Situating Power and Locating Knowledge: A Paleoethnobotanical Perspective on Late Classic Maya Gender and Social Relations



Viewing household production in terms of a political economic balance of “give and take” circumvents difficulties related to gender attribution in archaeology and challenges timeless gender stereotypes. This chapter proposes such an archaeological approach to gender by examining the charcoal assemblages from two Late Classic period Maya archaeological sites in the upper Belize Valley of western Belize. These sites occupied distinct positions within a complex political economic landscape, and their charcoal assemblages reflect heterogeneity in household production. The type and the intensity of activities, including wood procurement and craft production, were socially contingent. We propose that household activities and forms of knowledge were conditioned by the positions of households within broader political economic landscapes, not conforming to the timeless social stereotypes imposed by archaeologists.

Confronting the assumed universality of a gendered division of labor has become one of the central goals of an engendered archaeology (Conkey and Gero 1991; Conkey and Spector 1984; Conkey and Tringham 1997; Pyburn 2004; Robin 2002a, 2006). Often positioned within the functional arguments of a social evolutionary framework, this assumption maintains that certain activities are universally and more efficiently associated with either males or females and that evolution affects gender in ways that are transhistorically uniform and, hence, predictable according to a particular evolutionary stage (i.e., Klein 2004). Yet, following Conkey and Spector, “[a] division of labor between males and females should not be assumed but rather be considered a problem or a feature of social structure to be explained” (Conkey and Spector 1984:19).

Confounding the issue of a gendered division of labor, many archaeological statements about gender are based on ethnographic information, positioned within either an evolutionary or a particularistic explanatory framework. Drawing on ethnographic descriptions of the activities of men and women, archaeologists then attribute particular classes of artifacts or archaeological features as within the domain of men or of women. On one level, such an approach can reproduce the often androcentric biases inherent in many ethnographies that excluded both women and gender from ethnographic research design (Conkey and Spector 1984; Stahl 1993). On a more fundamental level, the uncritical and direct application of ethnographic analogies to the archaeological record potentially dehistoricizes social configurations in both the past and the present. When gender is presented as historically homogeneous, archaeologists promote a static vision of social relations, unintentionally reinforcing the conclusion that understanding change and variability is not a worthwhile endeavor.

Identifying these uncertainties and biases has led several archaeologists to question the entire nature of archaeological research, from design to analysis to interpretation, as providing both a technology and a discourse of power (sensu Foucault 1980). While recognizing the significance of this critique, the perspective taken here is based on the position that there are historical and material realities that constrain the range of archaeological interpretation (Wylie 1992) and that the problem with gender bias in archaeology lies not with data but, rather, with the questions that archaeologists ask and the modes of inference that they then construct (Brumfiel 1992, 1996:459).

This chapter examines social relations by viewing household production in terms of a political economic balance of “give and take.” We present an alternative to uncritical, ahistorical gender attribution in archaeology that challenges timeless stereotypes of a universal gendered division of labor. Productive activities are situated between broader political economic relationships and the internal structure of households and communities. Within this matrix, individuals must balance the daily rounds necessary for social reproduction. Productive activities can be seen as connected to a political economy that differentiates people's activities as well as the scale of their social networks, creating a heterogeneous landscape of knowledge and practice (Morehart 2006). The practices and systems of knowledge of household members vary not according to timeless gender roles but according to the material reality of this nested landscape. We consider this model by examining wood exploitation practices, especially the use of firewood, by the Late Classic Maya. First, we examine how the ethnographic record has been used to create static pictures of gender, particularly how accounts of household activities and cosmology have been used to support a gendered division of labor. Analogy often is employed in a contradictory manner that subverts variability and distorts historical realities. Our archaeological perspective is based on the charcoal assemblages from two sites in western Belize, Pook's Hill and Chan Nòohol, which occupied distinct positions within the Late Classic milieu. By analyzing the distribution of wood charcoal between these sites and articulating these data with other artifact assemblages, particularly weaving tools, we attempt to situate household production within a historically flexible political economic landscape and, in the process, challenge static social stereotypes projected onto the ancient Maya.

Paleoethnobotany, Ethnographic Analogy, and Gender

Little has been written on firewood and social relations, particularly gender, in Maya prehistory. Maya archaeology's focus on “high culture” rarely includes plants. The belief that organic remains do not preserve in the tropics has created a gap filled in with observations from disparate ethnographic sources. Such an approach perpetuates static models that obscure historically contingent variability in the past and the present (Robin 2002a).

