ABSTRACT
For over a century, explorers, scholars, and the interested public have marveled at the engineering accomplishments of the ancient Maya. From the construction of monumental architecture to the excavation of vast irrigation systems and drainage canals, the Maya transformed their landscape for social reasons, to mirror ideological constructs, and to adapt it to maximize agricultural yields from their local environments. Studies that explore the relationship between commoners and the social elites who organized the construction and maintenance of such works have always been popular in the eld, and remain a central research theme among archaeologists (Chase 1992; Chase and Chase 1992; Lohse and Valdez 2004; Masson and Freidel 2002). This focus on the sociopolitics of the engineered environment diverts attention from other important material relations that supported such monumental endeavors. In Marxian terms, labor and the relations of production have been well studied while the means of production have received little attention. This bias has been exacerbated by the antiquated, although still somewhat pervasive, idea that the Maya lowlands is fairly homogeneous in the distribution of critical economic resources. Certainly, greater effort has been expended in illustrating regional differences in access to jade and salt than differences in access to chert. Non-obsidian lithic data traditionally have been an ancillary concern in lowland Maya research. Prior to Kidder’s (1947) work at Uaxactun, stone tools were regarded as discoveries of little value
and were seldom included within the main text of archaeological site reports, instead being relegated to report appendices (Chapter 1). Figure captions often provided the only description of these artifacts during the early period of Maya archaeological research. Tools were classied using broad functional terminology based on their general form with little regard for their archaeological context or technological attributes, and there were no use-wear studies conducted to test such functional assertions. Archaeological site reports and other manuscripts produced during the late nineteenth and early twentieth century, largely prior to the Carnegie Institute of Washington excavations at Uaxactun, reect the biases governing Mesoamerican research during that period in their exclusive attention to architectural form and style, settlement patterns, monumental sculpture, ceramics, and art objects (e.g., Maler 1908-1910; Maudslay 1889-1902; Morley 1937-1938; Totten 1926; Tozzer 1911). Other artifacts, including faunal remains, botanical remains, and stone tools, were mentioned infrequently in these early works, and substantive analysis was wholly absent. Signicant advances have been made in the study of Maya chipped-stone artifacts since these early decades of research. Nevertheless, there remains a pervasive under-appreciation of the applicability of non-obsidian tool assemblages for addressing complex questions of social history. Although archaeologists readily examine such data to detect activity areas, discuss producer-consumer relationships, and identify engagement in specic economic activities, they do not incorporate these data when addressing higher-order questions related to the dynamics of socio-political relationships. This is a puzzling oversight, because stone tools were fundamental to the techno-economic structure of Maya society, and they occur in near ubiquity throughout the lowlands (cf. Chapters 6 and 7). In contrast, obsidian data have been used extensively for understanding the development and maintenance of inequitable political economies, as well as for mapping spatial and temporal patterns of regional and long-distance interaction (Braswell 2002; Clark 2003; Clark and Lee 1983; Dreiss and Brown 1989; Ford 2004; Hay 1978; Santley 1983). Considering that for sites in the Maya lowlands, the volume of recovered tools crafted from chert is most often substantially greater than that made of obsidian, it is imprudent that the latter consistently receive far greater analytical attention (see Hirth 2003). To be sure, this imbalance has been inuenced as much by the greater ability to identify the source of obsidian tools as it has the generally accepted-and inaccurate-assumption that chert tools were crafted from ubiquitously available raw materials. If tool-quality stone was indeed abundant across the lowlands, we would need to ask why a craft industry, such as that observed at Colha, developed in the rst place. There would be little motivation or scal support for such large-scale product specialization without a market demand predicated on the disproportionate allocation of natural resources. Lithic Resources and the Ancient Maya
