A final archaeological future

I just got back from a long trip to Northampton, Williamstown, Oberlin, and Cleveland. The trip began with a talk entitled “Doing archaeology in a digital age: challenges and opportunities” at Smith College’s The Futures of Classical Antiquity. I gained a lot from my interactions with colleagues on social media and from the blog posts that friends wrote in response to my blogging on the topic (see, e.g., Bill Caraher’s post on professionalization). Here’s the talk as I gave it, with some minor annotations and some links to references:

 

—

“Archaeology is at an exciting juncture” (Atalay 2012: 1). This is true for the discipline of archaeology as a whole, and for Classical archaeology. There are a million ways I could proceed. There is the issue of how archaeologists engage with the public generally, and local communities specifically. There is the issue of ethical practices, both with respect to the profession and with respect to the outside world. Archaeologists are struggling with issues of preservation, against the rising tides of looting and development, and against ourselves: how do we work sustainably so that existing sites are adequately maintained once they’ve been excavated and so that materials are effectively stored and made available for study? How do we deal with the “data avalanche” that we’ve unleashed so that our work is carefully archived, and how good are existing structures for dissemination, academic and popular? Are our curricula working for our students? These are all interesting questions that most practitioners, I suspect, are worried about, and I am certain that there are many more questions that I could pose. But because this is a symposium about the futures of Classical antiquity, Ι want to focus on the relationship between Classical archaeology and the discipline of Classics.

I want to preface my remarks with two brief comments: “Classical archaeology” is a problematic term that I’ll use here as a shorthand for “the archaeology of the ancient Mediterranean.” And in this talk I’ll be focusing almost exclusively on field archaeology rather than the many other things that Classical archaeology could refer to: art history, topography, and so on; in part because field archaeology is what I know best and where I think there is the most potential for transformative change.

Let me start with what’s good about Classical archaeology. Even bad Classical archaeology for 2017 is pretty good Classical archaeology in the grand sweep of its history (I owe this point to Donald Haggis). I think this mostly has to do with increasing professionalization: Classical archaeologists see themselves as professionals that need to uphold basic standards that correspond to the field’s expectations. That wasn’t always so: Classical archaeology has traditionally been a rich white man’s game. As I entered the field in the mid-1990s, the last of the great self-funded excavations were coming to an end. These kinds of projects have a long history stretching back to the very origin of the discipline: Schliemann’s excavations at Troy and Mycenae, paid for from his own pocket. The current crop of archaeologists, although they largely come (I suspect) from affluent backgrounds, do not have the financial wherewithal to pay for their own field projects, and so typically they fund them through a mix of external grants (public and private), university research funds, and private donations. It is worth noting at this point that Classical archaeology is still a rich man’s game, in that the institutions of Classical archaeology are relatively wealthy and rich private donors continue to exercise influence. But individual practitioners in the Mediterranean are now part of a broader professionalized academic archaeological community and see themselves as such. This opens the field to those with the requisite abilities and skills, rather than those with the economic and symbolic capital. This is not to say that problems don’t remain. Representation of women in directorial ranks remains a serious problem, and one that is especially prominent considering how many female students and scholars there are in Classical archaeology. A quick count of American projects in Greece shows that male directors outnumber female directors 18 to 4 (cf. the Canadian Institute: 14 men, 8 women).

The second part of the improvement is that the archaeological tools and methods at our disposal have increased infinitely in sophistication since the time when I entered the field in the mid-1990s. For those of us who entered the discipline then, it wasn’t uncommon to be trained to use a dumpy level and a stadia rod, a 19^th century technology for measuring elevation. Excavation was essentially just stratigraphic digging, using traditional tools: big picks, hand picks, and trowels. We recorded our work in notebooks, at various levels of standardization and consistency. We measured using tapes and plumb-bobs, we drew by hand on graph paper. The only real technological innovation in the previous 100 years had been film photography (Rabinowitz 2016). The lack of technical innovations at that time reflected not so much the availability of technologies but rather the research questions that Classical archaeology sought to address. In what is, I think, an apocryphal story, when a Greek prehistorian asked the director of a major American excavation why he didn’t use a water sieve, the (joking) response was “Because inscriptions don’t float” (cf. Whitley 2001: 57).

