Deconfounding the Effects of Competition and Attrition on Dialect across the Lifespan: A Panel Study Investigation of Swabian
R. Harald Baayen, Karen V. Beaman, and Michael Ramscar
Rising mobility, increasing levels of education, and intensifying immigration are bringing more diverse people into more frequent contact, more prolonged interactions (Auer 2007; Britain 2013, 2016; Britain and Trudgill 1999; Dodsworth 2017; Trudgill 1992). These factors, coupled with continuing globalisation and ubiquitous social media, are pushing standard languages into the forefront of people’s experience and relegating nonstandard varieties to the background. As a consequence, a growing body of research suggests that many dialects, i.e., nonstandard language varieties, are receding across the globe (Britain 2009; Schilling- Estes and Wolfram 1999; Smith and Durham 2012), and nowhere is this more evident than in Europe, notably in Germany (Auer 2005, 2018; Auer et al. 2011; Auer and Spiekermann 2011; Kehrein 2012; Pedersen 2005; Schmidt 2011; Streck and Auer 2012).
Dialectologists generally focus on changes in lexical items and the use of dialect-specific words (e.g., Swabian Grombiere versus standard German Kartoffel ‘potato’), variationists tend to target changes in the frequencies of various phonological, grammatical, and discourse-pragmatic forms (e.g., Alemannic Fescht [fr/t] versus standard German Fest [fast] ‘party’), and corpus/computational linguists often look at the competition between grammatical forms and changes in frequencies between different word forms (e.g., colloquial geb versus standard German gebe ‘go’). Cumulatively, these metrics reveal that when different language varieties come into contact, accommodation occurs, and most commonly, it is the dialect variants that lose out and the more prestigious, standard variants that “win” (Britain and Trudgill 1999; Giles et al. 1973; Trudgill 1986; Wieling et al. 2011). Indeed, this pattern is also seen in individuals: as they age, adult speakers appear to lose dialect as they develop greater experience with the standard language, gained through their participation in various educational, commercial, and public institutions (Britain 2010; Eckert 1997; Labov 1964; Sankoff and Laberge 1978).
The idea of dialect attrition is the dominant way of interpreting these patterns of language development. This interpretation assumes that standard languages encroach on dialects, such that, at the lexical level dialect words are replaced by their standard language counterparts, resulting in the attrition of individual dialect vocabularies. There are, however, problems inherent in this assumption that are particularly relevant to lifespan studies of dialect usage. First, the lexical distributional properties of natural languages (Baayen 2001) ensure that the lexical knowledge of healthy individuals increases continuously across their lifespan. These same distributional properties also guarantee that the majority of lexical types any individual knows are relatively rare and that many of these types will be shared only with subsets of the wider community. As people age, their knowledge expands as they gain new experiences (e.g., in schools, on the job, at leisure), face various new life events (e.g., graduation, marriage, childbirth), and tackle new challenges (e.g., driving a fork-lift, climbing Kilimanjaro). In the course of these undertakings, speakers encounter new words and add them to their vocabularies. Many of these new words are specific to particular areas of knowledge, such as medicine, plumbing, or linguistics, and are not in the vocabularies of other speakers in the community. In an increasingly technology-driven world, this increased lexical knowledge may involve words for new inventions and technologies (e.g., cell phone, fax, emoji). Importantly, it is likely that many of the specialisation-specific words, as well as words for cultural innovations, have the same form in both the dialect and the standard language.
The second challenge in investigating dialect attrition across the lifespan relates to the differing social settings in which the use of standard language or dialect is appropriate. A local dialect is lexically strong for discussing traditional methods of farming and socially appropriate for informal interactions with family and friends in the local community. The standard language comes into its own for interactions with speakers from different backgrounds or to cover topics for which the dialect does not offer the relevant specialised words. These two considerations thus suggest an alternative account of the changes in speech patterns as individuals age: specifically, lexical change across the lifespan does not necessarily represent skill loss, as has often been claimed (e.g., Корке and Schmid 2004), but rather reflects the fact that experience tends to make individuals more skilled when measured in terms of their ability to communicate about an expanding repertoire of topics. Thus, many changes in speech patterns merely reflect the increased influence of later acquired, standard language lexical knowledge, and not necessarily a substantial loss of the dialect itself.
Furthermore, as we will argue, earlier acquired dialect forms are likely to be more deeply embedded in speakers’ repertoires than later acquired standard forms and, therefore, are more likely to be reactivated as individuals age. Our view of lexical change across the lifespan can be visualised as a diamond. In their youth, speakers’ experiences are naturally quite limited, and the breadth of their active vocabulary usage reflects this narrower range of experiences, which are primarily with the local community in the native dialect (the upper point of the diamond). As individuals move into the workforce, their vocabularies expand along with their experiences, to express a widening range of interests and activities (the wider midpoint of the diamond). Then, in later life, particularly in retirement, the range of activities and the breadth of social contacts slowly decrease, resulting in talk about a smaller subset of topics (the lower point of the diamond). Our Diamond Model of vocabulary development over the lifetime (see Figure 10.1) proposes that dialects are typically the primary medium for communicating in the early and the later stages of life (the upper and lower parts of the diamond), with standard languages playing a greater role during mid-life when individual experiences and specialisations are most varied (the widest part of the diamond, cf. linguistic market (Sankoff and Laberge 1978)). As a result, if there is active vocabulary loss1 across the lifespan as individuals advance in age, narrow their circle of contacts, and reduce their exposure to standard language settings, we expect it to be primarily visible in the standard language, following the principle of “use it or lose it” (Shors et al. 2012).
Figure 10.1 Diamond Model of vocabulary development over the lifespan.