Learning by Doing: Archaeological Excavations as "Communities of Practice"

The international and multidisciplinary Keban Dam Rescue Project between 1966 and 1975 in Eastern Turkey brought scientists together to document and study the past of a landscape about to be submerged. The archaeological teams at Keban each constituted separate groups united around what they were to do in the field. This article examines the manner in which members of these archaeological "communities of practice" learned how to undertake Turkey's first salvage excavations. If such communities can form the basis for both archaeological knowledge and learning, they can also become the source of exclusionary practices, historical erasures and epistemic injustices. Introduction Archaeology as a discipline exists thanks to the interconnectivity of multiple communities of practice across the world. Such communities have not been the focus of any explicit attention on the part of archaeologists yet. The history of archaeology, for instance, has examined the ebband-flow of different theoretical paradigms, the evolution of scientific technology in the field, the manner in which ideologies such as nationalism, colonialism or imperialism influence what we know about the past. Many of these (Trigger, 1989; Murray and Evans eds., 2008) are excellent studies. As far as I am aware, however, none of them detail what archaeological practice really is. Elsewhere, other scholars have provided more critical accounts of archaeological practice (Edgeworth ed., 2006; Hamilakis and Anagnostopolous eds., 2009). These ethnographies of archaeology, however, never focus explicitly on community as the source of practice. Matt Edgeworth, for instance, in his thought-provoking Acts of Discoveries, examines archaeologists' "social transactions" (2003, p.41) without necessarily considering the team itself as a whole. More generally, if the decisions or interactions of particular individuals on a site are described, it is the act of "digging" as a group that is never entirely mapped out. The term community is far from absent in the archaeological literature of course. It is generally employed to describe the ancient communities of the past or the native communities of the present, that is local stakeholders living near the archaeological sites themselves (Marshall 2002; Atalay 2012; Rosenzweig and Dissard 2013). In both cases, these communities are almost always them and only rarely us. And yet, teams in the field oftentimes operate as communities of practice without necessarily being aware of it. In fact, it is perhaps because communities are so pervasive in archaeology that we tend to forget about them. This article remedies the lack of analysis of archaeological practice at the level of the group, an unchartered territory between the individual and the institutional, and deliberately adopts community as the scale of analysis in order to better examine what it is that archaeologists do in the field. The Keban Dam Rescue Project undertaken in Eastern Turkey from 1966 to 1975 serves here as a specific case-study to examine archaeological excavations as communities of practice.


Introduction
Archaeology as a discipline exists thanks to the interconnectivity of multiple communities of practice across the world. Such communities have not been the focus of any explicit attention on the part of archaeologists yet. The history of archaeology, for instance, has examined the ebband-flow of different theoretical paradigms, the evolution of scientific technology in the field, the manner in which ideologies such as nationalism, colonialism or imperialism influence what we know about the past. Many of these (Trigger, 1989;Murray and Evans eds., 2008) are excellent studies. As far as I am aware, however, none of them detail what archaeological practice really is.
Elsewhere, other scholars have provided more critical accounts of archaeological practice (Edgeworth ed., 2006;Hamilakis and Anagnostopolous eds., 2009). These ethnographies of archaeology, however, never focus explicitly on community as the source of practice. Matt Edgeworth, for instance, in his thought-provoking Acts of Discoveries, examines archaeologists' "social transactions" (2003, p.41) without necessarily considering the team itself as a whole. More generally, if the decisions or interactions of particular individuals on a site are described, it is the act of "digging" as a group that is never entirely mapped out.
The term community is far from absent in the archaeological literature of course. It is generally employed to describe the ancient communities of the past or the native communities of the present, that is local stakeholders living near the archaeological sites themselves (Marshall 2002;Atalay 2012;Rosenzweig and Dissard 2013). In both cases, these communities are almost always them and only rarely us. And yet, teams in the field oftentimes operate as communities of practice without necessarily being aware of it. In fact, it is perhaps because communities are so pervasive in archaeology that we tend to forget about them. This article remedies the lack of analysis of archaeological practice at the level of the group, an unchartered territory between the individual and the institutional, and deliberately adopts community as the scale of analysis in order to better examine what it is that archaeologists do in the field. The Keban Dam Rescue Project undertaken in Eastern Turkey from 1966 to 1975 serves here as a specific case-study to examine archaeological excavations as communities of practice.

