Résumé Migge :
Much of the work in language documentation has focused on describing language as a closed system and on documenting structural linguistic phenomena that are of interest from a typological and theoretical interest. In fact, a leading figure in the area of language documentation has argued that the main goals of language documentation are the discovery of new facts about human language and the testing of theoretical claims (Everett 2001). The expected outcomes of such an endeavour are a comprehensive reference grammar, a corpus of texts with interlinear glosses and annotated for grammatical and social interest and a “reasonably full vocabulary” (Dixon 2007: 19). Most of the data for this type of research come from formal elicitation sessions and the recording of cultural narratives with a selected few, relatively conservative, members of the community. This research tradition has indeed produced valuable insights into linguistic phenomena, but it has also led to a shewed perspective of the linguistic landscape of many regions in the world. Essentially, it is widely assumed that in most regions of the world languages are like self-contained islands that co-exist in the same location but do not interact with each other. However, even cursory contact with any of these regions will make it very clear that this is a myth that cannot be upheld because most languages exist in multilingual contexts and their speakers have multilingual rather than monolingual linguistic repertoires.
In this presentation I argue that research on any language should take a more holistic approach from the get go in order to both gain more comprehensive insights into a language and in order to also make the products of this research more applicable to the speaker communities themselves. Starting from Dell Hymes’ assumption that language competence does not simply entail grammatical but also communicative competence(s), I argue that research on any language should embrace variation and that it should focus on the linguistic practices and repertoires of the entire community rather than that of a select subset thereof. Such research will not only provide us with deeper insights into the structural makeup of a language, but it will also tell us what people do with language and why, thus helping us gain insights about issues around language maintenance and shift and the culture of a people, a third pillar of documentation research (Everett).
However, a holistic approach to language documentation will crucially require a change in research methods, both in terms of data collection and data analysis approaches. In this presentation I will mostly focus on the former. Based on my own and collaborative research (e.g. Migge & Léglise 2013) on the Maroon languages in Suriname and French Guiana, I show that a combination of traditional ethnographic or linguistic anthropological, urban dialectology (Labov), sociolinguistic survey, participatory and linguistic descriptive data collection methods provides overlapping and mutually enriching insights into the ordered heterogeneity of language practices that characterize most if not all communities.