ABSTRACT
One of the longstanding and rarely challenged conventions of sociolinguistic research is that linguistic data should be both ‘spontaneous’ and ‘naturally occurring’. This convention was probably derived in early sociolinguistic work from traditions in
dialectology, an approach that was careful to exclude speakers who are not authentically representative of the speech of the particular region. Within sociolinguistic work, however, the notion of authenticity was extended to the collection of speech styles. Not only were speakers to be deemed as authentic speakers of the regional variety, but the authenticity of the speech style (i.e. ‘casual style’ representing ‘vernacular speech’ versus ‘careful style’ representing ‘standard speech’) should also be validated. In his early work on the stratification of /r/ in New York, Labov (1972: 61) addresses the problem of authenticity in what he calls the ‘Observer’s Paradox’ by noting that ‘our goal is to observe the way people use language when they are not being observed’. In the same volume he suggests a way to overcome the observer’s paradox:
We can also involve the subject in questions and topics which recreate strong emotions he has felt in the past, or involve him in other contexts. One of the most successful questions of this type is the one dealing with the ‘Danger of Death’: ‘Have you ever been in a situation where you were in serious danger of being killed?’ Narratives given in answer to this question almost always show a shift of style away from careful speech towards the vernacular.
