Énonciation, corpus et representativité : le cas de « along »

• Main Author: Graham Ranger
• Format: Article
• Language: English
• Published: Association Française de Linguistique Cognitive 2019-07-01
• Series: CogniTextes
• Subjects: enunciation; Theory of Enunciative and Predicative Operations; corpus linguistics; preposition; particle; British National Corpus
• Online Access: http://journals.openedition.org/cognitextes/1517

The representativeness of a corpus is a function of the relationship between the corpus and a target language or language variety the corpus is intended to sample. However, the second term of this relationship can only be apprehended via a finite set of language occurrences, i.e. a corpus. For this reason we prefer to consider representativeness with respect to one’s specific research goals. In the enunciative approach to language, markers are described in terms of an invariant schematic form, configured by the pressure of surrounding operations into contextually-situated shapes. The elaboration of a schematic form is carried out on the evidence of authentic examples studied in context and submitted to manipulations and judgements of acceptability on the basis of the linguist’s intuition, essentially. This paper argues that corpus methodologies can provide invaluable quantitative support for these intuitions and, from there, for the theoretical constructions of enunciative approaches to language. We argue the case for such an approach with a study of the marker “along,“ using data from the British National Corpus. Collocational affinities of “along“ will allow us progressively to identify four particularly significant contextual configurations, yielding spatial, temporal, subjective and argumentative values for the marker.