The Keban "Spirit"
! 1 The rescue project eventually ended in 1975 when the area studied was no longer accessible. Over the course of eight seasons, a total of 28 sites had been excavated (see table below). Despite the initial absence of funds, the relatively small number of volunteers, and the fact that an international and multidisciplinary rescue project of this scale had yet to be undertaken in Turkey before, Keban is remembered today as a "successful operation" in the history of Turkish archaeology (Erder, 1978, p.3).
Keban does indeed represent a unique moment. The project was initiated by METU's Kemal Kurdaş and Istanbul University's Halet Çambel whose energy and vision rallied different scientists around their cause. It was supervised, for the most part at least, outside of Turkey's usual heritage institutions by members of METU's relatively young faculty. It was the first time, and perhaps the only time, that such a large number of excavations were monitored all at once by such a loose and improvised structure. The excavations themselves were led by archaeologists whose enthusiasm and determination forged solidarity in the field.
The participants I was able to interview often described the rescue project as having taken place in a collegial, even friendly environment. Some of them even referred to its intellectual atmosphere as the Keban "spirit." But, what exactly does that mean? I believe the rescue project's "spirit" is to be found in the group itself. Again, notwithstanding differences in disciplinary expertise and national origins, archaeologists and other scientists were able to coalesce as a community of practice in an effort to rescue the past of a region about to be submerged. Other factors such as the adverse working and living conditions, the necessity to rescue quickly, and the vicinity of the sites brought researchers to unite in the field.
Keban, finally, was carried out during an unusually peaceful period for the region, just before the start in the 1980s of the armed conflict across Eastern Turkey between the Turkish military and Kurdish guerrilla groups like the PKK. The Keban Dam was followed by the building of other mega-infrastructures, first on the Euphrates (Karakaya and Atatürk Dams in the 1980s, Birecik and Karkamıș Dams in the 1990s) and later on the Tigris River (Ilisu Dam today). All of these constructions were preceded by salvage excavations like the ones carried out at Keban. For different reasons, however, including the heightened presence of the military in the region due to the civil war, the same collegiality would not be found again in these subsequent projects.
It is all of these factors combined that created Keban's collegial atmosphere. However, it is not the "spirit" itself that created the community. A spirit, just like a shared passion for the past, might create some kind of solidarity in a group, but it will not be enough to sustain a community. The notion of spirit describes Keban's collegiality well enough. Something else, however, was necessary to reinforce the members' already strong bonds and push researchers to coalesce as a community. And this is to be found, I argue, in what they did on the sites themselves. It is through their practice, in other words, and not just in spirit, that archaeologists in the field came together as a group.
What is it that archaeologists do? "Digging" is perhaps the first word someone will associate to archaeology. What archaeologists do on a site, however, can not be narrowed down to "digging" only. The tools of archaeology are not limited to its pickaxes, shovels and trowels. Archaeological practice is constituted instead of ! 4 hundreds of different acts, most of them having nothing to do with removing earth from the ground. Ufuk Esin again illustrates this point: "The time remaining from the actual work of excavation was, as in previous years, devoted to the restoration, classification, filing, card-indexing, photography and sketching of the excavation material. At the same time "systematic surface collection" was carried on with the aim of establishing the distribution area of both the Tepecik and Tülintepe mounds." (Esin, 1974, p. 123) Here, a series of activities (restoring, classifying, filing, card-indexing, photographing, sketching, collecting) carried out in the field is enumerated. Usually not mentioned in a site report, a wide range of techniques, specific means to an end, used to produce scientific knowledge about the past is disclosed. Elsewhere in the report we read about drawing, recording, organizing, cleaning, cataloguing, archiving, sorting, analyzing, identifying, processing, filing, as well as a countless number of other acts performed to transform a mound of earth into an outdoors laboratory.
