Cothrom air suirbhidhean coimheadachd cànain gus data air cleachdadh na Gàidhlig a chruinneachadh

Gaelic has been categorised as ‘definitely endangered’ in the UNESCO Atlas of the World’s Languages in Danger (Moseley, 2010). This language shift has been ongoing for a number of centuries with many factors contributing to the process which has led Fishman (1991) to recognise that it is difficult to understand this process fully. One of the clearest indicators of language shift from Gaelic to English is the decline in Gaelic speakers in Scotland. The speaker numbers are one of the indicators used to assess the vitality of a language, with some minority language expert suggesting that a language needs at least 100,000 speakers to be “safe” (Krauss, 1992). The notion of speaker numbers as a language vitality indicator is based on the principle that all speakers are users of the language. Munro (2011) has , however, recognised that although the census has collected data pertaining to language competences in Gaelic, no information has been collected about the level of competency or level of language use and that, therefore, this data cannot be used to evaluate the extent to which the language is used by those who report to speak it. Information pertaining to the use of Gaelic has been collected through various research projects. These projects have mainly been focussed on Comhairle nan Eilean Siar, the local authority with the highest self-reported level of Gaelic competence in the national census. This research has shown that the extent to which Gaelic is being used has declined across many domains, but especially those associated with Gemeinschaft or intimate community (home, family and community) (Fishman, 1991), especially with the younger members of the population in the area. At the same time it would appear that the way in which the language is acquired is changing, from intergenerational transmission to the education system, with initiatives to promote the use of Gaelic in education, the media and language management (McLeod, 2010). All Gaelic speakers are at least bilingual (Dunbar, 2011) and this means that speakers of Gaelic need to make a choice over which language to use in each conversation. This language choice is based on a number of different factors, including (perceived) linguistic competences of the other participants in the interaction as well as language ideologies and current language use. Different models have been constructed to provide an explanation to this language choice, with le Myers-Scotton (1988) suggesting that there is an unmarked code (or language) – a language which is used naturally and without thought by the speakers in the community. In situations where there has not been an established code there is a need for negotiation of these language norms based on specific markers (Gumperz, 1964), for example in the social situation, the relationship between speakers, and the reasons for the interaction as well as the ideologies associated with the language. Research in Scotland (Munro, Armstrong, & Mac an Tàilleir, 2011; NicAoidh, 2006) has shown that English has become the unmarked code-choice in many social situations, even where there is an acknowledgement that Gaelic can be used. Research by Birnie (2018) has shown that the choice of code is often based on prior knowledge about the linguistic competences of the speakers as well as ideologies towards the use of the language and also that ‘the inertia condition of language choice’ (Spolsky & Cooper, 1991) means that the location or situation does not affect the choice of code and that, therefore, the measurement of language use in one domain can be used as an indicator for the use of that language in other domains or spaces. Evaluation of methodologies to evaluate language use In addition to the census data, which has provided information on the number of Gaelic speakers and the areas in which they are resident, there have been a number of research project which have aimed to evaluate the state and use of Gaelic in the community. The majority of these studies have been based in Comhairle nan Eilean Siar. These research studies have, by and large, been based on self-reporting methods – questionnaires and interviews, both with speakers and those resident in the area without (many) competences in the language. These research studies have been very valuable but, according to Urla (2013) and Hill (2008) there is a danger that the participants in these projects have recalled the use of Gaelic more frequently than the conversations in English, because these were more unusual and less common than the conversations in the majority language. As such, there is an acknowledgement that other research methodologies need to be used to evaluate the de facto language use. Research of this type has been conducted by Smith-Christmas (2016). In this study the conversations taking place within a Gaelic speaking home were recorded throughout the day and analysed. Other methodologies, such as the ‘walk-along’, where a researcher accompanies a participant to record the language use have also been used. In Montréal the conversations of groups of young people was observed covertly, a methodology also used by Ni Dhuda (2011) in her research in Ireland. The methodology applied in the Basque Country to evaluate language use has, however, been more systematic. From the 80s onwards language surveys have been conducted in the streets of the different communities in the Basque Country, the Kale Neurkata. These surveys are based on observations of conversations which are conducted in public spaces, especially the streets. This research systematically records the language used in a conversation on a data collection sheet, as well as information about the participants or speakers in the se interactions. This methodology has been used in Scotland as well, by Birnie (2018) but in indoor public spaces - more suitable to the context in Scotland. The research instrument used in Birnie’s study (conducted in Stornoway) and in the Basque Country was very similar, with additional information collected on the general purpose of the interaction (personal or business) and also the designation of the participants (member of staff or member of the public). For this pilot study the methodology employed in the Basque Country and that by Birnie (2018) were analysed to create a survey that would be suitable for measuring the language use in the Gaelic language context, especially in the geographical areas where the language remains strongest. Research For this pilot study a new research instrument was created, based on the work conducted in the Basque Country and also in Stornoway. This research instrument was designed to collect information about the main language of the conversation (Gaelic, English or another language) as well as information about the participants in the conversation. This information includes the gender of the participants and the age demographic with five categories identified; under 12, 12 – 18, 18 – 30, 30 – 60 and age 60 and over. Suitable locations for this study were investigated. This identified three different types of locations; the first two types were open locations – public spaces and community events, with the third type of location identified being closed spaces to which members of the public do not usually have access. In this pilot study different locations within the community were visited and data collected. The nature of the communities -rural with houses dispersed over a large area – meant that data collection was slower than it would have been in the towns (such as Stornoway) or even in the villages of the Basque Country where the set-up and social circumstances are different from Scotland. A transient approach was adopted for the research in the public spaces; in each of the spaces there were a limited number of individuals (members of staff and members of the public) and therefore only a short period of time was spent in each of the locations. Each location was revisited at different times of the day and different days of the week. The surveys were also conducted at community events. During these events more conversations were recorded in a short period of time, due to the larger number of individuals present. A comparison could be made between the general public spaces and during the community events to create a complete overview of the vitality of the Gaelic language in the community. This methodology did, however, prove not to be as successful in closed spaces; the majority of workplaces were small, and the research instrument was not effective in the recording of code-switching or translanguaging and it could not be used to identify the “triggers” for a change in language. The number of different conversations that could be recorded in each of the locations sampled was not sufficient to be statistically reliable and valid. In addition to this, the language practices in these locations might well have been influenced by the overt presence of the researcher and therefore other methodologies, such as language use diaries (see De Meulder & Birnie, forthcoming) might be more suitable to evaluate the extent to which Gaelic is used in these small organisations. The data collected through the language use surveys can be used to evaluate the overall use of Gaelic in the community but also provide an indication of who uses the language, and with whom. The data can be used to assess the extent to which Gaelic is used across the generation and can therefore be used to establish levels of intergenerational transmission, an important indicator of language vitality (Fishman, 1991). Recommendations This pilot study has shown that this methodology can be used successfully in Scotland, not only in towns but also in the more rural areas. Data collection is more time-consuming in order to gather a statistically valid and reliable data set of the linguistic practices of the community. Community events can be used to gather additional data; these events ha

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.