Grand Challenges for Archaeology
Archaeology as a social science, and its discontents
Michael E. Smith’s post is part of the Grand challenges for archaeology blogging carnival.
We do archaeology in order to learn about the past. This is a pretty broad purview. We learn about an amazing variety of past things from archaeology, from Richard III’s posture to the causes of the Maya collapse, from what the Natufians ate for breakfast to how the Plains peoples hunted bison. We work hard to recover treasure, garbage, dirt, tools, cathedrals and latrines, and we use them to make statements about what happened long ago. We ask all sorts of questions: Whodunnit? What the heck is that? Why is this particular coin sitting in that specific layer? What were they thinking?
We do archaeology for many diverse purposes, from reconstructing the lives of ancient kings and queens to creating historical narratives, from helping communities reconstruct their past to complying with government regulations. In this essay I will discuss one of those purposes—creating reliable information about past human societies. This is the primary goal of the archaeology I do, and to my mind this is the most important contribution archaeology makes to human knowledge. Another way of saying this is to claim that archaeology is a social science (Smith et al. 2012). That is, we contribute to the stock of knowledge about human societies around the globe and into the past. Our special brand of knowledge is distinctive in several ways that most of us can rattle off easily. We have access to human societies not documented by any other discipline (e.g., Natufian society). We can study change over longer periods of time than can historians. And we learn about aspects of past societies that cannot be studied well by other fields (e.g., material culture).
Over the past decade I have become involved in several transdisciplinary research projects. I have learned to interact with scholars in other disciplines (including geography, planning, sociology, political science, economics, and physics), and I have had to read widely in disciplines far outside my comfort zone of archaeology and anthropology. These experiences have made me simultaneously optimistic and pessimistic about the ability of archaeology to generate reliable information about past human societies.
My optimism has several components. First, our findings really do stand out as unique and important in bringing to light societies not accessible in any other way. I take examples from comparative urbanism, one of my research foci. Archaeology now describes quite a few fascinating and unique past cities and urban traditions—huge dense Tripolyan settlements (urban or not? How did so many Neolithic people live together?); the low-density urbanism of Mesoamerica, Africa and Asia (how did these cities work in a jungle setting?); or Çatalhöyük. Second, archaeology really does produce unique and important data—even on “historical” societies. No other discipline can study past settlement patterns, and archaeology has an incredible record of data on settlement sizes in many past regions. Third, a growing number of scholars in other disciplines actually take archaeological findings seriously and want to engage with archaeologists to learn more and to use our data. I am still surprised at this situation. Prior to going interdisciplinary ten years ago, my colleagues in cultural anthropology rarely expressed any interest at all in archaeology or my research. Now I interact with scholars in different disciplines who think our data can contribute to their research agendas. Amazing.
But I am also depressingly pessimistic about whether archaeological data can really contribute to social science knowledge and research beyond a couple of small projects here and there. Another way of putting this is to say that I think it will be hard to respond adequately to the many “grand challenges” to archaeology identified by Kintigh et al. (2014a, 2014b) until this higher-order challenge is pursued (click here for my take on those earlier grand challenges). This is my grand challenge here—to make the effort to change many entrenched archaeological practices to allow us to create reliable knowledge about past societies. We do good fieldwork, and our methods are constantly improving. We have access to a growing suite of sophisticated analytical techniques. But the results of our fieldwork and analysis are not yet building a solid foundation of evidence and data about past societies. Why not?
I see two major roadblocks:
- Our data are not available; and,
- Our epistemology is inadequate, particularly in the areas of theory and argument.
(1) Our data are not available
The “gray literature” of unpublished and poorly available contract archaeology reports is a vast sink of archaeological data. Some of it is reported in formats that can be used by other scholars, and some is not. Some is posted online, much is not. There is more contract archaeology going on around the world than grant-funded academic research, yet most of the results contribute little or nothing to building a systematic foundation of knowledge about the past. This needs to change. There are far too many academic archaeological research projects that are never adequately published, leaving important artifacts moldering in a dusty lab or analyzed data locked up in individual idiosyncratic data formats. And even archaeological data from projects that are published can be difficult to access; we need datafiles—not printed tables of numbers—and more excavation photographs—not the few that made it into the report.
This is getting uncomfortably personal now. My current project will be published and the data archived on tDAR before too long (I hope!). But the primary data for my earlier projects are sitting in ring binders, negative holders, a bunch of Excel and Access databases, and my own head. Should I start another fieldwork project, or spend my time archiving old data? The former is more fun, but perhaps the latter is of greater value to the discipline.
The challenge here is to change the behavior of individuals (like me) and to promote institutional solutions that will make our data available. Even when data ARE available, they will often need cleaning and sorting and recoding if we are going to compare our data to social science data about the present (Smith 2010). Creating and using data archives is a big part of this challenge, and that is why archives like tDAR are so important. But a change in the culture of archaeology may also be required. When an economic historian publishes a study that uses quantitative data, the datafiles are regularly posted and made available for others to use and reanalyze. Could anyone reanalyze my data that way? Not yet. We don’t have a culture within archaeology that promotes the easy sharing of data. A change in these norms would be a big improvement.
(2) Our epistemology is inadequate
Archaeology is a science, in the standard social-science conception of science as having these traits:
- Knowledge is responsive to evidence
- Claims are exposed to challenge
- Findings should be internally coherent
- Arguments should be judged on the basis of explanatory power, generality, simplicity, and replicability. For discussion of these traits, see: Gerring (2012) or Wylie (2000).