The lack of concrete research into Maya wood utilization is surprising. Wood charcoal is one of the most abundant archaeobotanical remains (Hastorf and Johannessen 1991; Johannessen and Hastorf 1990; Smart and Hoffman 1988). People burned a lot of wood in the past, every day, and for many different reasons. The need for domestic firewood has structured daily routines in many 20th-century Maya communities in the Highlands and Lowlands, and household hearths commonly are kept lit throughout the day and night (Redfield and Villa Rojas 1934; Vogt 1969; Wisdom 1940). The exploitation of wood and tree resources is also central to models of ancient Maya environmental impact and societal collapse (Abrams and Rue 1988; Abrams et al. 1996; Culbert 1988; Hansen et al. 2002; Paine and Freter 1996; Rice 1993). Abrams and others have argued that overexploitation of hardwood and pine populations in the Copan Valley of western Honduras led to ecological degradation and, eventually, political fragmentation (Abrams and Rue 1988; Abrams et al. 1996). They calculated that, regardless of the consumption rates employed for analysis, the demand for domestic fuelwood exceeded other wood demands by several hundred times. Although based on an experimental approach, this research emphasizes the importance of firewood in household economies and stresses that the consumption of firewood was connected systemically to macro-level sociopolitical processes.

Archaeologists have confronted the paucity of archaeological data on Maya wood utilization by drawing on ethnography. Relying on ethnographic literature, scholars have proposed women and girls collected firewood. For example, Sharer's Daily Life in Maya Civilization states, “We can assume that in the past, as today, girls were trained to take on the traditional roles associated with wife and mother in the Maya family … they undoubtedly played an essential part in subsistence by collecting firewood, wild foods, and condiments” (Sharer 1996:120, emphases added). Yet variability in gender roles within and between commonly cited ethnographic texts rarely is considered. Redfield and Villa Rojas's (1934:66, 71) ethnography of the Yukatek Maya, a mainstay in Maya analogy, records that women collected firewood when young boys aged six to seven were unavailable to perform this task. Wisdom's ethnography of the Ch'orti' states, “The men usually collect it individually for their own kitchens” (Wisdom 1940:85–86, emphasis added). During the late 20th century, many Lakantun (a.k.a. Lakandon) Maya women incorporated the collection of firewood into their many daily tasks (McGee 2002:56), but they often relied on their unmarried male children to collect firewood (Boremanse 1998:40). Furthermore, Lakantun firewood-collecting often is a joint, family task performed in tandem with agricultural activities and on the walk back from the field. Although these accounts associate firewood collecting with particular groups, the high degree of variability is clear.

Scholars likewise have employed cosmology to reinforce these more economic assessments of a gendered division of labor. A binary division of the world into a domesticated, ordered, social space versus an undomesticated, wild, dangerous space constitutes a cosmological model discussed in several contemporary Maya ethnographies. This model also appears in ethnohistoric texts, and it seems to have correlates observable in the iconographic corpus of the Classic and the Postclassic Maya. Archaeological applications of this cosmology often have confined past Maya women to the domesticated realm, arguing the undomesticated, dangerous realm could only be navigated by men (e.g., Stone 1995; Taube 2003). Although this claim is parallel to assumptions that segregate men's and women's worlds into public versus private spaces, this conclusion evades a more fundamental and historical contradiction relevant to this chapter.

As the Spanish began to reorganize the social, political, and economic worlds of the Maya following the conquest, they began to record data on Maya landholdings and tenure (Restall 1997). Employing a European model of agricultural practices, they separated Maya land into that of field, known in Yukatek as kool (“milpa” or “sementera”), and that of forest, k'áax (“bosque” or “selva”) (Bastarrachea et al. 1992; McAnany 1995). Their basic assumption was that land not in agricultural production was unused forest. Although classifying k'áax into unused land certainly had implications for restructuring and appropriating indigenous landholdings, their assumption was fundamentally flawed. This is a clear example of the often vast misunderstandings of colonial officials—the “uneasy relationship between knowledge and power” (Dirks 1992:176). Rather than unused, unproductive land, k'áax was a highly productive source of economic resources (McAnany 1995). Better viewed as either managed forests or managed fallows (Gómez-Pompa and Kaus 1990; Peters 2000), k'áax was one of the principal locations for hunting, gathering medicinal plants, collecting fruits from economically important trees, and collecting firewood.

