Linguistic ‘mudes’ and the de-ethnicization of language choice in Catalonia

Abstract

Catalan speakers have traditionally constructed the Catalan language as the main emblem of their identity even as migration filled the country with substantial numbers of speakers of Castilian. Although Catalan speakers have been bilingual in Catalan and Castilian for generations, sociolinguistic research has shown how speakers' bilingual practices have always been sensitive to keeping a clear sense of the boundaries between the languages and between their communities of speakers. The norms of language choice in everyday life have reflected this as Catalans have tended to use Catalan basically between those considered to ‘be’ Catalan. This article shows that this situation is gradually changing due to new conditions of mobility and access to language, that is, because most native speakers of Castilian are now bilingual and speak Catalan often in everyday life. On the basis of a corpus of 25 interviews and 15 group discussions conducted in Catalonia with a sample of young people of different profiles, we show that young people in Catalonia increasingly rely on prima facie linguistic behavior rather than ethnolinguistic classification to decide which language to speak in specific contexts, so that language use loses its earlier function of ethnolinguistic boundary maintenance.

Keywords:

• Catalan
• ethnicity
• multilingualism
• language choice
• second language acquisition

Notes

1. Accents and evidence of ‘linguistic interference’ are not good indicators of native language when those who have Catalan as a family language grow up in the Barcelona Metropolitan Area and adopt many features of those who learn it at school. In a variationist study conducted some time ago, Argenter, Pujolar, and Vilardell (1998) found that ‘weak pronouns’ en, ho, and hi, a traditional feature of spoken Catalan, were very little used in the Barcelona Metropolitan Area and normally used outside of it irrespective of the first language of the speakers. This indicated that L1 speakers of both languages tended to display the same patterns of linguistic interference of Castilian into Catalan.

2. Gonzàlez et al.'s (2009) data are based on the Enquesta d'usos lingüístics (2003), the ‘2003 linguistic usages survey.’ It is a wide-ranging sociolinguistic survey conducted by the Generalitat de Catalunya through telephone interviews to a representative sample of 7257 residents of the Principality of Catalonia aged 15 and above.

3. Figure 1 does not reflect respondents’ direct answers in the survey; but an elaboration of their claims as to the extent to which they used the two languages in different contexts, such as with parents, siblings, at school, at work, with friends, and so on. Thus, respondents classified as using ‘only or mostly Castilian’ were those estimated to be using Catalan < 20% of their time either in the contexts characterized as ‘primary socialization’ or as (current) ‘habitual language.’ On the other hand, people's ‘own’ language reflects direct responses to the question ‘what is your language?’ For more details, see Gonzàlez et al. (2009).

4. These data could be presented in many ways because people may experience more than one ‘muda’ during their lifetime and in different directions. In this case, we have included all significant mudes, which is why percentages add up to more than 100. It is noticeable that Catalan speakers have many fewer mudes and they also adopt much more restricted uses of Castilian than the reverse. The figures obscure the fact that adopting Catalan is very usual among Castilian speakers who access higher education. This is because adopting Catalan is strongly correlated with academic achievement, so that many academically-oriented Castilian speakers start using Catalan already during their teens.

5. The word ‘suggest’ in this sentence is meant to provide the necessary room for qualification of what is after all a small sample of respondents. The survey data at our disposal (see endnote 2) contained a query as to the language used by respondents when addressed in Catalan or Castilian; but this does not allow us to discriminate between groups 2 and 3, and 5 and 6 (the difference being the higher or lower readiness to accommodate). So far, discrepancies between the qualitative sample and the survey do not undermine our argument (e.g. group 1 makes up 16% of the former and 12% of the later, 10 and 12 in the case of group 7); but the accuracy of the figures yielded by groups 2, 3, 5, and 6 cannot be confirmed through survey data.