Reflexivity Redux: A Pithy Polemic on "Positionality"

It is now taken for granted that a good ethnography should be "reflexive." But what exactly does that mean? Most basically, reflexivity describes the capacity of any system of signification, including a human being—an anthropologist—to turn back upon or to mirror itself. Twenty years ago, Jay Ruby's edited volume, A Crack in the Mirror: Reflexive Perspectives in Anthropology (1982), confirmed the arrival of reflexivity in our discipline, although the concept itself was not entirely novel. In their jointly authored introduction, Ruby and Barbara Myerhoff recounted the "ancient" history of reflexive thinking, citing the case of stories about storytellers telling stories, and drew attention to some of the "early" experiments in reflexive approaches to ethnography, such as Elenore Smith Bowen's Return to Laughter (1954) ¹ and Gerald Berreman's Behind Many Masks (1962).

Twenty years ago, reflexivity was proposed as a corrective to a mode of ethnographic writing in which factual material was presented by an omniscient yet invisible author-narrator whose methods of fieldwork and data collecting were not always manifest, and who did not address the effect of her or his presence on others, much less the various effects that others may have had on her or him. Twenty years ago, there were no established conventions for writing a single manuscript reflexively, as opposed to a factual ethnography followed by a more diary-like "confessional" text. Since then, many modes of reflexive writing have been explored and reflexive lexicons introduced. Twenty years ago, it may have been premature to question the excesses of the nascent concept of reflexive anthropology, but now that "reflexivity" is a commonplace, it is time to revisit the concept and to initiate a critical but hopefully salutary appraisal of its many uses and guises, beginning with a consideration of the relationship between reflexive thinking and mirroring. What follows are some informally presented thoughts, or "talking points," about reflexivity and its conceptual offshoots provoked and quickened over the year by ideas, insights, and concerns gleaned from a broad range of books and articles as well as from working with graduate students of anthropology. ²

The word "mirror"—both the noun and the verb—was frequently encountered in anthropological circles, whether as part of a title or as a description of didactic technique, well before the publication of A Crack in the Mirror. In Clyde Kluckhohn's Mirror for Man (1949), anthropological inquiry was likened to a mirror held before us to allow and encourage a better understanding of ourselves through the study of others. However, a mirror is not an inert device and can be deployed to contain or control differences and oppositions. Thus, in The Chrysanthemum and the Sword (1946), written during WW2, Ruth Benedict's mirror rendered Japanese national character intelligible as American national character the other way around. In order to make America's "most alien enemy" appear more human, Benedict positioned Japanese and Americans as the mirror image of each other: "[t]he arc of life in Japan is plotted in opposite fashion to that in the United States" (Benedict 1946: 253-4; see Robertson 1998). However humane her anthropological motives, Benedict made getting to know Japan too easy, and the Japan she profiled was all too knowable, a legacy that Japan anthropologists continue to grapple with in different ways.

The potential of mirrors to create the conditions for both solipsism and reification should be obvious. Moreover, an anthropologist's metaphoric mirror can render her or him the de facto ethnographic subject, and turn the self into the fieldsite. Egocentrism is one of the pitfalls to avoid in exercising reflexivity. A laudable theory about epistemology, reflexivity is also a method for conducting fieldwork and constructing ethnographies. A reflexive anthropologist intentionally or self-consciously shares (whether in agreement or disagreement) with her or his audiences the underlying assumptions that occasion a set of questions. Those interactions in turn guide the ways in which answers to...