In other words, the wild and undomesticated space modeled in cosmology would seem to have been also a useful space. But if this space was the realm of men, how were women collecting the firewood? On one level, this apparent contradiction is the product of equating cosmological and ideological statements to the lived reality of social actors (Morehart 2006). On another level, the contradiction suggests an interpretive fuzziness—a chaos of disarticulated inferences and analogies that are not in dialogue with one another but rather are grounded in the assumption of universal gender roles.

Most approaches to analogy are markedly “continuity-centric” in that they seek to make explicit conditions of enduring continuity between ethnographic and archaeological cases. The “direct historical approach” (Gould and Watson 1982), common in Maya archaeology, maintains that we can compare the ancient Maya to the modern Maya because they are “Maya.”1 Archaeological and ethnographic cases become homogenized in terms of one another, and archaeological narratives are linked by a concern with the “traditional” over the “modern” (Stahl 1993:243). When stemming from an unquestioned acceptance of the assumedly invariant ethnographic record, such “continuity-centric” approaches lack the historical imagination necessary to adequately explain or to contextualize variability. The tyrannical legacy (Conkey and Spector 1984:13; Robin 2002a:13; cf. Wobst 1978) of this approach has influenced reconstructions of almost all aspects of Maya society, including agricultural systems, iconography, ritual, politics, and gender.

One can examine particular configurations past and present without making a blanket claim of continuity. Furthermore, archaeologists cannot ignore certain aspects of the ethnographic record that they feel are “modern” and, consequently, irrelevant; as Talal Asad observes, “much of what appears ancient … is itself recently invented” (Asad 1991:316). The ethnographic record is not a stagnant data set to be drawn upon selectively. Archaeologists can view the ethnographic record critically without disposing of its potential for elucidating the past. A critical utilization of the ethnographic record is one that frames ethnographic and archaeological comparisons within historically situated political and economic contexts. Such an approach flips commonly held notions of continuity on their heads by articulating the past with the present as an ongoing and unfolding historical process. The present is continuous with the past not through a compartmentalized vision of the survival of “traditional” cultural forms but, instead, as a real, material consequence of History.

Case Examples: Pook's Hill and Chan Nòohol

This study contrasts wood use by people who occupied different socioeconomic and political positions in the Maya area. Two major factors influenced the selection of sites for this study. The most pragmatic consideration involved the availability of data. Collecting archaeobotanical data still is not a standard component of excavation strategies at Maya sites, though research has shown that systematic archaeobotanical sampling can yield considerable social, economic, and ecological information (e.g., Lentz 1991, 1999; McKillop 1994; Miksicek 1991; Morehart 2002, 2003; Morehart, Wyatt, and Lentz 2004). The second consideration focused on selecting settlements that represented distinctly different types of habitations in scale and size. The sites of Pook's Hill and Chan Nòohol met these requirements. First, researchers at both sites systematically collected archaeobotanical remains during excavations. Second, Pook's Hill is an affluent plazuela group, while Chan Nòohol is a cluster of seven commoner farmsteads. Moreover, both sites are located in the greater upper Belize River watershed of western Belize, providing a somewhat similar environmental context (Helmke 2001, 2003, 2006a, 2006b; Helmke et al. 2003; Robin 1999, 2002a, 2002b, 2004) (Figure 5.1).

Figure 5.1.

Map of the upper Belize Valley showing sites discussed in the text.

Pook's Hill is a medium-sized plazuela group located in the karstic foothills that form the western perimeter of the Roaring Creek Valley, overlooking a fertile alluvial valley below (Figures 5.1). Pook's Hill lies 4.7 kilometers north of the major center Cahal Uitz Na and one kilometer northwest of the minor center of Chaac Mool Ha. Cahal Uitz Na appears to have served as the capital of the local polity during the Classic period (Awe et al. 1998). Chaac Mool Ha was the principal northern satellite to Cahal Uitz Na, similar to other satellites in the greater Belize Valley (Driver and Garber 2004). Pook's Hill likely was incorporated into the Cahal Uitz Na polity through Chaac Mool Ha, possibly via tributary networks.