Archaeological practice, however, is not everything archaeologists do in the field. None of the activities listed above, for instance, are ever carried out alone. An isolated act on a site would not make much sense on its own, quickly becoming futile or absurd if undertaken outside an archaeological community: "Practice does not exist in the abstract. It exists because people are engaged in actions whose meanings they negotiate with one another." (Wenger, 1998, p.73) Excavations are made up of actions whose significance is constantly being mediated within a group. What archaeologists do, in other words, only makes sense when shared with others, to the point that archaeological "practice" and archaeological "community" can never be understood separately. It is collectively that archaeologists make what excavations are in practice, and it is in their practice that archaeologists make what excavations are collectively.
The concept "community of practice" was first developed precisely to grasp this relationship within a more general social theory of learning (Lave and Wenger, 1991). Moreover, community and practice are inseparable when articulated around the following three dimensions (Wenger 1998, p.73-84). First of all, a community of practice exists thanks to its participants' mutual engagement. An archaeological community of practice is more than a group of friends spending time together. Archaeologists assembled for a relatively short period of time in the field sustain dense relations with one another. They all share an active participation in the project, mutually engaging around what they are to do. It is important here to remember that such communities are less defined by what its members do, but more by the way they do it collectively. And, what often differentiates a community of practice from, let's say, a personal network, is the quality of the relationship among its different members, a particular kind of bond between participants continuously reinforced by their mutual engagement in what they are to do.
Likewise, a community of practice is more than just a group of people mindlessly completing chores because they were told to do so by a superior. Second of all, a community of practice consists in a joint enterprise. This is more than just the tasks that need to be completed by its members and more than the totality of individual goals and common objectives in the group. Instead it consists of all the problems and solutions systematically being shared and discussed within the community. Here, Wenger places the emphasis on joint precisely because the enterprise is constantly up for negotiation. A sort of passive participation on an excavation site might create unity in a team, but it will not be enough to sustain a community. Individuals ! 5 are brought together instead to work within an area of knowledge that is continuously being explored and developed, questioned and negotiated.
Third of all, a community of practice possesses a shared repertoire. This is defined by "routines, words, tools, ways of doing things, stories, gestures, symbols, genres, actions, or concepts that the community has produced or adopted in the course of its existence, and which have become part of its practice." (Wenger, 1998, p.83) On an excavation site, a shared repertoire can correspond to a team's specific style or discourse, including such things as excavation tools, artifacts unearthed, daily schedules, site reports, or even shared stories and inside jokes among team members. Again, members of a community unite around what they are to do and how they are to do it. An excavation's shared repertoire is made up of the objects and ideas around which archaeological practice begins to make sense. It is, in other words, archaeological practice reified. And, oftentimes, it is the most visible sign that a community of practice is indeed thriving on an excavation site.
It is around these three key terms (mutual engagement, joint enterprise, shared repertoire) that practice becomes the source of a community. At Keban, rescuing the past took place following a tacit knowledge of archaeological practice. Despite differences in academic interests, theoretical inclinations, disciplinary specialties or national traditions, archaeologists and others united in the field around what they were to do. These researchers coalesced as a group in their shared commitment to study the past of this soon-to-be-submerged region, constantly negotiating the manner in which historic landscapes should be surveyed, ancient monuments preserved, and archaeological sites excavated. They exchanged information, helped each other, learned from one another, and sustained dense relations to achieve their joint objectives. Again, it was in their practice that researchers made what the rescue project was collectively, and it was collectively that researchers made what the rescue project was in practice.

