Modeling and testing polity boundaries in the Classic Tuxtla Mountains, Southern Veracruz, Mexico

Abstract

Combined application of GIS modeling of polity boundaries and qualitative characterization of those boundaries in the Tuxtla Mountains, southern Veracruz, Mexico generated interesting questions and some answers regarding the regional political landscape. Not only did this process elucidate possible boundary locations, the process of evaluating these different hypotheses proved to be a useful exercise in characterizing the boundary and the nature of interaction between the polities controlled by Matacapan and Totocapan during the Classic period. I develop a seven variable classification for political boundaries to be used with coarse-grained settlement data.

Introduction

Political economic models previously presented for the Classic (Table 1) Tuxtla region of Mesoamerica have been based largely on archaeological excavation and survey at Matacapan and settlements within the Catemaco Valley (Arnold et al., 1993, Pool, 1990, Pool and Santley, 1992, Santley, 1994, Santley, 2007, Santley et al., 1989, Stoner et al., 2008) (Fig. 1). Missing from these reconstructions is the nearby regional center of Totocapan¹ within the neighboring Tepango Valley (Ortiz Ceballos, 1975, Valenzuela, 1945b). In 2007, I conducted the Tepango Valley Archaeological Survey (TVAS), which included the valleys for the upper Tepango River and almost the entire Xoteapan River. This survey focused on Totocapan and its southern hinterland for the purpose of comparing the prehispanic cultural evolution of the Tepango and Catemaco valleys. Intervalley comparisons suggest that Totocapan and Matacapan were the capitals of distinct polities. In many ways, the two capitals were very similar, however, they also display several important contrasts.

During the early stages of this analysis, it became clear that any comparison of the polities headed by Matacapan and Totocapan would require some estimation of the political boundary between them. Characterization of the boundary between these political entities can help to describe the nature of interaction (or non-interaction) between them. Additionally, political boundaries serve an analytical function by dividing the study region into political units that can be entered into various statistical analyses or models.

Polity boundaries are often passively assumed by archaeologists to fall on physiographic features, such as rivers or mountains between watersheds. Other methods, like Theissen polygon analysis, divide a region into roughly equal parcels of land at different ranks within the settlement hierarchy (e.g., Blanton, 1996, Christaller, 1966; Johnson, 1972). More recent attempts to estimate polity boundaries make use of predictive models using GIS software or other quantitative models (e.g., Hare, 2004, Kintigh and Ammerman, 1982, Renfrew and Level, 1979, Ruggles and Church, 1996). Some archaeologists simply characterize political organization within their respective survey boundaries and ignore substantive boundary reconstruction. Such an approach assumes that the survey boundary itself ‘fills-in’ as a political boundary until additional data are collected on a broader scale. All of these approaches rely on several assumptions about how space is politicized that may not be valid in all cases (see Smith, 2004). As with so many other archaeological problems, drawing upon several lines of evidence is the best approach to the characterization of polity boundaries in ancient states.

I rely on a combination of different settlement data to characterize the boundary between Matacapan and Totocapan polities. I employ a simple geo-spatial model to allocate ‘territories’ to small and large centers in the west-central Tuxtlas. This model uses the relative size of centers to weight political power and physiographic terrain is considered to have some effect on the size and shape of their territories. It must be understood that geospatial models must never be employed irrespective of the archaeological data that support or refute them. The archaeologist must evaluate them using more traditional data from survey and excavation. Qualitative evaluation of the geo-spatial model aims to characterize boundaries between polities according to seven variables.