Investigations at Pook's Hill have focused on the site's terminal occupation, which dates to the Late to Terminal Classic period (ca. 830–950 c.e.) (Helmke 2001, 2003, 2006a, 2006b; Helmke et al. 2003; Morehart 2001). Primary contexts date back to at least the Middle Classic (ca. 550 c.e.). On the other hand, the earliest ceramic specimens from secondary contexts can be attributed to the Late Formative (ca. 300–100 b.c.e.), while the latest belong to Early Postclassic complexes (later than ca. 950 c.e.). The site is comprised of the remains of nine masonry building platforms enclosing a central plaza (Figure 5.2). The architectural group as a whole encompasses an area of approximately 1,106 square meters of which the plaza occupies just over 430 square meters. The largest building measures 16.5 meters long and is over 2.5 meters high above the plaza surface. The smallest structure measures 7.1 meters long and is approximately 0.8 meters high.

Figure 5.2.

Plan map of the plazuela group, Pook's Hill.

Most of Pook's Hill's structures are rectangular in plan and can be categorized as “range structures” (long, linear, masonry structures delimiting an open patio or plaza). A dome-vaulted sweatbath or pib'naah has been identified at the northwest corner of the plazuela (Helmke and Awe 2005), and an ancestor shrine structure is located on the eastern side (Helmke 2003). The group conforms to the so-called Plaza Plan 2 configuration defined at Tikal by Becker (1971, 1999:139–147) based on the presence of the ancestor shrine defining the eastern perimeter. As is common, the largest concentration of funerary and votive deposits at the site was found in association with this eastern shrine. Evidence of ritual feasting was associated with the western range building, which Helmke (2001) has interpreted as a possible part-time “feasting hall.” The site's residents also maintained access to minor quantities of exotic goods (i.e., jadeite, marine shell, obsidian, and pyrite). Pook's Hill's relative affluence may indicate strategic flexibility based on the mobilization of social, political, and economic ties.

Chan Nòohol is located along the limestone uplands in the interfluvial zone between the Mopan and Macal rivers, the two principal branches of the Belize River (Figure 5.1). The site contrasts with Pook's Hill in several ways (Figure 5.3). Whereas Pook's Hill consists of a single affluent plazuela group, perhaps a community node itself, Chan Nòohol is a cluster of seven commoner farmsteads and was part of a larger farming community, Chan, during the Late Classic period (Ashmore et al. 2004; Robin 1999, 2002a, 2002b, 2004). The site is four kilometers southeast of the civic center of Xunantunich, which was the polity capital of the area during the Late Classic period. Although the farming community of Chan has an occupational history that spans from the Middle Preclassic to Early Postclassic periods (ca. 1000/800 b.c.e.–1150/1200 c.e.), the Chan Nòohol farmsteads included in this study date to the late facet of the Late Classic period (ca. 660–780 c.e.). Moreover, the establishment of the Chan Nòohol households corresponds to the Chan community's maximum settlement expansion and to the community's incorporation into the Xunantunich polity.

Figure 5.3.

Plan map of Chan Nòohol. Redrawn from Robin 1999:Figure 17.

The spatial organization of Chan Nòohol's architecture is notably distinct from that of Pook's Hill (Figure 5.3). Chan Nòohol's seven farmsteads (CN1–CN7 in Figure 5.3) were spaced between 50 and 100 meters apart. Each farmstead was composed of one or two main residential structures and in many cases ancillary structures surrounded by work areas and agricultural terraces. In contrast to Pook's Hill, Chan Nòohol's structures were constructed as small pole-and-thatch buildings on low rubble platforms between 0.1 and 1.1 meters high. The interior of most residences was taken up by a low bench for seating or sleeping. The labor necessary to build these structures easily could have been accommodated by an individual family. Most daily activities occurred either in outdoor spaces or in the ancillary structures. Chan Nòohol, thus, is marked by an open organization of space, which may have shaped social behavior by promoting more complementary and egalitarian interactions among residents (Robin 2002b).

