Language in context seminar
Speaker: Frans Gregersen (University of Copenhagen)
Title: The role of the situation in providing comparable data for studies of language change and variation
Abstract: In any study of language change and variation we need data. Such data can be generated in many ways. Now, video cameras can be mounted on head-sets so that the interviewer/field worker documents gestures and actions on the part of the informants along with the spoken data. But audio-recording of spoken data may also be as regular (sociolinguistic) interviews or as part of some kind of ethnographic field work: The situation recorded may be ‘real’ in that it has not been caused by the data collection process, or it may be ‘artificial’ in that the field worker has invited him- or herself to an informant’s home or work place in order to get data. In such cases there are a number of obstacles to overcome.
In the New York Study (Labov 1966) William Labov set an example which has been followed by many since then in laying down some ground rules for sociolinguistic interviewing. Labov 1966 (1972: chapter 3) distinguishes contexts likely to lead to a more casual style: Speech Outside the Formal Interview, Speech with a Third Person, Speech not in Direct Response to Questions, Childhood Rhymes and Customs and The Danger of Death Question. If we analyse the difference between these five contexts, we notice that for the first three the field worker is at the mercy of life unfolding around the interviewee, while for the latter three the sociolinguist may bring about a change which may or may not lead to casual style. Common for the first two contexts is that some other person trumps the field worker as the audience (Bell 1984) and common for the latter three is that some theme has an attraction or reportability which overrides the constraints of the situation at large.
In the LANCHART (Language Change in Real Time) project at the University of Copenhagen which I directed 2005-15, we re-recorded 243 informants who had been recorded once before included in five previous projects. Re-recording that many people generated a lot of new data but in order to make data comparable we had to process the data for sources of variation coming from the situation. In my paper, I will discuss how we may make claims to have controlled for the important parts of the recording situation – which by the way is extremely dynamic – and how credible such claims are if we demand that what is represented is the so-called vernacular, in both the old recordings and the new ones.
If you would like any further information about the Language in Context Seminar Series, or have any recommendations or feedback you’d like to give us, you are warmly invited to contact either Stephen McNulty or Sarah Van Eyndhoven at:
To sign up to the mailing list, and be kept up to date with future events, please follow the link below, log in and click “Subscribe”:
To find out more about the Language in Context research group within the University of Edinburgh, you can visit the page below:
We hope to see you very soon!
Co-ordinators of LinC
Language in context seminar
Room 1.17, Dugald Stewart Building, 3 Charles Street, Edinburgh, EH8 9AD