ABSTRACT
The study of linguistic elements as they actually are used, in contexts larger than sentences, has already proved its value in providing an understanding of many synchronic phenomena that were intractable to more traditional approaches (among many recent examples of the newer approach, see e.g. the various articles in Giv6n 1979b and in Klein-Andreu 1983). I would like to take this opportunity to apply some of the concepts and techniques of discourse-pragmatic analysis toward elucidating a historical problem: namely, the semantic development of the Spanish verb-form ending in -ra (as in (que) yo amara, now translatable as "(that) I would love").
