 |
|
|
|
|
|
John P. Hale Department of Anthropology Ball State University
GIS in Archaeology: a Survey
Introduction History is littered with instances where a piece of technology has completely revolutionized the methodology within a discipline. Certainly the science of astronomy came about solely because Galileo invented the telescope. Biology really only blossomed after the invention of the microscope. While not all changes are as dramatic as the changes wrought with the invention of the telescope or microscope, many other fields have been transformed, though in much subtler ways. It may well prove to be just such a revolution within archaeology with the coming of age of geographic information systems. The purpose of this paper is to introduce and explore some of the advantages of using a geographical information system (GIS) in the field of archaeology. While the exact definition is still a matter for debate, very simply, a GIS is an information management tool for organizing, storing, and manipulating geographically-oriented data, with the added ability to perform analysis and predictive modeling (Savage 1990:22-3). A GIS is more than a simple mapping system, such as a computer-aided drafting (CAD) system or a database management system, because it has the ability to generate new data from existing data. "GIS…are systems that interrelate, manipulate, and analyze a variety of geographically distributed data in addition to mapping" (Kvamme 1987:10). The usefulness of a GIS in archaeology cannot be considered without first establishing a clear definition of archaeology. According to Vincent Gaffney, archaeology is "the study of past societies in their entirety, from the analysis of their cultural and environmental remains, and through the inferences which may legitimately deduced from such remains" (1995:371-2). Since the focus of archaeology is past societies and on cultural and environmental remains, it follows that the vast majority of archaeological data is temporally and spatially referenced. Until the advent of GIS systems, printed maps were the only means by which an archaeologist could record spatial data. But although the switch to a GIS may seem like a natural movement for archaeologists, Ezra Zubrow (1990) warns that using a new analytical tool such as a GIS "is not equivalent to a mechanic changing wrenches." According to Zubrow, the changes go much deeper, and the methodology, and even the theory, adapts to take advantage of the new tool. The very nature of the questions being asked undergoes change. Like ripples on a pond, the repercussions will move outward and eventually encompass the entire field (67). The archaeological universe is not made up of discreet, defined collections of artifacts; rather the landscape itself is a palimpsest that extends in all four dimensions. One of the ripples being felt on the archaeological pond already is the advent of landscape archaeology. According to Savage (1990), landscape archaeology is the "study of spatial relationships among humans and their physical, social and cognitive environments" (29). Or, as stated by Crumley and Marquardt (1990), "landscape is the spatial manifestation of the relations between humans and their environment" (71). As the methodological advantages of GIS are felt, landscape archaeology provides the necessary theoretical framework in which GIS-driven research can be applied. Previous to the introduction of GIS systems, research aimed at social organization and spatial distribution could only be handled through complex mathematical models that were arduous in both implementation and interpretation. Without the integrated mapping and database management of a GIS, landscape archaeology was at "a methodological dead-end" (Savage, 1990:29). Advantages to Using a GIS in Archaeological Research Synthesized Records and Data Exchange One of the most immediate advantages to utilizing a GIS in archaeological research is data management. Digitized records of archaeological records provide for data exchange on an unprecedented scale, and allow the synthesis of information across large areas. As a result, information can be incorporated into a much larger arena, particularly when information can be shared — in both directions — with local and federal government agencies (Allen, Green and Zubrow, 1990:385). The convenient, speedy exchange of information will help to expedite important archaeological work, particularly in the area of cultural resource management (CRM), but also for research archaeologists. The exchange of information across borders extends beyond the relatively transparent borders of state governments in the United States, however, and in Europe the exchange of data made possible by GIS systems is becoming particularly important. In Europe, current boundaries certainly bear little resemblance to the boundaries of prehistory, and research interests can easily extend across multiple borders (Hansen, 1993:229). To that end, many European archaeologists are attempting, through various projects such as the EUARCH project, to establish a common database format for easy exchange of archaeological data (Hansen, 1993:232). Systematic Testing of Hypotheses A second major application of GIS technology is the degree to which hypotheses that deal with culturally relevant landforms can be systematically from a variety of angles and perspectives. Existing hypotheses can be tested for validity, and new hypotheses developed and tested again. For example, Madry and Crumley (1990), investigating Celtic hillforts in the Arroux River valley in Burgundy, France, used line-of-sight analysis to attempt to determine how ancient roadways were placed. The results of their investigations revealed previously unseen relationships between the hillforts and the placement of ancient roads; they discovered that the roads tended to stay within a line of sight of a hillfort. Interestingly, when Madry and Crumley tried to recreate the roadway network, they were unable. While all of the variables that determine the placement of ancient roads are not yet known, they have isolated many that had not been known before. To attempt to test their hypothesis without the GRASS GIS that they used would have been so time-consuming and involve so many variables as to be impractical. Marcos Llobera conducted another line-of-sight study on the area surrounding enigmatic ditches carved into the chalk downs of the Salisbury Plain in Wessex in Southern England. According to the prevailing theory, the purpose of the ditches was as territorial markers