A discourse of classification: what type of "thing" is the myers-briggs type indicator and what makes it "work"?

Theorists who study discourses are interested in the social construction of reality through talk and text. The discursive construction of reality occurs at different (though interconnected) levels - from transient and situated instances and episodes oflanguage use (Potter & Wetherell 1987), through to the circulation of durable sets of interrelated knowledge claims, usually created and maintained by certified experts attached to institutions such as schools, universities, hospitals, prisons and courts (Foucault 1972, 1978). In this paper, we explore some aspects of these different dimensions of discourse, through a consideration of the creation, circulation and use ofthe Myers-Briggs Type Indicator (MBTI), a popular personality typing tool. We argue that closer attention to the characteristics of discourses and the practices that surround them - that is, how they are created, maintained and structured - can help us to understand how they work to construct reality across a variety of contexts.