That joke reflects the old orientation of the discipline, which was to illustrate the world of the texts. For instance, the 1996 excavation summary of the Athenian Agora begins with the claim that the excavations for that year shed light on (1) the city’s destruction at the hands of the Persians in 479 BC and (2) the form of the great statue of Athena Parthenos. This approach was, of course, logical. Classics is a discipline defined by a narrow canon of texts and the philological skills required to read them in the original languages. The canon focuses, of course, on particular times and places, like 5^th century Athens, and so why not devote enormous resources to understanding the greatest moment of Athens’ history and her greatest artistic productions? Training archaeologists at a high level in the ancient languages also followed logically from this orientation: if the telos of archaeology was to follow a trail left by texts, then what could be more important than being able to read and understand them? The other part of this is that the ancient world began with the text such that the material remains were understood as epiphenomenal. It was thus natural for Coldstream in 1976 to understand tomb cult (attested archaeologically) as the product of a text, namely Homer.

No doubt this is a too-simple characterization of a complex situation, and sub-fields like Greek prehistory have always had different priorities and therefore made use of a broader range of methodologies. (For instance, the water sieve was introduced to Greece in 1969 by the excavations at Franchthi Cave). But we can see that the traditional orientation of Classical archaeology towards texts is preserved in the way that graduate programs and other institutions train and evaluate their students.

Today archaeological excavations are much more sophisticated with respect to analytical techniques. Virtually all projects now have a systematic strategy for flotation, and some analyze phytoliths. The use of soil micromorphology, especially on excavation, has become increasingly common in the last ten years. Remote sensing, which was essentially limited to analysis of aerial photographs in the 1950s, now includes multispectral satellite images as well as archaeological geophysics. Palaeoenvironmental studies have helped us to reconstruct ancient coastline and climate, as well as anthropogenic episodes of landscape change. Scientific methods for the study of archaeological materials has also flourished.  Technical analyses of human bone make use of macroscopic forensic analysis as well as scientific analyses of various isotopes and DNA; so too does analysis of animal bones make use of a combination of macroscopic and scientific analysis. Ceramics are being studied in a wide variety of ways, including petrology and chemical analyses, and their residues are increasingly being tested. I could go on and on: scientific and microscopic analyses are now de rigeur for stone tools, metals, and indeed every type of archaeological material. Scientific methods used for absolute dating are still rare outside of prehistory, however.

These analytical techniques are increasingly central to the shape of the discipline. That is, the research being produced by archaeological scientists is increasingly central to the research questions of the field, and that’s reflected in archaeological literature, field practices, and institutional structures. Not only are journals of archaeological science flourishing, but central issues are increasingly tied to scientific research. Whereas these methods were once especially associated with prehistoric archaeology, they are increasingly being used on field projects of all types, from long-standing urban excavations to newer projects like the excavations at Gabii or the Roman Peasant Project. Specialists aren’t just coming in at the end of the field season to examine their materials, but are increasingly part of the daily practice of field archaeology: the lab is coming to the field. And archaeological sciences increasingly appear in our teaching, too: a course in the archaeology of the Aegean Bronze Age taught today includes a lot more discussion of scientific techniques than one taught twenty years ago.

This shift in archaeological practice comes both from a surge in technological innovation in the aftermath of the Second World War, but also from important shifts in research priorities away from culture history and towards paradigms that emphasized the historical processes at the root of cultural change. The influence of the “New Archaeology” that dominated North America in the 1960s and 1970s certainly encouraged interest in the “interaction between humans and the natural environment,” (McDonald 1972: 3) the application of scientific approaches in archaeology, and the kind of interdisciplinary scientific collaboration that is now common in the Mediterranean. In Greece, at least, the standard-bearer was the Minnesota Messenia Expedition (MME), a project that combined regional survey and excavation with historical studies, materials analysis, ethnography, geomorphology, soil science, physical anthropology, geochemistry, geophysics, and so on, all with the goal of “reconstructing a Bronze Age regional environment” (McDonald and Rapp 1972). The penetration of ideas from the New Archaeology entered the Mediterranean largely via prehistorians like Bill McDonald, the co-director of MME, in part because these ideas had already circulated in nuce among them (Blegen 1941). Already in the interwar period, prehistorians were advocating for research projects that focused on long-term change and “the conditions of life of the humble commoner” (Blegen 1921: 125-126). The focus on the everyday is still visible nearly a century later in recent work in social and economic history and regionally-oriented archaeology. I’m thinking here of Horden and Purcell’s The Corrupting Sea, Paul Halstead’s ethnographic magnum opus, Two Oxen Ahead, and the Roman Peasant Project, a collaboration between Penn, Cambridge, and Siena that seeks to identify and excavate the dwellings of the Roman rural poor between the 2^nd c BC and the 6^th c AD in western Tuscany through a combination of archaeological field survey, geophysical survey, and excavation; archaeobotany, zooarchaeology, organic residue analysis, and microtopographical soil reconstructions.

