A methodological consequence of Foucault´s paradox "L´énoncé est à la fois non visible et non cache"
Department of English, American and Romance Studies Emmy Noether Independent Junior Research Group "Travelling Goods//Travelling Moods" Erlangen, Germany
In 2001 and 2003 the "Handbook Socio-scientific Discourse Analysis" was published, compiling different theoretical approaches to discourse analysis and presenting empirical researches on discourse. Here and elsewhere both Reiner Keller (2004, 2005, 2006, 2007, 2008) and Rainer Diaz-Bone (1999, 2006, 2007, 2008) set out to think about methodical devices for empirical research on discourse and consequently think about methodological issues, like design, inference and sampling, too.
Following Keller?s ?sociology of knowledge approach to discourse? ? which integrates important insights of Foucault?s theory of discourse into the interpretive paradigm (Keller 2006) ? in my trans-cultural case study "Ford in Interwar Berlin" I am tackling the task to analyse the process of social appropriation of discourse on Ford cars. Hence, I am confronted with the above mentioned methodological problems. Especially when it comes to the description of the relations between statements and of the statements themselves, we find connected to Foucault?s notion of the ?statement? this paradox: ?The Statement is neither visible nor hidden.? (Foucault 2006, p.109) ?It is still difficult to understand how something ?neither visible nor hidden? can be represented and analyzed.? notes Simon During (1992). Well, I agree with him and from my point of view this paradox has at least one far reaching methodological consequence: it affects sampling. Why? Since the ?statement? is the modality of existence for a formulation, formulations have to be inspected concerning this feature before data gathering and subsequent analysis. Consequently, we are adopting not probability, but purposive sampling. Hence, Foucault?s discourse analysis ? despite Foucault?s ?happy positivism? pronouncement ? displays an affinity to qualitative methods and interpretive methodology, for now in the planning and reflection of the process of data gathering in general. This is more than to say ?theoretical sampling? informs the collection of data, as Keller (2006) did, and it is more than to call the probing phase an exploring one, as Diaz-Bone (2006) did. Foucault?s theoretical concepts, like formulation, statement, discourse and so on, have a fundamental impact on methodological issues like sampling.