The power of this theory is that it can begin to explain failure as well as success. Students who have not been sufficiently immersed in a context, who have not been sufficiently exposed to its speech genres, will not be able to ventriloquate, to respond dialogically. Of course, sufficiency of exposure will vary from person to person: Threshold levels may differ accordingboth to individual needs as well as to the degree to which the social languages or contexts to which learners were exposed earlier resemble new ones. This raises the question of "discourse transfer," in Popken's (1992) terms. lowe to my colleagues Yalden and Jones the analogy of fish scales to suggest successive stages of approximation, or overlap, in terms of social languages, genres, and learning contexts. However, as Popken (1992) illustrated, arguing on the basis of the literature on interlanguage in second language transfer, both positive and negative transfer can occur, and how students learn to select appropriately remains to be investigated. More specific and concrete implications for the teacher of writing will become apparent in the discussion of research findings that follow.