The mention of The Corrupting Sea points at a significant convergence between archaeology and Classics over the past two generations. When Classical archaeology’s purpose was to illustrate texts, it didn’t really have much to offer Classics. Excavations in Athens might illustrate some aspect of the Persian sack in 479 BC, but ultimately the literary narrative remained primary and the archaeological evidence secondary. The Corrupting Sea, on the other hand, is a work of ancient history that is built up from the concerns and evidence of archaeology, regional or landscape archaeology in particular, especially the micro-region, the emphasis on environment, the infinite complexity of economic strategies, the element of risk, the constant movement of people and goods. Even their micro-regional case studies are drawn from those covered by survey projects (e.g., Melos). Here and elsewhere in the past 30 years we can see a real engagement with archaeological evidence by ancient historians, and a willingness on the part of archaeologists and historians to use archaeological evidence to contravene well-established, textually-derived truths.

Similarly, shifts in the study of Greek literature under the influence of a variety of approaches have allowed for a convergence between Classics and archaeology (broadly construed). As Classicists have increasingly understood texts as the products of various historical forces, social, economic, and political, and the more interest is paid to marginalized groups in the ancient Mediterranean, the more of a role archaeological evidence has to play. Three years after Coldstream declared hero cult to be an epiphenomenon of epic, Greg Nagy’s Best of the Achaeans put hero cult and epic poetry on equal footing, such that the elaboration of hero cult and epic poetry were parallel developments that were intimately linked to each other. This trend only accelerated with new historicism. Essentially, once Classics focused on power and politics, it began a kind of historical and archaeological turn.

Yet while archaeology and Classics are converging in such a way that archaeological evidence and analysis has much more to contribute to mainstream discussions in Classics, I also see an accelerating split for two main reasons: (1) the rapid proliferation of archaeological evidence and scholarship and (2) the rapid expansion of the core skills of archaeology, which are becoming increasingly technical. Professionalization has led to a super-specialized mode of knowledge production.

I’ve already discussed the second point above briefly with respect to archaeological sciences, but the other piece is the rapid expansion of digital technologies. A number of projects have gone 100% digital. Even the most committed digital skeptics (I count myself in this group) are functionally digital archaeologists: as a co-director of an archaeological field project I need to use ArcGIS, a spatial database program, and other database applications that interface with ArcGIS. We also use basic drone photography and photogrammetry on our survey. In my capacity as a co-director of the Pylos Tablets Digital Project, I am in charge of RTI, a kind of computational photography and I deal with three-dimensional scans and evidence from X-Ray Fluorescence. Because most all archaeological projects, whether field or museum in orientation, make use of large quantities of data, including specialist data, digital databases are really necessary, not just to keep track of the data, but to integrate it in ways that allow it to be usefully queried and used.

We can get a sense of how complex this operation can be by briefly looking at the Gabii Project. Gabii migrated from a paper project in which forms were entered into a database to a direct-to-digital project, with all data, photographs, etc. entered directly into the database. Spatial data from total stations were directly exported to ArcGIS, which was versioned daily and managed by a dedicated digital data team; multiple users could edit spatial data simultaneously. For almost all contexts, the project also made 3D photogrammetric models. The Gabii Project recently published a mid-republican house in digital form with the University of Michigan Press (for $150!) which makes use of all these data to provide an extremely data-rich and high-tech presentation. When you consider that Gabii is a large project – I counted 37 staff, and there seem to be as many students working on the project in any given season – you can start to imagine how complex their data structures necessarily are and the sheer size of their data avalanche. I am certain that Gabii’s data are enormous compared to what I deal with at the moment. In my small museum-based project digitizing Linear B tablets, we produced nearly one terabyte – one terabyte! – of digital photographs in our first season. Just backing up our data was a significant undertaking. On my field project, we have nearly 7,500 survey units, each of which is mapped in GIS and has a significant quantity of attached data, photographs and descriptions, and we have almost 19,000 records of artifacts that we’ve collected and analyzed from those units. After three field seasons, our data archive is only 346 gigabytes but includes almost 60,000 files in over 1,000 different folders.