intended to be seen from within the area they bounded. Through the use of a GIS, Llobera was able to prove that, in fact, the ditches could not have been territorial markers based on their orientation and visibility from key points. While Llobera could not explain the actual purpose of the ditches, his study opened the door for other ways of looking at the ditches and perhaps developing a more accurate understanding of the ditches. Predictive Modeling One of the most widely applied uses of GIS systems are site location models, that, much like a suitability study, highlight areas that meet certain criteria and thus will probably contain an archaeological site. Archaeological predictive modeling seeks to establish a causal relationship between certain environmental parameters and known archaeological site locations. The use of predictive modeling is primarily a CRM tool, since it allows contract archaeologists to identify areas that are likely to contain archaeological materials in advance of the actual need, allowing them to respond quickly to record check requests required for state and federally funded construction projects (Savage, 1990:26). Martijn van Leusen, after using a GIS to perform "cartographic modelling" in the South Limburg area of the Netherlands, predicts that the GIS will become as important a computer application as the wordprocessor in the archaeologist’s toolbox (1993:122). Visualization of Patterns All of the advantages mentioned so far have in common the fact that they all allow archaeologists to see patterns that might not otherwise be visible. In the case of landscape studies, the complexity of the data, expressed mathematically, does not lend itself readily to the identification of patterns. As Kvamme (1995) argues, the visualization of patterns in data is a crucial tool in archaeological research (7). In fact, the skill is so important in landscape archaeology that Marble (1990) argues that human spatial behavior has not been explored before because we hitherto lacked "the tools which would permit us to organize and comprehend the data defining the real and extremely complex spatial environment in which human behavior actually takes place." According to Marble, the myopic view of spatial behavior in the past was due to our inability to visualize the range of human spatial interaction, and therefore we could not hope to effectively model that behavior (18). Examples such as the Wessex linear ditch project show the skewed way in which questions can be visualized, and how GIS can quickly place the landscape into proper focus. In another analysis based on viewsheds, David Wheatley (1995) noted a marked difference in the intervisibility between Neolithic long barrow tombs in the Stonehenge and Avebury regions of Southern England. While the older tombs in the Stonehenge area are all within a line-of-sight of other long barrow tombs in that area, the tombs in the Avebury region show no such relationship. Additionally, long barrow tombs in the Stonehenge area are primarily of an earthen construction, while those in the Avebury area are of megalithic construction. Previously, differences in the construction of the tombs were attributed simply to the materials available, however, the change in intervisibility between the tomb sequences indicated by the viewshed analysis suggests that something more complicated was at work. Wheatley suggests that this shift may represent a change in the social structure from one of integration into the landscape, as represented by the tombs in the Stonehenge sequence, to one of authority and importance, suggested by the tombs in the Avebury sequence. The introduction of the GIS allowed Wheatley to ask questions that may not have occurred to archaeologists before, simply because the tools were not available to answer such complex questions. As a result, greater tool/theory interaction can be expected in the future, as the true capabilities of GIS are explored (Savage, 1990:29; Marble, 1990:18). More and more complex questions can be asked as experience and added capabilities improve. But while the impact on archaeology may be strongly felt within the discipline, the repercussions may spread to other areas, as well. As an example, GIS was extensively used in reconstructing ancient monuments in Japan. The reconstruction, which included ancient Keyhole tombs, brought about a new impact on Japanese social and scientific fields, including cultural resource conservation, museums, education, TV, and archaeological research. Through pseudo-3D graphics, entire ancient villages, including the mysterious Keyhole tombs, can be reconstructed at a fraction of the cost of physical reconstruction of the monuments. The village reconstructions are of extensive usefulness in education, public awareness, and museum work because of their portability and the various methods of display available (Ozawa, 1993). Even the city of Pompeii, which was originally excavated in 1748 and has been almost continuously excavated ever since, has benefited from the use of a GIS. In order to document over 8,000 frescoes, mosaics, and other features that are slowly decaying from weather and visitors to the ancient city, the Archaeological Superintendent for Pompeii began the Neapolis Consortium, a cooperative high-technology venture between IBM and Fiat. The primary goal of the program is to catalog the many artifacts excavated and to link images and detailed descriptions to the specific point from which they were excavated (Bruschini, 1990:27). Once cataloged, the GIS database will allow artifacts to be "cross-analyzed to determine relationships between lifestyle and distribution of wealth" (Bruschini, 1990:29). Only through the use of GIS could so many variables be correlated. Presentation of Materials A very important aspect of GIS use in archaeology cannot be overlooked: presentation of materials. It is an extraordinarily important part of the archaeological process is the publication and presentation of the data resulting from an investigation or excavation. Without the dissemination of information, the data might just as well not exist. Archaeologists working in the CRM field, and archaeologists working for primarily non-archaeological agencies, such as the National Park Service, will frequently find themselves presenting the results of archaeological research to laymen. Archaeologists in academia can also benefit from the presentation capabilities of a GIS at conferences, in lectures, and when coordinating with other archaeologists or staff on projects. As noted by