ABSTRACT
Arguments have been made in earlier chapters (e.g., chapter two) as to why a parametric approach to analyzing social skills, though almost universally adopted in that field, is in fact mistaken and should be replaced by a structural approach. It would seem obvious that linguistics is the natural heir to this particular throne, and it seems strange that social skills investigators have not turned to this long-established discipline. Linguistics, and in particular that branch of development termed discourse analysis, seems an obvious choice in that not only is it structural, thus befitting the nature of the subject, but speech has a much greater significance in social skills compared with other elements of behaviour (Marzillier & Lambert, 1976; Trower, 1980). Accordingly, in attempting to tackle the problem of abnormal social behaviour, it would seem we should incorporate assessment of and training in spoken interactions, using an applied linguistics approach. However, here, as in other areas of applied linguistics, progress has been hampered by the lack of an adequate linguistics. The authors of the most substantial piece of work on abnormal discourse so far reported (Rochester & Martin, 1979), admit that “the arguments for the study of dialogic patterns are very persuasive” and they “would like to study conversational behaviour in schizophrenic speaker-listeners”; nevertheless they in fact confined themselves to monologue on the grounds that the model they were using was reliable only for monologue.
