It should come as no surprise that I’m a big fan of Shawn Graham’s work. After all, my press not only published his previous book, Failing Gloriously and Other Essays, but also collaborates with with him to publish the paper version of the journal Epoiesen. Shawn was also among the first scholars to push back against my ideas of a slow archaeology and his critiques have continued to influence my current work.
I was pretty excited to read his latest book, An Enchantment of Digital Archaeology, which appeared as the first volume of Andrew Reinhard’s new Digital Archaeology: Documenting the Anthropocene series from Berghahn Press. The book is short and good, and I’d encourage anyone who is interested to read it. Shawn argues that digital archaeology can enchant by engaging the practitioner in the act of creating new knowledge largely through the media of agent based modeling. These digital models – primarily developed on the Netlogo digital modeling platform – allow archaeologists to create programs that allow them to test hypotheses by bringing their models to life. The practice of “singing” (i.e. fr. chant) these ideas to life on the screen foregrounds the creative and generative aspects of digital archaeology and emphasizes their relationship to craft as opposed to the industrial practices that I developed at the core of my calls for slow archaeology.
(As an aside, Shawn’s view of enchantment in his models reminded me so much of Robert Coover’s The Universal Baseball Association, Inc., J. Henry Waugh, Prop. which I blogged about a couple years ago here.)
Instead of offering a review (which would be premature anyway as the book has just appeared), here are a few thoughts:
1. Big Tent Digital Archaeology or Digital Archaeologies. A few years ago, there was a good bit of talk about big tent digital humanities (which Ethan Watrall connected to digital archaeology). The gist of this conversation is that by expanding the definition of digital humanities, we also open the field to a wider range of voices and participants. There was (and perhaps to some degree continues to be) concern that digital humanities has become rather narrowly defined with perhaps too great an emphasis research questions of particular interest to scholars of literature and a fairly narrow set of tools and practices associated with producing digitally mediated critical editions, various forms of distant reading, and public facing corpora of historical sources. While this critique may seem outdated today, the debates surrounding the “big tent” digital humanities almost certainly nudged the nascent field to embracing a wider range of digital approaches under its purview.
Digital archaeology has perhaps avoided a small-tent mentality, although it would be fair, I think, to discern the clear emergence of certain communities of practice in the field. That being said, it might well be time to think about digital archaeologies rather than digital archaeology.
Shawn’s book, for example, focuses heavily on the development and use of agent based modeling and the enchantment he experiences watching his ideas come to life in the interaction between agents in these digital mediated environments. This is intriguing work with applications that go well beyond the analysis of Roman brick stamps (which were the focus of his dissertation) or patronage systems in Roman Italy. One could imagine applying agent based models, for example, to Justin Leidwanger’s recent work on the Roman maritime economy and the distribution of amphora or Roman fine wares.
At the same time, this application of digital practices in archaeology is quite a bit different from the kinds of issues explored in Averett, Counts, and Gordon’s 2016 book, Mobilizing the Past which really emphasized the use of digital tools in field practices. It is also distinct from the brilliant work done by the Gabii team in producing a dynamic publication of a Middle Republican House or the work of Maurizio Forte and his DIG@Lab to create immersive archaeological environments. This is not to say that all these approaches can’t create the sense of enchantment, and the goals of digital heritage practices often seek to do just that, but rather to suggest that digital archaeologies embrace a wide range of practices, goals, and genealogies from the magical to the mundane.
2. Models and Meaning in Archaeology. Making and defining archaeological knowledge is difficult and the digital environment offers a way to think about how and what we know about the past as a product of the tools and practices that we use. In some ways, the key to Graham’s book is understanding that his idea of enchantment depends upon his view of archaeological knowledge making and this, in turn, is grounded in his reading of R.G. Collingwood.
Collingwood, an archaeologist, historian, and philosopher famously argued that all history is the history of thought. The work of the historian was to rethink past thoughts and communicate these thoughts to other people. The historian’s work at explaining the past, then, depends upon their ability to articulate in a critical way their own thoughts. These critical articulations, mostly structured as formal arguments that make visible the reason why a historian thought in a particular way, are the objects of historical knowledge. Historians cannot debate what happened in the past, because this is something that is fundamentally unknowable, but we can debate what historians think happened in the past.
