Classification and Interpretation, and the Construction of Digital Resources

In “The Alchemy of Turning Fiction into Truth” (Journal of Scholarly Publishing, [July 2008]: 354-372), David Henge examines the LC classification system and its treatment of “historical” works. Noting that works catalogued under the LC classification system’s D-DX, E and F categories are generally assumed to be factually based, Henge demonstrates the error of this assumption. He opens by discussing four types of works devoted to studying the past—“history based on solid evidence and argument, history based on less acceptable forms of these, pseudo-history, and counterfactual history” (354)—but his key concern is with the cataloguing of the last kind of history. Counterfactual histories or “pretend histories”

immediately and unabashedly depart from accepted versions of the past in order to hypothesize about what the course of the past and present might have been., if only different events and outcomes had taken place. They never quite pretend that these alternative histories did occur, but they clearly often wish they had. (357-58)

Despite addressing themselves to a past that never occurred, these counterfactual works are more often than not given LC designations that place them among works of actual history. Such placement seems all the more odd if we consider that the LC system does have other categories that would better signal their status. For example, the HX806-HX811 call numbers represent Utopias, the Ideal State, and these categories often seem a far better fit for the titles Henge discusses (368). Although most of these works end up in history, a few have been correctly placed under the classification designations for fiction. That some do end up in fiction ironically boosts the factual nature of those fictional works that remain classified as history. Further clouding the status of these “pretend histories” is their frequent adoption of the trappings of authoritative scholarly work—the appearance of “maps, footnotes, numbers, and pictures with false captions” (362) as well as the imprint of a university press.

While Henge identifies general readers as the population at greatest risk for viewing titles bearing D-DX, E or F designations as credible and factually based, his study does address issues relevant to the creation of scholarly digital resources. Henge notes that although “guides to the LC classification scheme spend considerable time classifying history, they ignore the equally important task of defining it” (363). Similarly, building digital resources entails designing classification schemes, and it is important to make the logic of those systems transparent. Henge’s article usefully reminds us that classification is an exercise in interpretation and that users must understand the rationale and assumptions behind the interpretative processes employed in the various classificatory designations. Even a cursory look at the description of the TEI header on the Text Coding Initiative’s website makes the link between classification and interpretation abundantly clear.

From another, less technical perspective, the desired feedback sought by Julia Flanders and John Melson for “Exploring Reception History in Women Writers Online” represents the type of forethought necessary for anticipating users’ needs and assumptions effectively and for creating the type of supporting contextual documents that will help lay bare the thought processes involved in creating a digital resource. In the process of discussing visionary failures of the LC classification designers, Henge points out that its originators assigned the essentially the same amount of classification space to the history of Asia as they did to the history of gypsies and labels this case “the most egregious example in the D-DX (history properly speaking) classes of the failure to anticipate growth” (360). While this decision seems inexplicable, generally it can be very difficult to predict future needs and build a resource capable of growth. This difficulty is compounded by the potential of digital resources to create new perspectives and new areas of inquiries not yet imagined.

The cataloguing of “pretend histories” as actual history that Henge identifies underscores that even accepted authorities like the LC classification scheme are not infallible. A parallel to Henge’s critique, work by Jim May, Stephen Tabor and others on problems with the ESTC have already received attention on emob, and both cases suggest a healthy dose of skepticism is often warranted even when dealing with respected and well-established resources.