In January, I am contributing to a panel at the annual Archaeological Institute of America meeting on legacy data. I’ve already blogged a bit on this last week.
One of the unanticipated aspects of this work is that I’ve had to think about what constitutes “data” in an archaeological setting. For example, we’re studying a small corpus of lamp fragments from a particular area at the site of Polis. The “legacy data” consists of a notebook of preliminary observations from a scholar who has more or less abandoned the project. The notebook entries range from cursory descriptions to detailed documentation with measurements and comparanda. There is little in the way of analysis or synthesis.
At the same time, these legacy notebooks are data points that can be integrated into larger contexts. In fact, part of the lamps data already exists on Open Context where a version of the inventoried finds database for Polis currently lives.
Recontextualizing legacy data was perhaps the most interesting part of our work this summer. Part of the challenge is negotiating the flow between streams of data that constitute arguments. As I think more about flow – whether workflow or flow in a Deleuzian sense – I’m wondering about the relationship between flow and the character and structure of archaeological arguments. Historically, I think, archaeologists have seen data points as the structuring element of archaeological argument. In this highly empirical form of knowledge making (not to say positivistic) “data” forms foundation upon which stable archaeological arguments are built. In its most extreme manifestation, the presentation of archaeological data points can be rather “siloed.” In this situation it becomes difficult to navigate between examples of objects found at one site or in one region, for example, and those found at another site or another region. There is a tension, then, between describing objects at a site effectively and aligning an object with a type common across a region.
For archaeologists, interpretation and analysis is often about resolving this tension. At its best, archaeological work is tied to organizing and understanding objects, buildings, and contexts between the level of the site and that of the state, region, time period, or proposed trajectory of development. In other words, archaeology is concerned less with objects (however defined) and more with the relationships between objects. These relationships are navigated – vividly in a digital context – by the flow between sets of data. Workflow describes both the production of datasets from contexts and negotiating and structuring the relationships between contexts.
Analyzing and interpreting legacy data is all about finding this flow.