Learning by Doing
Keban constituted a learning experience for an entire generation of scientists. It offered a young cohort of academics, many of them in their 30s, their first fieldwork experience. Cevat Erder who led the initial survey with a team of graduate students, for example, was a newly appointed professor at Middle East Technical University. The project was a productive learning experience for Mehmet Özdoğan as well, who launched his career at Tepecik after participating in the joint Istanbul-Chicago Southeast Anatolian Prehistoric Project and before becoming one of Turkey's most prominent archaeologists. Elsewhere, Harald Hauptmann at Norşuntepe felt the responsibility, bestowed upon him by the German Archaeological Institute in Istanbul, and its director Rudolf Naumann, to successfully excavate the region's largest mound. Others also had to demonstrate their worth to the ones skeptical of the rescue project in the first place, like David French at the Aşvan sites, for instance, who felt his superiors in London "breathing down his neck." 3 This new generation experimented with freshly developed tools at the time, many of which like the surveyor's levels and theodolites were still in a trial-and-error mode. Water sieving machines used to sort micro faunal and floral remains were also tested by some at Keban. Another groundbreaking invention and original tool at the time, a computer, was later used to refine the pottery classification established during the 1967 survey. Not all of these new techniques were successful at first. The radio-magnetic surveys carried out to locate buried architecture before actually physically unearthing it seemed promising at first (Yaramancı, 1970(Yaramancı, , 1971). In the end, however, the magnetometer achieved limited results, leading some to wonder what precisely the machine could do (Hauptmann, 2010, personal communication, January 22).
Similarly, a few samples were taken to a newly established carbon-14 dating laboratory at Middle East Technical University from Keban. The results of this pioneering technique were less than satisfactory but are nonetheless remembered today as a first in Turkey. Elsewhere, the grid system, which all archaeologists take for granted now, was also developing into one of the discipline's most common tools. Ufuk Esin tested the grid at Tepecik, but quickly realized that it had not been measured correctly during the first season: "In 1968 some errors were discovered in the topographic plan of the mound and its environs. Until the mistakes were corrected the trenches were labeled not according to the grid system, but in alphabetical order... During the 1969 season these mistakes were corrected. Consequently in 1969 we began to use the grid system again." (Esin, 1971, p. 120, note 2) Such innovative techniques like the grid system had to be learned and experimented with in the field itself. We should not forget that Keban took place at the height of the New Archaeology, a time when Science and Technology promised more objectivity in the production of archaeological knowledge. Thus, new tools were accompanied by a renewed faith in science, which also participated in creating the Keban "spirit." Moreover, the surveyor's levels, theodolites, sieving machines, computers, magnetometers, grid system, Carbon-14 were key in forming the project's communities themselves. The tools helped some researchers who knew, or learned, how to use them integrate the group, while also excluding others who might have been less accustomed to them.
I mentioned earlier how the term "community of practice" was originally developed by Jean Lave and Etienne Wenger within a broader social theory of learning. In their book Situated Learning: Legitimate Peripheral Participation, the two authors describe the process of learning in a variety of ethnographic case-studies such as elementary school children learning how to solve math problems through collaboration; apprentice tailors in Liberia who learn their craft by "stealing with their eyes" from their masters as well as their fellow tailors; members of Alcoholic Anonymous groups learning to articulate a narrative about their struggle with alcohol by listening to the stories of older members of the community who have already undergone the transformative process from "drinking non-alcoholics" to "non-drinking alcoholics." (Lave and Wenger, 1991) Most learning, Lave and Wenger explain, does not take place passively, as one might assume, reading a book or sitting in a classroom. It is instead a situated practice that happens when individuals actively participate in the social world. Above all else it is a transitory, fluid and directional process. In all of their examples, learning occurs as an individual moves from the margins to the core of the community of practice. This movement is what they refer to as legitimate peripheral participation. It allows an individual to learn what to do and how to do it while simultaneously integrating the community from the outside in. It is this same movement inwards one observes on an archaeological site as new participants integrate the team while learning how to execute some of the steps necessary to carry out the excavations.