As you might expect, dealing with these data requires a certain amount of expertise, like the ability to code, and a lot of patience and time. I don’t expect that this data deluge or the need to manage it will decrease anytime soon. If the past ten years is anything to go by, it will only accelerate. Indeed, as more and more projects publish their raw data, it will become increasingly attractive for ambitious students to embark on “big data” analyses that integrate and organize data generated by multiple projects.

I don’t think there’s any doubt that the “digital turn” in archaeology is enormously important. It wouldn’t be impossible to go back to an analog archaeology, but it would be incredibly difficult. One of the best arguments for digitization and born-digital data is (in my opinion) that it gives supervisors more interpretive tools in the field, by allowing them to call up previous years’ work easily on their computers or tablets, not to mention things like digitized reference works and scholarly literature (Poehler 2016). This is a way to bring as much information as possible to bear on solving interpretive problems in the field. I stress this point because if field archaeology is anything, it is that: thinking through problems in the field, at the trowel’s edge. Excavators are not just collecting data – they are interpreting as they go, and their interpretations shape their data. To do better archaeology, we need better interpretations more than anything else, and digital tools have allowed us to stretch our capabilities to pull up information of various kinds, especially outside of the library. So the science lab and the computer lab have both arrived in the field.

The digital turn does present some challenges that I want to briefly discuss. (Although I’ve discussed them at some length on my blog, so I will try to be brief). First, it seems to me that there is almost a fetishism of new digital technologies, a misplaced technological solutionism. This faith leads to the denunciation of old technologies as if they were false idols. For example, a number of publications in the past 30 years have denounced GIS as “reductionist, positivistic, and lacking engagement with cultural and social factors” (Sullivan 2016) because it represents the world as a flat Cartesian space that ignores the realities of ancient lives and experience. I never understood this critique. Human beings don’t live on two-dimensional maps, sure, but none of the tools at our disposal really capture the lived lives of ancient people anyway. (In fact, GIS doesn’t even represent the “natural” surface of the earth accurately!) If we get a sense of false objectivity from GIS or any other digital tool, it is because we are not doing enough thinking. Digital tools are research tools for organizing and presenting evidence, not models that produce answers.

There is on the other hand a tendency among digital archaeologists to obsess about data. Digital practitioners celebrate their ability to produce “high-quality data with manifold improvements in accuracy and efficiency” (Roosevelt et al. 2015). This move creates an artificial separation between data collection and interpretation. The old analog practices displaced by digital photography and photogrammetry were both “data collection” and interpretation at once: I’m thinking here primarily of drawn top plans and sections and illustrations of all kinds, as well as text descriptions of drawn features. Drawing is a “slow” process in which the act of recording is also an act of careful attention and interpretation. This is being replaced by a much more accurate and efficient mode of documentation, but it can be an unthinking one. When I do photogrammetry or other kinds of photography that will be computationally manipulated, I am thinking about making an accurate computational model, and not about the artifact or feature that I am photographing. The benefit, I tell myself, will be that I will have a beautiful model to interpret at a later date, and one that I can publish and share with my colleagues. There is a real power to that sharing. For example, in the Pylos Tablets Digital Project, we will be able to publish and share high-quality dynamic digital models of each and every administrative document from the Mycenaean “Palace of Nestor.” This will, we hope, have the effect of redistributing the editorial authority from a single editor who has examined the texts via autopsy to all qualified users. But this mode of data capture is boring. Anyone who has spent days on end doing photogrammetry or RTI will tell you: it is mind-numbing. We are told the gains in efficiency that our digital tools afford us provides more time in the field for quiet contemplation, but I am a bit skeptical of that claim. In my experience in digital projects, the impulse is to collect increasing quantities of data. You can’t throw a stick at an archaeology conference without hitting someone who is at that moment nodding and saying, “we just need more data.”

