Review: Trends in Archaeological Simulation

For a subject with a comparatively short history, the history of computational modeling in archaeology has been written many times before. The earliest attempt to establish a chronology of archaeological simulation appeared in Doran & Hodson’s Mathematics and Computers in Archaeology. This was followed over the next decade and a half by reviews by Dyke in 1982, Bell in 1987, and Aldenderfer in 1991, all of which were more or less pessimistic about the sum of contributions from archaeological simulation (with Chippindale portending its untimely demise at the hands of paradigm-fickle prehistorians with a quantitative bone to pick).

In the recent special issue of the Journal of Archaeological Method and Theory, Lake reverses course on what might be described as a rather grim assessment of simulation’s prospects presented in a special session at the 2005 SAA meeting. In that paper (published in the Simulating Change edited volume from Utah Press), he argued that archaeological simulations suffered from overdevelopment and limited application. As a result, simulation studies were likely to remain marginalized because much of the archaeological methods and theories of the day were ill-suited to make appropriate use of them.

In this new paper, titled “Trends in Archaeological Simulation”, Lake credits a revival in archaeological simulation to the advent of agent-based approaches and to simulation finding its niche in areas such as human evolution, dispersal, and household decision-making. Like other reviews (such as those appearing in Simulating Change), it breaks down the history into distinct phases. But rather than apply the first/second/third wave scheme, Lake uses a more refined timeline in which periods may overlap one another to some extent as a result of publication lag:

  • An early Pioneer phase, taking place between the late 1960s and early 1980s, prompted in no small part by Doran’s 1970 exhortation , and featuring the works of Thomas, Zubrow, and Wobst.
  • A Hiatus phase, almost entirely ensconced in the 1980s, when archaeological simulation was reeling from the post-processual critique and the recognition of computational limitations.
  • A Renaissance phase, taking place mostly within the 1990s and continuing into the early 2000s, heralded in part by the publication of edited volumes by Mithen, Kohler and Gummerman, and McGlade and van der Leeuw.
  • An Expansion phase, from the beginning of the 21st century, in which archaeological simulators begin to control their own destiny

What sets this review apart from others is Lake’s contextualization of the trends, particularly those during the 1990s. For example, the 1980s and 1990s are frequently considered to be a period when archaeological simulation was in decline, coinciding with the postmodern critique and initial disappointment in technical constraints imposed by mainframe computing of the day, and evidenced by drooping publications on the subject. Lake agrees that the 1980s represented a hiatus (although this downplays the important contributions made during that time, particularly Reynolds’ work on the Guila Naquitz project), but he argues that while the number of papers applying simulation did not increase appreciably during 1980s or 1990s, projects became longer-lived, showed increased utility, and the focus shifted away from highly specialized simulations on the periphery of larger traditional studies to more generalized applications centered on simulation. This trend is, in part, due to interest from both archaeologists and simulators in complex systems theory, a shared interest which has had an annealing effect on the theoretical position of simulation within the discipline. Rather than simulation being a niche tool used for dramatic effect by those with computer programming skills, simulation is viewed by some as an essential way of getting down to the tasks of archaeological heuristics and explanation.

There are several areas that Lake argues have seen growth for archaeological simulation: reaction-diffusion models, long-term societal change or human-environment interactions, and human evolution. Many of these have benefitted from the advent of agent-based modelling, and its successful wedding to GIS. Some are large, multi-component models, while others remain abstract, incorporating as few variables as possible. Premo’s response to Barton et al.’s paper of Neanderthal settlement patterns is a good example of this latter type, as well as the recent paper by Vegvari and Foley on cultural complexity reviewed here last month. Lake argues that these types of papers are accounting for a larger share of archaeological simulations, and that rather than being superfluous to a larger study, these often are simulation-centered studies.

There is a final set of simulations Lake calls “miscellaneous”, and these include Surovell and Brantingham’s important use of simulation to understand taphonomic biases in the use of cumulative radiocarbon data, and a recent study by Rubio-Campillo et al. which melds ABM and GIS to simulate historic battles with the aim of understanding the distribution of musketballs over space. Lake argues that what binds these models together is an interest in testing archaeological methods; this is true, there’s more to them than simple method-checking. In many of the studies that have been conducted, particularly over the past decade, the target to be generated is a social or demographic phenomenon, which itself has typically been constructed from traditional archaeological inference.  The target of these “miscellaneous” studies, on the other hand, is the number, arrangement, and qualities of objects in the archaeological record itself. If archaeological simulation continues to grow and become more incorporated into the mainstream as predicted, it will be interesting to see how the theoretical connection between model outputs and the stuff in the ground plays into justifications for and critiques of future simulation in archaeology.

After re-reading Lake’s opinion of archaeological simulation in 2005, one can’t help but agree that its pessimism was hasty (perhaps even untimely with its publication in 2010), and that “a real increase in the use of simulation was underway” even before it was being written. This new paper could serve as a touchstone for that movement, one only achievable now that some of the fledgling projects of the 1980s and 1990s have come to full fruition and encouraged a new generation of simulators. It offers a clear narrative of how simulation in archaeology has changed over time, a sense of how it came to produce the diverse types of studies now being published, and perspective on where it may be headed.

Featured image: Old Timey Computer With Black Keys by user realityhandbook


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s