Heath and the Enormous Ethnography

Heath acknowledges in her prologue to Ways with Words that the study it contains is unlikely to be repeatable and in fact was conducted under what we might consider to be ideal conditions, given that "no deadlines, plans, or demands from an outside funding agency set limits on the time or direction of the cooperative arrangements between teachers and anthropologist" (7). This lack of constraint ended up producing an incredibly detailed ethnography with what seems to be a reasonable ambition (i.e. to track and compare how language development occurs among two different populations) but an enormous scale.

I thought it was an effective rhetorical choice to begin with the acknowledged limitations of the study (e.g. the lack of quantifiable data it produced). Yet she defends her research by saying that more quantitative results wouldn't necessarily have acknowledged the social and cultural contexts of the study. She also refers to the book as an unfinished story -- placing the importance on the narrative nature of the book, again moving it away from expectations of empiricism. This narrative tendency is demonstrated even in the first sentence. Rather than beginning with her claim, the theoretical framework for the study, a description of her method or her participants, Heath begins by setting the scene: "A quiet early morning fog shrouds rolling hills blanketed by pine-green stands of timber, patched with fields of red clay" (19). She is orienting the reader to the fact that this is a story of a place and its people. It's something different than we might expect when we think of a "study."

I take that interest in narrative to be part of the reason why she spends so much space orienting the reader to the material conditions of the participants' lives as well (particularly in  Chapter Two, where we learn everything from the layout of the towns and the decor of the homes to how money is typically spent). Yet I couldn't help but wonder if, for the sake of a coherent "story," there isn't a problem with some of the essentializing that Heath does. The gender and racial identities of these participants is often generalized (i.e. "the men do this" or "Trackton children do this"). It goes back to my post last week, where I asked about when we can/can't generalize as researchers. It occurred to me that a study like this almost requires generalization, because without it, how do you represent this (presumably) massive amount of data? Given that we're only halfway through the book, I'm reserving judgment about her conclusions, but I think it's an interesting quandry: Is there such a thing as too much data?