Willard McCarty’s recent post to the Humanist discussion list, number 22.008, was a response to a 1984 talk by folklorist Bruce Jackson, “Things That from a Long Way off Look like Flies,” on the history of folklore as a discipline and the nature of folkloristic evidence. Willard’s post, “The Fragility of Boundaries,” ended with the following observation:
The manipulatory abilities of our digital tools…are of course just the thing to translate anxious fragility of categories into amazing agility for categorizing and re-categorizing raw material. That much is plain. But what about the equally plain fact of the (truly) exponentially increasing volume of data? The problem, it seems to me, is not the hermeneutic nightmare of arbitrary, unjustifiable choice but the ease with which evidence for just about anything may be found. As Northrop Frye used to say, given enough data any statement can be connected with any other statement. Now we actually have the data, at the push of that lovely button.
Willard’s post will be archived, eventually, here. Bruce Jackson’s talk, published in the Journal of American Folklore, is available for subscribers to JSTOR. My own response, as posted to the list, is as follows:
I am reminded of two bits of advice from my academic training, both of which seemed massively intimidating at the time, and therefore memorable. The first was from an undergraduate literature seminar: “pay attention to everything.” (‘Really, professor? Everything?’) This was a mandate, of course, to read closely but also to be alert to the wider social world in which the written word is embedded.
It was repeated in a graduate-level seminar on material culture, through Ian Hodder’s Reading the Past. Hodder describes context as “the totality of the relevant environment,” which, for the purposes of archaeological interpretation, is always a subset of that totality: that portion of it that is accessible to the interpreter. Scholars have always had to decide what is relevant to the explanation of a thing, and these decisions represent boundaries.
Disciplines, then, are the product of long-established patterns of boundary-drawing. Scholars of architecture long held that all that was relevant to any discussion of a great building was the form of other great buildings (Pevsner was famously explicit about this: “A bicycle shed is a building; Lincoln cathedral is a piece of architecture”). In time, with the emergence of architecture as a profession, attention turned to the biography of individual architects as a significant piece of evidence. Not relevant, however, were questions of contemporary social and economic practices, so it has only been in the last 30 years or so that slavery, for example, has been introduced into discussions of Virginia’s plantation houses, or gender relations into the study of domestic space.
Richly networked digital evidence represents, surely, a greater subset of relevant material than what a single researcher has previously had at hand. What this situation demands is that scholars who work digitally are self-conscious, and explicit, about how they draw those boundaries, which potentially extend quite a bit beyond where disciplinary habits might once have drawn them. I’m not sure that this situation shows the fragility of boundaries so much as their malleability. I will continue to study buildings; another will concentrate on novels; some others animals. If everything is digitized, everything is available as context, so our choices may be made more freely about where we situate our subject and how expansively to draw our boundaries.
But digitizing everything is a more serious problem for some pieces of evidence than others. For those of us who work on the material world, or who understand the material world as a meaningful aspect of context, digitization is far from straightforward. I share Bruce Jackson’s envy of literary scholars, whose subjects can be dis-embodied from their physical carriers and remain computable in a useful (if imperfect) way. Scanning a building, or a fork, or a landscape, is quite a different matter, as the results are very poor substitutes, as evidence, for the original. As with folklore, material evidence is always mediated and always collected according to contestible assumptions about what is significant.
As we learn to do research in an ever-expanding environment of digital evidence, we should be mindful of how that evidence has been collected, which varies significantly according to subject (or, if you like, discipline).