Kenneth Kvamme (1995), "a well-designed graphic can…more effectively communicate results" (7). The ability to create pseudo-3D images, color-coded maps, maps at almost any scale necessary, as well as the ability to create layouts specifically for presentation purposes makes the GIS the ideal tool for the presentation of archaeological material (Bruschini 1990:28-9; Kvamme 1995:6-7). Barriers to Using GIS in Archaeology Although many possibilities exist for the use of GIS systems in archaeology, many barriers to the effective use must be overcome before the true value of GIS can be realized. Stephan Shannon has warned that "GIS has reduced archaeological research and problem-solving to the making of ‘pretty pictures’" (1993). Data Standards One important impediment to the effective use of GIS in archaeology is the issue of data standards. According to Kvamme (1990), GIS development is market-driven, since it was designed as business software, and archaeology is but a small part of that GIS market (113-4). As a result, accurate data is simply not available in many areas that are important to archaeologists. Hitherto, the concentration has been on large urban areas where companies that utilize demographic data center their interests. Moreover, in areas that are of interest to archaeology, many small topographic features are smoothed over, inaccurate, or simply missing. It is these features, argues Kvamme (1995), that would have probably been the most important to human settlement decisions (5). With these features missing, predictive modeling loses much of its effectiveness. Gruel et al (1993) argue that the only real solution is to design a system from the ground up, specifically for archaeological applications (81). With that in mind, they designed the Arkéoplan project, which first examined the needs of a system, then designed a system based on the technology available at the time of writing. Standardization of Archaeological Terminology Another barrier to the use of a GIS in archaeology is the necessary standardization of terminology within the field of archaeology if any meaningful exchange of data is to occur. Although it has no direct connection with GIS, it will have enormous impact on its implementation. Even the method by which locations are recorded is not standardized, though most GIS systems can convert relatively easily. Still, Arroyo-Bishop and Zarzosa (1995) recommend UTM as the standard grid system for the field of archaeology because it eliminates any ambiguity in the coordinates of a site (45-6). Other standardizations include the definition of a site, agreement on terminology for the description of artifacts, and many others that, while not directly affecting the digitization of archaeological catalogs, will seriously impede data exchange and the more global outlook that such a free exchange of data would permit. Theoretical Concerns While not, strictly speaking, a barrier to GIS use, the issue of the technology driving the research is an important one. There is an old saying, "when the only tool you have is a hammer, everything starts to look like a nail." Allen, Green, and Zubrow (1990) argue that it is probably the most dangerous pitfall in using a GIS. "Good research," they argue, "is asking good questions, which a GIS cannot do" (383). Referring to the deterministic nature of a tool such as GIS, the term environmental determinism was applied to the fact that research utilizing a GIS tended to produce results that pointed to the environment as the most important factor in social organization, cultural change, and other phenomenon being investigated. Historically, the reaction against environmental determinism first appeared in the literature of archaeology when Julian Steward advanced his theory of cultural ecology in the 1930s. Steward’s theory was that the primary mechanism for social change was constraints placed on a society by their environment (1977). Since a GIS is primarily an instrument for evaluating the physical environment, Llobera (1996) argues that environmental determinism is primarily a result of the adoption of systems in spatial models from geography and applied to archaeology. Llobera states that the models are oriented around the study of the institution, groups, and systems, rather than on the individual, which ought to take priority (613). It is precisely this type of error, however, that landscape archaeology seeks to avoid. Conclusions The advantages of using GIS systems in archaeology are certainly more than have been mentioned in this paper, and there are many more obstacles that must be overcome. The advantages of cataloging artifact collections in an electronic database without the added benefit of being spatially-referenced are enormous -- particularly in the realm of data exchange -- are almost reason enough to adopt the technology. But the added ability to map archaeological phenomenon, and the incredible advantages offered by the ability to recreate entire landscapes have opened the door to entirely new approaches to viewing the past. Landscape archaeology promises to provide a theoretical basis for research that utilizes the enormous power of a GIS that would help to eliminate bias created by artificially imposing site boundaries that are almost certainly completely arbitrary (Savage, 1990:24). As we proceed further with GIS applications in archaeology, we can only hope that our understanding will be able to keep pace with our ability to collect information. Ultimately it is the interpretation of the data that will determine whether a re-construction of the past is accurate or not, and not the technology used to support it. As long as wisdom and prudence guides the archaeologist in those interpretations, GIS can be an enormous asset in understanding the vast amount of data available.
References Cited Allen, Kathleen M. S., Stanton W. Green, and Ezra B. W. Zubrow Arroyo-Bishop, D. and M. T. Lantada Zarzosa Bruschini, Stefano Crumley, Carole L., and William H. Marquardt Gaffney, V. and M. van Leusen Gruel, Katherine, Olivier Buchenshultz, Jean-Françios Alliot,
and Hervé Murgalé Hansen, Henrick Jarl Kvamme, Kenneth L. 1993 Spatial statistics and GIS: an integrated approach. In Computing the Past: Computer Applications and Quantitative Methods in Archaeology, edited by J. Andresen, T. Madsen, and I. Scollar, pp. 91-104. Aarhus University Press, Aarhus. Llobera, Marcos Madry, Scott L. H., and Carole L. Crumley Marble, Duane F. Ozawa, Kazumasa Savage, Stephen H. Shannon, Stephan Steward, Julian van Leusen, P. Martijn Wheatley, D.
|
