It Doesn't Matter What You Say But Where You Say It

If I were to undertake a discourse analysis what would be the major methodological critiques? Here I am concerned with both scoping discourses during analysis and evaluating final analyses.

On the issue of scoping, are all discourses as clearly delimited as the whaling discourse that Charlotte Epstein defines in The Power of Words in International Relations (2008)? Is the whaling discourse as clearly definable as Epstein presents it?

Defining the boundaries of a discourse and assessing the power of one discourse over other discourses may be easier in certain cases. How did Epstein judge that the cold war democratization and environmentalism discourses/meta-narratives are the most important ones for backgrounding the whaling discourse? I am not arguing that they are not important. I simply aim to point out the role of the researcher in steering the analysis. What about the civil rights movement (alluded to but not expressed, p.100) that unlocked citizen agency, and the discourses around pop culture that arose with the “televisual” revolution (p.100 and 143)? What criteria did Epstein use in deciding (or not) to limit their analysis?

It seems as though the definition of a discourse’s boundaries is important for two methodological implications. The first is that since discourses prescribe “subject-positions” and their related “articulations” and “interpellations”, the researcher’s choice to foreground certain discourses or to interpret the displacement of one discourse by another will determine the possible subject-positions under analysis. These subject-positions (and subjectivities) create the context for possible paths within a discourse, as it is the actor choices and power relations with other actors that then reproduce or alter/substitute the discourse.

For example, the anti-whaling discourse was powerful because it created categories of engagement that shaped actor identities. Within an identity, such as anti-whaler, the actor has certain choices, such as the choice to not consume whale meat. Thus the researcher’s choice of discourse boundaries will influence how she/he interprets actor moves.

The second implication of a discourse’s boundaries and perceived power over other discourses corresponds to my initial concern with evaluating a discourse analysis. For a new researcher choosing a dissertation topic, Epstein’s endorsement of the “test of time” (p.173) to evaluate the validity of a discourse analysis provides little consolation.

My guess is that unlike essentialist, positive truth claims which are based on hypotheses using quantitative bodies of evidence, discourse analyzers must constantly check the consistency of their truth claims, as the constitutive pieces of the discourse (e.g. “words, actions, music, and centrally on the case of whales, images,” p. 5) build on each other, or unfold in the researcher’s narrative.

When judging the final product, the validity and value of the analysis will depend of course on the accuracy of the information constituting the discourse. But also, a discourse analysis must be judged within the context it is set, substantively, academically and culturally. It matters less which story the research chooses to tell than how (and where) she/he chooses to tell it.

Remember: "You are who you say you are, or rather you are who you speak as" (p. 169). The old axiom, "it doesn't matter what you say but how you say it" is relevant here as well. The "how" in the context of discourse analysis translates to the "where." The analysis locates speaking-actors in terms of their discourses and subject-positions. Well, maybe it matters what you say as well, but the point of discourse analysis is to highlight the importance of context.