This paper discusses the relationship between the concept of reference and that of inference; the point is to indicate that contrary to the usual view it may be good to see the former as "parasitic" on the latter, not the other way around. The paper is divided into two parts.