Reaching beyond sociolinguistic corpora of interviews: Montreal 1995

Marty Laforest, Diane Vincent & Guylaine Martel

Sun. 9-10:40 C

The sociolinguistic interview, defined by Labov and used since by many variationist sociolinguists, allows one to obtain a good sample of comparable oral data, at the expense of a relatively rigid situation: the interviewer/informant relationship is neither the most spontaneous, nor the most frequent, of verbal interactions. The advantages outnumber the disadvantages when the objective is the analysis of variation and linguistic change. Previously used twice in Montréal (the Sankoff-Cedergren 1971 and the Montréal 1984 corpora), with the same informants, this methodology yielded a fairly clear picture of the state of French as then spoken in Québec.

The disadvantages predominate, however, when the objective is to analyze the specifically interactional component of informal discourse. To attenuate the disadvantages, analysts of interactions have often favoured making recordings here and there during meals among friends, situations which constitute a readily accessible source of continuous discourse. But the data acquired under such conditions permit no comparison at all between speakers, and the sound is generally of poor quality.

As an extension of the existing megacorpora in Montréal, the research undertaken in 1995 has three objectives: to pursue the analysis of linguistic change in real time; to update the study of the conversational habits of the informants, using a more ethnographic perspective; to obtain a large an diversified data base of conversational data. We have therefore chosen a hybrid methodology, one that might insure comparable data and, at the same time, offer a wide spectrum of situations of communication. Sociolinguistic interviews were conducted with twenty subjects from Montréal, who had been interviewed previously, in 1971 and 1984, and some of whom accepted to record themselves in a large number of different events of their daily lives.

Our paper will list the problems that arise with the method of self-recording, and the solutions to these problems. It will indicate the wealth of information contained in the corpus and will include as well a brief summary of the conversational habits of the informants.

References

Labov, W. 1972. Sociolinguistic patterns. Philadelphia. Pennsylvania University Press.

Thibault, P. et D. Vincent, 1990. Un corpus de français parlé. Québec. CIRAL, Université Laval.