For those of my colleagues who view archaeology as a branch of the humanities, or who are hostile to science for reasons postmodern or other, please ignore the following discussion; it does not apply to you. But if one accepts the notion that archaeology does indeed (or can) conform to the model of science as outline above, then one has to be depressed about the current state of archaeological epistemology. Propositions are rarely tested, claims are too infrequently challenged, and formal arguments are rarely examined for adequacy. I explore this situation and provide suggestions for improving our arguments in Smith (2015). One result of our sloppy epistemology is that we have failed to create a solid body of empirical knowledge that can be improved, refined, and extended as research proceeds.
When postmodernism hit the social sciences in the 1980s, disciplines such as sociology and political science gave it a look, made a few changes, and got back to work. But in anthropology or archaeology, postmodernism settled in as a systemic infection and pushed empirical, scientific approaches to the margins. In the archaeology of complex societies, postmodernism is still festering (post-processualism, post-structuralism, post-humanism, etc.). The assumptions of postmodern approaches contradict the principles of science as listed above. Postmodern approaches are incapable of testing empirical propositions or carrying out rigorous comparative analysis. This is not just my peculiar view of the world; this is basic social science epistemology, discussed in a rather extensive literature rarely considered by archaeologists (6 and Bellamy 2012; Gerring 2012; Hedström 2005; Mjøset 2001; Tilly 1994, 2008).
Although few archaeologists use the phrase “postmodern” these days, this anti-science perspective is rampant in the discipline. As noted by a noted postmodern geographer, scholars now prefer to use the term “post-structuralism” because it is a “safer sounding label” than postmodernism (Soja 2001:11863). A major goal of scholarship from this perspective is to deconstruct or problematize knowledge. That is, the idea is to break down established knowledge. In science, on the other hand, the goal is to build and extend reliable bodies of knowledge. In several blog posts—see here, and here—I have discussed two types of “science” in archaeology. Science type 1 is research with a scientific epistemology as discussed here. Science type 2 is work in an interpretivist or non-scientific framework that employs scientific techniques from other disciplines as part of the research process. Archaeologists pursing this model can sometimes fool scientific granting agencies by touting their use of archaeometry or “archaeological science,” while hiding the fact that their research is governed by a non-scientific epistemology.
Only research carried out with a scientific epistemology, making rigorous empirical arguments about data and about the past, serves to build a body of archaeological knowledge that is capable of generating reliable conclusions about past human societies (beyond descriptions of individual sites or finds). The prevalence of non-scientific epistemologies in archaeology (all the “post-“ approaches) makes it difficult to create this reliable knowledge. Yet for many of us, the goal of our work is to say something useful about past human societies. As in the case of the first roadblock—data availability—the promotion of a scientific epistemology will require a change in the culture of archaeology. If we want to contribute to a body of empirical evidence about human societies and their change over time, then we have our work cut out for us. If we fail to meet this “grand challenge,” then it is hard to see how we can meet all the other grand challenges identified by Kintigh et al. (2014a, 2014b) and by the other bloggers participating in this event.
6, Perri and Christine Bellamy. 2012. Principles of Methodology: Research Design in Social Science. Sage, New York.
Gerring, John. 2012. Social Science Methodology: A Unified Framework. 2nd ed. Cambridge University Press, New York.
Hedström, Peter. 2005. Dissecting the Social: On the Principles of Analytical Sociology. Cambridge University Press, New York.
Kintigh, Keith W., Jeffrey Altschul, Mary Beaudry, Robert Drennan, Ann Kinzig, Timothy Kohler, W. Frederick Limp, Herbert Maschner, William Michener, Timothy Pauketat, Peter Peregrine, Jeremy Sabloff, Tony Wilkinson, Henry Wright, and Melinda Zeder. 2014. ‘Grand Challenges for Archaeology’. American Antiquity 79 (1): 5-24.
Kintigh, Keith W., Jeffrey H. Altschul, Mary C. Beaudry, Robert D. Drennan, Ann P. Kinzig, Timothy A. Kohler, W. Frederick Limp, Herbert D. G. Maschner, William K. Michener, Timothy R. Pauketat, Peter Peregrine, Jeremy A. Sabloff, Tony J. Wilkinson, Henry T. Wright, and Melinda A. Zeder. 2014. ‘Grand Challenges for Archaeology’. Proceedings of the National Academy of Sciences 122: 879-880.
Mjøset, Lars. 2001. ‘Theory: Conceptions in the Social Sciences’. In International Encyclopedia of the Social and Behavioral Sciences, edited by Neil J. Smelser and Paul B. Baltes, pp. 15641-15647. Elsevier, New York.
Smith, Michael E. 2010. ‘Sprawl, Squatters, and Sustainable Cities: Can Archaeological Data Shed Light on Modern Urban Issues?’ Cambridge Archaeological Journal 20: 229-253.
Smith, Michael E. 2015. ‘How can Archaeologists Make Better Arguments?’ The SAA Archaeological Record 15 (4): 18-23.
Smith, Michael E., Gary M. Feinman, Robert D. Drennan, Timothy Earle, and Ian Morris. 2012. ‘Archaeology as a Social Science’. Proceedings of the National Academy of Sciences 109: 7617-7621.
Soja, Edward W. 2001. ‘Postmodernism in Geography’. In International Encyclopedia of the Social and Behavioral Sciences, edited by Neil J. Smelser and Paul B. Baltes, pp. 11860-11865. Elsevier, New York.
Tilly, Charles. 1994. ‘Softcore Solipsism’. Labour / Le Travail 34: 259-268.
Tilly, Charles. 2008. Explaining Social Processes. Paradigm Publishers, Boulder, CO.
Wylie, Alison. 2000. ‘Questions of Evidence, Legitimacy, and the (Dis)unity of Science’. American Antiquity 65: 227-237.
Image credit: ‘Aztec Ruins NM’, jb10okie. Creative commons license.