You can see the emphasis on data collection in the publications of digital recording systems, which constantly harp on efficiency and accuracy to the virtual exclusion of any discussion of interpretation or theory. This is not to say that digital archaeologists are robots: surely not. But there is at the very least a rhetorical problem and a reluctance to deal squarely with the fact that the proof of archaeological pudding is in the interpreting. More data and more accurate data do not automatically produce better interpretations. We cannot regard data and methodological sophistication as ends in themselves: our goal must be better interpretation. That means that we need to discuss improvements in actual field methods as much as we discuss improvements in documentation – after all, excavation is itself an interpretive process, and if you dig the site wrong, all the 3D models in the world won’t save you – and there needs to be much more explicit and informed discussion about the relationship between methods and theory.

But, as I asserted before, we are all digital archaeologists now, and for good reasons. Steve Ellis (2016) elegantly summarizes some of the benefits: better data (cleaner, more accurate, and more efficiently gathered), more dynamic data, more secure data, and more easily accessed data. But there are also significant intellectual benefits. For example, archaeological field survey in Greece tends to produce highly complex patterns of artifactual distributions; they have been described as a “continuous carpet” of artifacts of different types and as “palimpsests” of activities of different periods. Early surveys would define archaeological sites in the field and immediately map them out and document them, but my experiences in the Peloponnese made me skeptical of the claims that sites were so easy to define. I (and others) have argued for the past 20 years or so that the best way to define sites was actually in the computer lab, once all of the relevant data had been collected and digitized. That’s because the in-the-field way of doing things is data-poor and not very sophisticated in terms of its interpretation; the computational way of doing it is data-rich and involves more thinking. The value of the digital way of doing it is that you can display and comprehend more evidence than is comprehensible than when you’re in the field. So although this way of defining sites separates data collection from analysis, this separation has definite interpretive benefits: more information and more space to think. In fact, one of the things that I’ve noticed about bad digital humanities is that it treats the interpretive process as something that is eased by big data and quantitative approaches. That is to say, once the digital model is produced, it’s felt that the hard work is somehow over, that the model gives us some clear insight into meaning. In fact, I think the value of digital approaches is precisely that they make interpretation more difficult by flooding us with evidence, including contradictory evidence. And that’s appropriate, because good interpretations take time and effort.

 

It’s time for me to start moving towards a conclusion. The future prospects for Classical archaeology are, despite the problems that I’ve highlighted, very bright. We have an explosion of exciting new methodologies that bring in new evidence, expose the discipline to new perspectives, and give our work new power. Projects like the Roman Peasant Project are asking important questions and answering them with a well-designed mix of modern methodologies. Our projects are becoming more professional and less hierarchical, due to gradual changes to the composition of practitioners, the increasing importance of specialists, and the explosion of evidence and scholarly literature. It’s now impossible for a single person to control the data produced by modern projects or the many approaches being used in the field. That makes the discipline necessarily more collaborative, and that is in my view a good thing.

Challenges remain, however. As Classical archaeology becomes more archaeological in approach, it also becomes less Classical. When I was applying to graduate school, I was told by my advisor that if I wanted to do archaeology in Greece, I should go to a graduate program that required significant training in both ancient languages. I took his advice, perhaps too literally, and consequently spent most of my time in graduate school working on languages and literatures. It turned out that I wrote a dissertation on a subject, Linear B, that required precisely those linguistic skills (at least the Greek), but my interests were always broader than Mycenaean epigraphy. This put me in an awkward position when I got to the job market: as a prehistorian, I couldn’t get jobs as a historian; as a person whose dissertation was about texts, I didn’t look like much of an archaeologist. I had to pick up most of my archaeological skills in my spare time and over the summer, when I spent as much time as I could in the field. As these skills multiply, even the most diligent and best trained students will find it difficult to keep up.