Lithic resources are distributed broadly across much of the lowlands. Nevertheless, material quantity and quality vary, and some areas are considerably richer in tool-quality stone than others (Black 1987; Black and Suhler 1986; Blom 1932; Hester and Shafer 1984, 1994). The natural availability of lithic resources in the lowlands is determined by the geological history of the substrate as well as the more recent erosional history that has exposed lithic deposits (Mosely and Terry 1980). Thus, the quantity of available resources may vary across lowland landscapes either in terms of the actual presence or the ultimate extractability of resources. Lowland landscapes can be categorized along a continuum of lithic resource availability. Some landscapes are best considered as decient zones with little or no quality resources, while others, such as the northern Belize chert-bearing zone, are accurately viewed as zones of considerable abundance. Where high-quality material is absent, lithic producers and consumers could have adapted to the use of poor-quality stone, imported tools from other regions, or employed other materials
(Chapter 6). Some zones of exceptionally poor availability, such as many coastal and island sites, adapted through importing stone tools from inland areas, using conch shells in place of stone in manufacturing tools, and being generally less ambitious in their architectural pursuits (Eaton 1974; Stemp 2001). The use of substitute resources was not always a viable alternative. There are few available alternatives for stone that can be crafted into tool forms. Tools of shell, bone, and antler have been recovered in excavations in some areas of the lowlands, and others were likely crafted from wood (Dacus 2005; Dreiss 1994; Guderjan and Garber 1995; Hester 1985; Tozzer 1941). Such tools, however, could not have been used for the full range of tasks performed in Maya society. Activities such as quarrying stone, cutting and shaping timber for architecture or manufacturing canoes, and various other resource-processing tasks required the use of stone tools. Stone must possess specic properties to be useful for tool manufacture (Kooyman 2000; Luedtke 1992). Siliceous materials with an amorphous compositional structure, such as chert, obsidian, and, to a lesser degree, chalcedony, were preferentially chosen for aked tool manufacture. Quartzite, dolomite, petried wood, and silicied limestone also were occasionally used. Highquality materials possessed a cryptocrystalline structure with few fossil or mineral inclusions, and appear in nodules large enough to be viable for the production of standardized tool forms. Several studies have reported nding heat-treated material in Maya lithic assemblages (Barrett 2004; Brian 2003; Meadows 1999; Shafer 1983), suggesting that thermal alteration was practiced to increase the workability of poorer-quality stone. Although this process would have converted some marginalquality stone into material viable for tool production, the overall durability of such tools would have decreased (Luedtke 1992; Whittaker 1994). There is no evidence that tool producers at Colha, Belize heat-treated their material, undoubtedly because of the exceptional quality of locally available chert. The mere presence of chert boulders in architectural ll and in-eld chich piles has been viewed as evidence for resource abundance, as have chert gravel deposits and production-waste accumulations (Kunen 2001; Kunen and Hughbanks 2003). Nevertheless, because tool-quality resources must possess a particular set of characteristics, including having a ne-grained, homogeneous matrix uncompromised by fossiliferous or crystalline inclusions, chert boulders chosen for use in architecture may have been unsuitable for craft production. Likewise, gravels are not large enough to allow for the manufacture of axes, knives, or large projectile points. Finally, the presence of production waste, as an indication of resource exploitation, must be observed as having a negative correlation with material availability. One of the most signicant oversights in lowland lithic research has been a failure to realize that stone represents a nonrenewable resource. The need for tool-quality stone increased as Maya populations grew, and there were technological limits to their ability to extract chert. For these reasons, the eventual depletion of this resource was inevitable. Thus, when we assess the availability of lithic materials, we must not only consider the natural proliferation of exploitable resources, but also understand how local, regional, and interregional demand changed over time, and how these affected a limited resource. The Archaeology of Northwestern Belize
The sites in northwestern Belize discussed in this paper-Blue Creek, Bajo Vista, Bedrock, La Milpa, Nojol Nah, and Sotohob-occupy different positions along the continuum of resource availability (Figure 5.1a). Ancient stone-tool production has been documented at several of these sites, and took place in the context of attached and nucleated workshops (Costin 1991; Lewis 2003). The diversity in natural resource potential and economic organization provides an opportunity to evaluate the responsiveness and adaptability of local political ecology.
