this article examines an inconsistency in the skinnerian proposal with respect to the relations between explanation and description. at some moments, skinner identifies explanation and description, committing itself to the machian principle " to explain is to describe" . at other moments, skinner detaches the terms, considering description as a preliminary stage of the scientific enterprise, which must be complemented by explanation. it is argued, here, in favor of the identity between explanation and description in the radical behaviorism on the basis of mach influences in the skinnerian philosophy of science. moreover, it is stated that these influences take place through selectionism-pragmatism, and not empiricism-descriptivism. this analysis differs from traditional interpretations of the relations between mach and skinner. it is concluded that the identification between explanation and description seems to express better the philosophical affinities of radical behaviorism with pragmatism.