Edited by Thierry Ruchot and Pascale Van Praet
[Lingvisticæ Investigationes Supplementa 33] 2016
► pp. 265–282
Starting out from the consideration that conclusions as to what is atypical regarding a linguistic feature have to be drawn on the basis of language norms, the paper discusses three types of norms: those defined in authoritative grammars or dictionaries, those apparent through judgements by native speakers, and those determined by corpus-based analyses of language use in large corpora. As grammars and dictionaries are generally based on – formal – written language productions (“folk belief” being strongly dependant on school grammar), the norms defined do not cover the phenomena of verbal constructions occurring in oral language production, especially in dialogical contexts. The study of naturally occurring conversations consequently reveals a great number of syntactic, semantic and conversational features a priori deviating from the “official” norm. The paper demonstrates that rather than being atypical these forms are specific for oral language production. What is more, atypicality cannot be defined on the basis of one specific type of speech exchange system, each different genre having its own typical properties. Atypicality is consequently what does not coincide with the forms of oral or written conceptuality applied by the majority of a community of language users in a given situation.