I’ve recently become fascinated by indexes. Partly this stemmed from a rather arduous effort to index our Oxford Handbook of Early Christian Archaeology (2018). Partly my interests stem from thinking about how and whether indices matter in the age of digital books. The ability to search a document for a particular word, for example, makes the conventional index of proper names and key terms irrelevant.
Indexes also have strange relationship to the world of the hyperlink. On the one hand, an indexes represent a one-to-many relationship. One terms links to many places. Hyperlinks are one-to-one links that connect one term to one place. In this sense, a conventional index is a helpful thing.
On the other hand, most relationships in a text are actually many-to-many. In other words, a range of possible relationships exist for any location in a text. These range from the relative simple relationship between words or concepts that are either identical lexically or so similar to be virtually synonyms to the much more complex and fuzzy relationship between related ideas, concepts, or even antonyms that require their opposite to produce meaning. Indexes, then, could relate to clouds of meaning, perhaps derived from text mining or other automated analysis of a work. This would offer a non-linear way to read a text and to understand its meaning.
Recently, however, there have been some creative efforts to engage with the indexing as an explicitly creative act. Anyone who has prepared an index (or edited someone else index) recognizes the intellectual and creative work necessary to make it a useful tool for engaging a work, but this is rarely noted explicitly. Indexers, for example, are rarely formally credited for their work.
In Lauren Berlant’s and Kathleen Stewart’s new book, The Hundreds, the authors invited five colleagues to prepare indexes to their book and these indexes with their authors offer strange wonderful, and intriguing ways of engaging the text. In Ana Paula Pais and Carolyn F. Strauss’s edited Slow Reader, they run the index on the margins of the page allowing a reader to find similar passages in other contributions and read across these passages rather than in a simply linear way.
Over the past few years, I’ve been trying to get a volume of interviews from the Bakken oil patch published, titled Voices of the Bakken and edited by Bret Weber. At various times, we’ve even released little previews of it. One of the challenges that we’ve faced is how to organize these interviews. Do we arrange them chronologically to map how attitudes toward the Bakken Boom changed over time? Do we arrange them thematically? Do we organize them according to location or the position of the individual interviewed?
One way to produce this book is not to worry very much about how the chapters are organized in the volume. After all, someone is unlikely to read this volume start to finish. More than that, since the book will be published as both a digital and paper form, simple queries can be conducted digitally with the search function on any PDF reader. Complex queries, however, require more complex reading and indexing the volume. More than that, more complex queries depend upon more subtle readings that are invariably idiosyncratic or, at very least, dependent on the particular questions and interests of a particular reader. I’d be particularly intrigued by an “affective index” that looks to understand the moods, feelings, and emotional character of the interviews. This would not, of course, preclude more conventional kinds of indexing that, say, explored relationships between individuals, a sense of home, or even just places or objects in the text.
What if we invited five or six readers to compile their own indexes to the interviews? These readers could engage these interviews in a range of ways that reflect their own research interests, which they could justify in a brief essay? Rather than indexing by page, we’d index by interview and include the key words that generated by the indexers at the conclusion of each interview, attributed to the authors, and with references to the other interviews.
This could get more wild, of course. We also have thousands of images that I started to analyze last year before getting distracted by other projects. These photos also need some kind of indexing to be useful and engaging. I’ve long considered publishing this data via, say, Open Context, but I wondered about the utility and value of a slightly organized dump of images. Maybe these images would be more useful if they were indexed according to some of the same criteria that our indexers would create for the Voices of the Bakken volume. After all, our interviews and archaeological investigation of workforce housing in the Bakken informed one another. There are obvious links between these two data sets, but also the potential for more creative ways to link these two sets of documents.
A project that links the interviews and the images would embody some of the ideas behind “slow data” that archaeologists have discussed recently. It would also demonstrate explicitly how publishing and curating data is work that creates new constellations of knowledge that revolve around critical engagement that starts in the field and continues through the organization of data for publication.
To be clear, I haven’t yet convinced the editor of these interviews to go along with this kind of approach, and I’m not sure that I could find willing indexers. More than that, indexing thousands of photos seems like a daunting task, but one that would be worth it even if done on a relatively small scale.