My research and teaching are two sides of the same coin. My current research projects ultimately are founded on the question, ‘how do digital media foster historical consciousness?’ From that question emerge ancillary questions: ‘How do digital media affect learning?’ and ‘What ways can digital media improve the teaching of history?’ To wonder about learning is to also wonder how digital media change the kinds of questions we can ask in history and archaeology, and to explore the new kinds of knowledge that can be generated. This places my research agenda squarely in the developing ‘big tent’ of the digital humanities. For my teaching, please see this page.
As an archaeologist, my research was focused on narrow questions concerning the development of the ancient Roman economy as evidenced in the Roman brick industry. That work led to my first book, Ex Figlinis (published in the British Archaeological Reports series), and to an interest in social networks in the past. Roman brick frequently carries makers’ marks, and I was able to stitch these together into a series of snapshots of the social worlds of power and exploitation in the immediate hinterland of Rome, the Tiber Valley. While I could discern these networks, the search for an explanatory framework for how they developed (and what that implied for Roman history) led me into the field of agent based simulation.
My work in agent based simulation for Roman archaeology was one of the earliest forays in the field, and I am still active in this area (where I now argue for explicit parismony in model building, and for models that are used to deform our thinking, rather than justifying our explanations). The problems of model building have led me to think more deeply about the methodological issues surrounding digital tool use in humanities research. My current research agenda involves exploring another kind of model building – topic models, a kind of natural-language processing that involves injecting semantic meaning into unstructured datasets. With Ian Milligan and Scott Weingart I published a ‘how to get started’ tutorial for topic modeling in The Programming Historian which is consistently the most cited and visited page in that e-book (with visitors staying on average 15 minutes at a time). For DH2013 I am exploring patterns in the Portable Antiquities Scheme database using a combination of topic modeling, network visualization, and geographic information system. Thinking through in public the problems of this method with this kind of data is a current stream on my research blog, Electric Archaeology.
Current major projects: