My thesis focuses on Breton, a Celtic language spoken in Brittany (northwest France). Breton has seen a steep decline in the number of speakers since the 1940s, the result of decades of political and social pressure. As a result, it is now an endangered language, with most speakers aged over 75. However, more recently there has been a resurgence of interest in the language, and language revival efforts have led to the establishment of Breton-medium schooling. Today there are two groups of speakers, separated by a gap in language transmission: older, traditional speakers who grew up speaking Breton, and younger speakers whose parents do not speak Breton, but who attended Breton-medium schooling. This thesis investigates what the impact of this decline and revitalisation has been on the structure of the language itself, and focuses on word order and initial consonant mutation.
To investigate this question, I conducted fieldwork in Brittany, interviewing older speakers, young adults and children who were in Breton-medium schools. Although Breton has been heavily influenced by the dominant and more prestigious language, French, and all speakers today are bilingual, it seems that Breton word order patterns are actually quite resistant to change. Breton speakers are not simply replacing the traditional Breton word order with the subject-verb-object order of French, and the general picture is one of language maintenance, rather than change – good news for an endangered language.
The two generations of speakers do differ in more subtle ways, however: first, the widespread regional variation found amongst the older speakers seems to be absent from the speech of the younger generation. This is unsurprising: this generation learn Breton at school, where a standardised form of the language is used to enable consistent teaching training and the production of Breton educational materials. Secondly, distinctions maintained by the older generation of speakers are being lost among the younger generation in a process known as levelling – some of the complexities found in verbal conjugations are not used by younger speakers. The fieldwork data also indicate that initial consonant mutation, a process whereby the initial consonant of the word changes in specific contexts, is difficult for the children to acquire: even those pupils in their early teens were not proficient in the use of mutation. However, the young adults seemed to have full use of this process, suggesting it is acquired late in speakers’ linguistic development. This is probably partly a result of the unusual context of language acquisition in immersion schooling.
Overall, the Breton of younger speakers does not differ as much as might be expected from that of older, traditional speakers, at least in terms of its structure. The gap in the transmission of the language has led to some changes in Breton, but it is not the case that the language is changing beyond all recognition. What seems to be crucial in children’s acquisition of Breton is continued exposure to the language beyond the early teenage years. Without this, young speakers never reach full competence in the language, and so are unlikely to continue using it once they leave Breton-medium schooling.