The on-the-fly, in-the-field instruction that characterized much of my training is often accepted as a necessity in Classical archaeology, but in fact it is a serious problem. Like all scholars, archaeologists need time to learn their materials in such a way that they can work creatively with them to solve problems. There are no short cuts here. To write her dissertation, my partner analyzed 4.5 metric tons of pottery from Corinth, which, she estimates, took her about 10,000 hours to study. That works out to about three years of working ten hours every day. I don’t really believe in the “10,000 hour rule” as popularized by Malcolm Gladwell – that 10,000 hours is some kind of magical threshold after which one is an expert – but I do think it points at something important, which is that good work requires time: time to become expert, time to be creative, time to make mistakes, and time to think. The consequence is that we cannot train well-rounded Classicists and expect them to become expert archaeologists.

My big worry is that there is a growing chasm between what makes good Classical archaeology and how an archaeologist gets a job in Classics. This mismatch between professional incentives and how archaeology will move forward is clearly unsatisfactory. I’m worried about brilliant students who do brilliant work that sheds important light on the ancient Mediterranean, but who can’t get jobs because their research is based on archaeological science. This problem is hardly new: it has been with us for the past thirty years at least (McDonald 1991), but I suspect that it is getting worse. Unfortunately, we are already behind our colleagues in anthropology and archaeology departments, and we will fall still further behind as our tools and approaches continue to expand. Classical archaeologists are archaeologists, and they need similar training to archaeologists in other departments (Newhard 2017).

We might imagine, then, Classical archaeology ripped out of Classics departments and redeposited in newly-minted archaeology departments. But perhaps that is a feeling that comes out of a misplaced longing for disciplinary coherence along formal lines. That is to say, that there should be archaeology departments as there are economics departments, organized around a more or less coherent set of methods (rather than subject matter). Compared to most academic departments, Classics is extremely compact and coherent. Whereas a cultural anthropologist might ask “when was the last time that research on hominid evolution or primates was helpful to you in thinking about your ethnographic data?” and expect the answer “never” (Segal and Tanagisako 2015), I suspect that the equivalent question in Classics would provoke the opposite reaction. That is, “when was the last time that Classical archaeology was helpful to your thinking about literature?” or “when was the last time that literary studies were helpful to your thinking about archaeology?”).

I don’t think that separation from Classics would be good for Classical archaeology. Many of its strengths come precisely from its close association with the Classics: a long tradition of fieldwork, robust data sets (archaeological, epigraphical, artistic, and literary), broad recognition and appeal, and strong institutions. Classical archaeology also has within it an enormous amount of expertise, and we have natural allies within our Classics departments. The success of historicizing approaches in particular has created a better environment than ever before for integration and cooperation in a sophisticated way.

Indeed, for me the interest and power of a discipline like Classics is in the application of different approaches that are focused on a relatively limited chronological and geographical expanse. When Classicists define the field in narrow and totalizing ways, I find it boring and small-minded. I’m not particularly interested in a Classics that defines itself, as a colleague once did, as a discipline knit exclusively around linguistic competence in Greek and Latin, control of essential textual skills, and familiarity with all ancient genres. As James Clifford (2005: 27) points out, “knowledge does not…naturally sort itself out in professional segments, and institutionalized domains of academic practice are necessarily dynamic and relational.” Clifford also observes that “a discipline does not actually need consensus on core assumptions. Rather like a hegemonic alliance…it requires consent, some significant overlapping interests, and a spirit of live-and-let-live across the differences.”

I think that Classics can do better than that, however. We can retain coherence and the flexibility for new directions within the discipline. If Classics will be a narrow discipline, then so too will Classical archaeology. But if Classics can accommodate methodological diversity, then Classical archaeology can be a strength and a source of energy and new ideas. It will be most successful if it can forge cooperative and collaborative bonds within Classics and between Classics and other disciplines, especially those in the social sciences and the natural sciences (and fields like “Digital Humanities”). I don’t want to imply that we can snap our fingers and achieve this: cooperation isn’t something that just happens; it is something that is “learned, practiced, and honed” (Perry 2017). It requires effort and experience, and it will necessarily entail different models of teaching and learning, publication and promotion, from the norms of the discipline. It is a radical change from the old models of knowledge production, but it would be a natural development from the current collaborative models of knowledge production emerging from modern archaeological field projects. If this becomes a reality, then Classical archaeology will also have contributed a great deal to Classics as a whole. I hope that none of us thinks that a smaller umbrella and a less expansive view of the discipline and its position within the Academy is any way forward.

Advertisement