! 7
At Keban researchers experimented with new tools while becoming members of the rescue project. As peripheral members of the community learned to use these new methods, they moved from its margins to its center. Keban might have been full of unknowns at first. Its participants oftentimes did not know what to expect in the field. They made mistakes and ran into unexpected problems. They had to overcome many trials and errors, as well many absences and uncertainties by experimenting and improvising solutions all the time. And, when experimentation or improvisation was not enough, they relied on each other to solve unexpected issues in the field. In the end, it was precisely these challenges that conditioned the archaeologists to unite as they learned to find solutions and solve problems through cooperation.
Again, a shared interest for the past is not enough to create an archaeological community of practice. "Communities of practice are groups of people who share a concern or a passion for something they do and learn how to do it better as they interact regularly." (Wenger-Trayner, 2015) What I am foregrounding in this article is the manner in which archaeological excavations are learning processes. What united the members of the rescue project at Keban is not only to be found in what they did and how they did it, but also in the manner in which they learned how to do it. It was the participants' regular interactions that not only sustained unity, but also brought forth learning in the group. In the end, archaeologists and other scientists who came to Keban in an effort to rescue its past learned how to do archaeology in the field by simply doing archaeology.

Conclusion: Inclusions and Exclusions
The term community emphasizes togetherness. Here, it conveniently expresses the rescue project's collegial atmosphere. I do not wish, however, to exaggerate the group's harmony. Keban as a community was far from ideal. There were plenty of tensions, disagreements, jealousies, cliques and gossip. It brought a few universities to collaborate, for instance, but did not erase their existing rivalries. Hierarchies were also strictly respected. Keban witnessed a clash between two generations. If younger scholars were indeed given a chance in the field, they remained under the strict supervision of more senior colleagues from a distance. A few researchers played a central role in the project, while others acted more passively on its peripheries. Moreover, the rescue project in spite of its multidisciplinary dimension remained first and foremost archaeological. Other "side" projects by botanists, architects or sociologists were carried out only if time and money allowed.
Furthermore, Keban's international dimension did little to disturb the unequal balance of power engrained in Turkish archaeology at the time, which made learning amongst teams somewhat unidirectional. European and American teams dominated the discipline in the 1960s and managed to excavate the Upper Euphrates' most prestigious sites. They imposed their research agenda on Turkish teams that remained for the most part passive recipients of foreign theory. Keban also took place at the height of the Cold War. As such, no scholars from the Soviet Union were invited to the Upper Euphrates. Armenian archaeologists, for instance, could not work in Turkey at the time, even if some would have liked to excavate in what had once been part of Greater Armenia. In addition, manual labour hired in nearby villages had almost no stake in Keban's research agenda. Even though the "digging" would never have taken place without them, these local people remained excluded from the excavations and were given few opportunities to learn and integrate the teams themselves.
Communities of practice are usually tightly defined groups with strict boundaries. As such, the term makes us think of both inclusions and exclusions. This last remark also has broader implications at the epistemic level. Archaeological knowledge can be understood as the property of archaeological communities because they are precisely the ones that decide what counts as the archaeological record, what constitutes acceptable facts, what represents credible explanations. As such, the theoretical concept I have adopted and developed in this article to think about excavations is not only key in grasping the process of learning, but also in understanding the social fabric of knowledge itself. It is important to remember that the shape of Keban's archeological communities--who made it in the group and who remained on its margins--directly affected the kind of knowledge produced.
In other words, as teams working in the Upper Euphrates defined their research goals, scientific identities and excavation styles, they simultaneously delineated the kind of discoveries they made about the region's history and prehistory, including and excluding specific facts, objects and stories, accounting for but also omitting certain archaeological sites, levels and periods. In the end, the term community of practice not only makes us think about what archaeologists do on a site, but also what they might forget to do, or what they might, more or less deliberately, sideline, erase or throw away. The rescue project at Keban could have been a chance in the 1960s to explore the Upper Euphrates' more inconvenient past for instance. And, to be fair, a few researchers attempted to do so, albeit without much success or determination. In the end, the more contested or taboo history of Eastern Turkey, if not completely ignored, remained sidelined by Keban's scientific communities, left for others in the future perhaps to explore more thoroughly and study more critically.