If one simply cannot measure the stated variables or the parameters with which the theory is constructed, or if their measurement is so laden with error that no discrimination between alternative hypotheses is possible, the theory becomes a vacuous exercise in formal logic that has on points of contact with the contingent world. The theory explains nothing because it explains everything. It is my contention that a good deal of the structure of evolutionary genetics comes perilously close to being of this sort.
Richard Lewontin, The Genetic Basis of Evolutionary Change (Columbia UP, 1974), p.12.
