Exactness and Pseudoexactness in Historical Linguistics

Abstract

[Introduction] In May of 1995, the Society for Exact Philosophy held its Silver Jubilee meeting in Calgary, Alberta. The invited speaker was Mario Bunge, one of the founding members of the Society. The title of his talk was “Pseudoexactness in the Social Sciences.” By “pseudoexactness” Bunge meant the assignment of numerical quantities to essentially qualitative variables, such as the level of dictatorship or happiness in a society, and expressing theories about the relationships between them in terms of mathematical equations. It was a terrific word, and a terrific paper. The following day, I presented my own paper, “Causal Decision Theory and Decision-theoretic Causation” (since published as Hitchcock, 1996), and much to my chagrin, found myself on the receiving end of the charge of pseudoexactness: hadn’t I assumed that degrees of belief have precise numerical strengths, and that events have precise numerical probabilities? “If you can’t beat’em, join’em” they say, so here is my own rebuke of pseudoexactness.