counterfactual on any statement that would not be true if the antecedent were true” (4th ed., p. 20).
Two studies on two of my favorite topics—Argentina’s globalization and American slavery—illustrate why this is such a problem. In my opinion, they also suggest there may be some fundamental problems with “quantitative spatial economics.”
