Leerlingen aan het Woord: Verborgen Dimensies van Communicatie op School

An ethnographic document of 'objet trouvé' nature was the starting point for this study: a school diary in two volumes, co-authored by two Dutch teenage girls in the early seventies, mostly during classes-in-progress. This lively collection of fieldnotes (texts, drawings, shrieks, written conversations) constructs an insider-relevant representation of what school is like, from class-to-class, day-to-day, year-to-year. The existence of this collusive document is itself a comment on our assumptions of what goes on in classrooms, what is heard and not heard, seen and not seen. I have selected two passages from "The Writ", as the girls called these unorthodox classroom data, that illustrate how current linguistic and educational paradigms, in embracing a 'conversational' view to learning and communication in school, underanalyze the complexity of multi-party interaction in social situations, both with respect to 'the' data that have to be accounted for and the analytic tools they use to account for them (cf. 'Where is the lesson in all this talk'; Van Dam van Isselt, 1995). The relevance of what is said is constructed in highly complex floors and collusive interactional positions (cf. Goffman, 1979, 'Footing'). In the first part of the paper, a refinement of hearer roles is central. In the second part, a different type of data is used: video-recordings of lessons in 'Dutch as a second language', taught to a group of refugee children recently arrived in the Netherlands. A close examination of these data shows that 'utterance that is not talk' is relevant to contexts of early socialization and language production (cf. Palloni, 1994). The main theme of the paper is that a dynamic notion of context is needed: as produced in real time on a moment-by-moment basis by the interpretative work of co-present parties. Two discourse models that partially fulfill these requirements and that could be adapted for the purpose are discussed: the Linguistic Discourse Model by Polanyi & Scha (Polanyi, 1988) and Van Kuppevelt's proposals for tracing the structure of narrative texts through underlying questions (Van Kuppevelt, 1995).