Graham’s agent based models are, if I understand correctly, the basic unit of archaeological thought. They represent his ideas about the past brought to life and attention to the variable and code around which he structured the models offer a way to critique Graham’s thoughts. The models are arguments and the enchantment that Graham experiences in producing these models is parallel to the enchantment a historian feels in spelling out their view in text.
3. Digital Archaeology and Work. Graham has long taken issue with my concept of slow archaeology as a counter weight to the discipline’s growing dependence on digital technology and practices. He rightly points out that coding is a creative process and parallels the kind of craft practices that I connect to slow work in the field. This makes sense; after all WordPress has long reminds us that “code is poetry” and few of us would argue that writing poetry is industrial labor.
I’ll gladly concede that the production of robust and carefully constructed datasets, thoughtfully designed maps, agent-based models and even the digital involved in creating and publishing books has closer parallels to craft than to industrial practices. In fact, I’ve been pretty convinced by Colleen Morgan and Daniel Eddisford’s argument that single context recording, in which an archaeologist is responsible for bringing together all that is knowable about a single context is an open area excavation, represents a the basis for a creative, anarchist praxis in our field.
At the same time, I think there are elements of scale that make these practices enchanting. After all, not all digital practices in archaeology involve the integration or construction of complex models or datasets. Most of what slow archaeology critiqued was the use of digital tools in the field to collect bits of data. Unlike an agent based model, these bits of data are not complete archaeological thoughts. A photograph used to make a structure-from-motion 3D image, for example, might not even be recognizable by a human viewer as representing a particular object or landscape. The photograph only makes sense in the computer-mediated context of a 3D model generated by software like Pix4D or Agisoft Metashape. The same can be said for entering data into a database in the field or filling out a form. In isolation, a check box or even a short description does not produce a sense of enchantment and, following Collingwood, these are not complete archaeological or historical thoughts.
I’d argue that this kind of digital archaeology, the work that anticipates Graham’s enchanting models, has more in common with the fragmented work of the assembly line and industrial practices than the integrated work of craft. That the collector of digital data, especially at scale, is rarely the individual responsible for the integration and interpretation of the data reinforces the relationship between fragmentation of digital archaeological work and the organization of archaeological labor. Slow archaeology, in this context, is very much consistent with Graham’s calls for enchantment, but it also offers a more cautionary note that encourages us to see the potential for enchantment alone as only part of a larger system of digital practices in our field that both draw upon longstanding forms of archaeological organization and intensify their effects.
4. The Future of Digital Practices. Shawn, of course, recognizes that not all digital practices lead to enchanting outcomes and that not all forms of enchantment are inherently good. If the first half of the book involves enchanting models, the final chapters considers how the creation of models from massive unstructured, human-language datasets opens the doors not only to viewing how computers create structure but also on the limits of computational methods for reproducing the thoughts that we think about the past.
Shawn presents a number of examples of plausible texts produced from the autonomous analysis of historical documents. The plausibility of these texts rests on our ability to understand the context of the original documents and the style of writing and kinds of thinking that Flinders Petri, for example, might offer.
This plausibility, however, also makes the potential of this kind of technology seductive. By divorcing bit of text from their specific context and reordering and remixing them in new ways, he reveals how the fragmentation of digital data can divorce the fruits of archaeological work from individual actors, spatial and relational contexts, and relevance grounded in a sense of place. It appears that the affective potential of digital enchantment can also obscure and complicate how we understand the past.
As the code for NetLogo agent based models in the appendices of the book demonstrate, digital work requires transparency.
5. Style. Finally, the book is written in Shawn’s honest, informal, but nevertheless precise style that should be familiar to anyone how has read his blog or Failing Gloriously. It may not be everyone’s cup of tea, but it certainly fits the goals of this book and has contributed to Shawn’s distinct place in the field of digital archaeology.