Language Endangerment and Language Documentation in Africa

Abstract

Language endangerment and language documentation in Africa 1 How global is language endangerment? 1.1 The global and the local Language endangerment and death is seen as a process that operates worldwide. The metaphors used to describe the situations of language change and shift 1 captured under this umbrella term cast them as a human tragedy of the largest imaginable scale. Yet, when shifting our gaze to Africa, stories of resilience and adaptivity, of mobility, multilingualism and creativity, flank stories of disappearance and extinction. In this introduction I gauge how useful global perspectives on language endangerment and loss (henceforth LEL; see Mufwene forthcoming) are; whether they are really global; and what kind of mould they provide for describing locally anchored practices. A critique of universal scales and models and a quest for different perspectives on communicative practices on the African continent and for new ways to describe and document them is made in the remainder of the introduction, but runs through the chapter, focussing on ontologies, concepts and models of linguistic description and documentation (LDD) and LEL in 2; on sites of linguistic vitality in 3; and on situations of linguistic exclusion and marginalization in 4. Section 5 briefly explores how new phenomenological approaches may not only serve to capture Africa's linguistic diversity more fully, but how they may contribute to a decolonization of African linguistics and to the conceptualization of new forms of language development.

Language as category: using prototype theory to create reference points for the study of multilingual data

“…It is the hypothesis of this paper that languages can be conceptualized as categories, structured according to principles laid out by Rosch and her colleagues in their work on prototype theory (Rosch, 1973, 1975a; Rosch & Mervis, 1975; Rosch et al, 1976). This approach can account for many of the ostensible conflicts observed in multilingual settings and related research paradigms as discussed above (see also Cobbinah et al, 2016, and Lüpke, 2017b, for earlier formulations of the hypothesis put forward in this paper). This is not the first time that aspects of prototype theory have been applied to language.…”

“…The notion of the ANCESTRAL CODE, 3 pervasive in the language endangerment rhetoric driving many documentation enterprises, means that the choice of which data to collect is often skewed to represent the target language, or a ‘pure’ version of it, rather than language use as it occurs (Childs, Good, & Mitchell, 2014, p. 171; Dobrin & Berson, 2011, p. 193). It is recognized that “approaches privileging one ‘language’ as ancestral are problematic, and potentially even pernicious, in highly multilingual and fluid linguistic contexts where language use is organized around multilingual repertoires rather than ‘native languages’” (Childs et al, 2014, p. 169; Lüpke, 2017b). Nevertheless, Goodchild (2016, p. 76) describes how, despite explicit calls for the development of new research paradigms for multilingualism, “language documentation has, until recently, continued in the same isolationist vein”.…”

“…These languages are thus typically the focus of ancestral code type documentation and description projects. However, focusing solely on ideological constructs of patrimonial language erases large portions of the multilingual populations of these communities (Lüpke, 2016a), as well as inherently positing some ‘pure’ variety which may well not exist (Dobrin & Berson, 2011). These languages, such as they are, will be referred to here as PATRIMONIAL LANGUAGE OF X, where X is a geographical locus (village or polity).…”

“…Languages can be used to index affiliation with a certain group, and thus difference from other groups (Irvine & Gal, 2000). Both Lüpke (2016a), for the Casamance, and Di Carlo (2016), for the Lower Fungom, Cameroon, have stressed that this differentiation has practical importance. Group identity, sanctioned and strengthened by linguistic identities, is important for strategic purposes, whether for spiritual security (Di Carlo, 2016, p. 73), or for claiming land rights and forging strategic alliances against invasion and slave raids (Lüpke, 2016a, p. 10).…”

“…This manifests in the maintenance of many named, but unstandardized, languages, with varying degrees of relatedness to each other, and which are spoken for different purposes by each person, and where each interaction involves a unique constellation of languages and sociolinguistic factors. It is observed that while named languages can, to varying degrees, be distinguished according to linguistic factors, the prevailing approach to linguistic description can obscure much of the nature of language use, as it is not able to effectively account for fluidity and variation (Lüpke, 2017b). I propose that languages can better be modelled using principles from prototype theory, which facilitate more accurate and flexible descriptions based on both the elicited judgements and observed usage of speakers, and both allow and account for variation.…”