I was recently discussant for a TAG conference session on Data Activism (session 14) organised by Jeremy Foot, Joana Valdez-Tullett, and Stu Eve, all of Wessex Archaeology. Their session abstract expressed some dissatisfaction with the current uses to which archaeological data are put and challenged participants to consider a host of questions surrounding our use of data: what more we could do with it, why we might want to do more, and what stands in our way. Their call attracted an interesting variety of papers, ranging across data reuse cases (and associated problems), the sustainability problems associated with unexpected datasets, issues of comparability between different datasets with differing underlying data strategies, the unexpected application of data to unanticipated research questions, and different ways of perceiving and representing temporal dimensions in imagery, for instance.
Reading the abstracts and listening to the papers, it seemed to me that underlying this broad range of issues and approaches were questions relating to our relationships with data and the consequent implications for knowledge creation. From my notes on the presentations, I identified three key areas:
1. Our changing access to data
We can’t assume that data are unaffected by the means by which we capture, store, manage, and process it. Databases are not neutral, structuring data is political as well as technological, and infrastructures act as gatekeepers presenting particular world views and providing certain modes of access. With tools and structures that both enable and constrain, the opportunities for innovative data reuse becomes a struggle between different perspectives, different terminologies, and different recovery and processing strategies, not all of which are equally visible to an end-user. And of course, making data accessible opens it to critique, revealing some – but not all – of its weaknesses, absences, and errors. All that’s assuming the resources are there to make the data accessible in the first place: after all, the ultimate barrier to data reuse is its unavailability.
2. The changing face of data
What is perceived as data varies according to perspective: it depends on the research questions that define the data collection strategies, and these determine what is selected for record and what is set aside, unrecognised and unrecorded. Data is not self-evident, ‘natural’ or even ‘raw’, but it is fluid, unstable, and subject to multiple viewpoints. For the most part, we capture an imperfect snapshot of data at one moment in time according to a particular research inquiry. But what if those different perspectives, those moments in time could somehow be represented in new forms of data? What would be the implications for archaeological knowledge and practice?
3. The changing character of data
Somewhat paradoxically, the instability of data is what provides its flexibility: digital data are easier to store, rearrange, duplicate, and share than any analog equivalent. Moving data from one format or another, re-ordering or restructuring data, extracting data from large datasets to create a smaller, more focused dataset, and merging different smaller datasets together into a much larger one are all far easier in a digital context. None of these actions are straightforward. This volatility of digital data and the way it can shift mode and purpose presents a range of challenges – for instance, tracing the journeys that data undergo, maintaining the data threads as it is transformed and removed from its original context or purpose.
I’d argue therefore that a precursor for data activism is an understanding of our relationships with data and a recognition that these are changing – or, if they are not, that there is at least the potential for them to do so. The act of digitisation, the moment of data capture, the selection of what constitutes data and the rejection of all else, all introduce a relationship between archaeologists and data, between the observer and the observed, which carries forward into its subsequent use and reuse. It’s a complex cyclical relationship: essentially, we create the data, and the data subsequently creates us.
Quite what constitutes data activism remains largely undefined, perhaps because we remain unsure of how or in what ways we might expand on current, traditional methods of data capture and use. But if data are to be used as a driver of change or as a means of advocacy then a clear appreciation is needed of its many shades and textures, its strengths and shortcomings, ambiguities and imprecision, as well as its richness and potential. To revert to the challenges presented by the session organisers, what do we want from the data – and what does the data want from us?
[This post is based around my comments at the close of the Data Activism session. I should like to thank Jeremy Foot, Joana Valdez-Tullett, and Stu Eve for the invitation to join the session as discussant, and to thank all the participants for their presentations which sparked these thoughts in the first place.]