Categories and Data

The readings this week discuss the ways in which literature has been studied and categorized in the past and the ways in which these categories have been affected by digitization and other digital technologies. According to the readings, literature has been traditionally been organized according to author, genre, language, and historical period. Ed Folsom, in his article “Database as Genre,” asks what would happen if we didn’t have such categories. By what criteria would we analyze and evaluate ‘literature’ (I use literature in quotes because this term is also problematic – what constitutes literature and what doesn’t? Also, what is a text vs a literary text?)? The Stanford site mentions some other ways that scholars have analyzed literature – style and gender for example. Another approach has been material and/or paleographic, or in other words, grouping and studying texts according to material criteria, such as binding, script, size, paper, ink, and whether a book is printed or a manuscript.

However, I’m not sure to what extent digital technologies really affect these categories. In fact, I might argue that these technologies, and databases and visualization techniques in particular, actually reinforce them. Folsom mentions that when creating a database one has to create very “particular” categories in order for the computer or a program to understand the data. The degree of specificity and rigidity required for some programs actually makes matters more even more complicated and problematic. Weingart, in his article on networks, discusses how these requirements create “artificial” categories that do not account for more complex or even uncategorizable data. Folsom uses Whitman to discuss this issue, and in particular, the problem of assigning his works a singular genre. Also, the fact that data is put online does not change the fact that categories are a product of the person creating them, even if the computer might process the data differently.

Narrative vs. Database

Folsom also discusses the supposed opposition between narrative and database. A database, he argues, does not inherently construct a narrative because its form allows for a non-sequential way of ordering information. An archive or a book must use narrative or some ordering mechanism to organize data for the reader. However, while data does not always or necessarily construct a narrative, data does often tell a story. One reason for this is that we decide what data to put into a database, which is not random; that is, the data is curated to some degree. Also, narrative is an important and useful way to make sense of data, especially if the data is presented to an audience. In addition, narrative is a very effective way to communicate the information that the database provides.

Advantages of Databases

Databases, particularly those with large amounts of data, allow scholars to collect diverse data and put it in one place and to find connections that might have been hidden. Another advantage of databases is that they permit people to create their own narratives and categories, whereas genre or other forms of categorization frame readership prior to the encounter with a text or texts.

Side note: I also think there is a difference between narrative(s) and a grand narrative. I look forward to discussing this in class!