Artifactual data from Chan Nòohol's farmsteads suggest a focus on productive practices geared toward local consumption and household reproduction (Robin 1999, 2002a, 2002b, 2004, 2006). Members of each farmstead undertook common domestic and agricultural tasks. With the exception of prismatic obsidian blades, granite grinding stones, slate artifacts, and two greenstone artifacts recovered from two households (CN5 and CN7), the majority of artifacts were made of resources available locally. High numbers of artifacts and activity areas associated with agricultural production and food processing suggest these were the principal economic activities at Chan Nòohol. Given the lack of evidence for other forms of specialization, residents at Chan Nòohol seem to have been tied into the Xunantunich political economy through their labor and agricultural production. The relatively short occupational depth of the farmsteads, their very modest size and composition, and their localized economic orientation suggest the farmers had limited access to certain resources and sociopolitical ties. Nevertheless, despite the site's lower economic standing, archaeological research has uncovered evidence of community-based rituals and feasting (Robin 1999:264,2002b:259–260).

Paleoethnobotanical investigations were undertaken at both Pook's Hill and Chan Nòohol in order to better understand human–plant interactions at the sites (Lentz et al. 2005; Morehart 2001; Robin 1999). A total of 69 archaeobotanical samples have been analyzed from Pook's Hill to date (Morehart 2001). Samples were collected for water-assisted flotation from several contexts, including middens, burials, caches, floor deposits, and architectural fill and collapse. Archaeobotanical remains from Pook's Hill indicate residents relied on multiple domesticated plants, such as maize (Zea mays), squash (Cucurbita sp.), and chile peppers (Capsicum annuum). The remains of fruits from economically useful trees, such as hog plum (Spondias sp.), calabash (Crescentia cujete), coyol palm (Acrocomia aculeata), avocado (Persea sp.), and chico sapote (Manilkara sp.), may suggest that the site's inhabitants maintained orchards (Morehart 2001).

Paleoethnobotanical efforts at Chan Nòohol focused on collecting macrofloral remains and soil samples for flotation during the course of excavations (Robin 1999:546). Seventy-one samples were taken predominantly from household refuse deposits and from agricultural terraces. Like those from Pook's Hill, archaeobotanical samples from Chan Nòohol contained a variety of remains from domesticated cultigens and fruit trees (Robin 1999:296–305). The distribution of maize at Chan Nòohol provides tentative evidence that farmers processed maize in their fields (Robin 1999:296, 2006); remains of maize glumes were found in terrace excavations, but none were located in domestic refuse deposits. Similarly, no maize glumes were found at Pook's Hill (Morehart 2001), which may suggest an initial stage of maize processing occurred prior to transport, particularly the removal of kernels from the cob.

Despite these similarities, the distribution of wood charcoal between the sites differs. Based on charcoal from flotation samples, the residents of Pook's Hill were burning pine (Pinus sp.) almost exclusively (Figure 5.4). The residents of Chan Nòohol, in contrast, were burning mostly hardwoods and little pine. Pine can be distinguished from hardwoods by the lack of vessel elements (cellular structures that conduct water and solutes) and the presence of resin ducts in the wood. Moreover, pine can be identified easily from sites in the Maya Lowlands as it is the only gymnosperm indigenous to the area (see Balick et al. 2000). Eighty-eight percent of all samples from Pook's Hill contained pine charcoal, whereas 25 percent contained hardwood taxa and three percent contained charred palm (Figure 5.5a). This disparity is even more observable when one compares charcoal weights, though weight measurements may be less reliable due to differential preservation between samples. Ninety-six percent of Pook's Hill's total charcoal weight is pine, versus four percent of hardwood and 0.2 percent of palm (Figure 5.5b). Samples from burials, caches, middens, floors, and architectural collapse all contained substantially more pine than hardwood charcoal. This widespread contextual distribution of pine at Pook's Hill suggests that pine was being used for a variety of tasks, from construction activities to firewood to ritual.

Figure 5.4.

Scanning electron micrograph of pine (Pinus sp.) charcoal from Pook's Hill in cross-section.

Figure 5.5.

Distribution of hardwood and pine charcoal at Pook's Hill by ubiquity (a) and weight (b).

Almost the exact opposite distribution of wood charcoal characterizes Chan Nòohol (Figure 5.6a, b). Ninety-four percent of archaeobotanical samples from Chan Nòohol contained hardwood charcoal. Only 2.8 percent of Chan Nòohol's samples yielded pine. Weight measurements of wood charcoal from Chan Nòohol reinforce this pattern. Less than one percent (>0.01 grams) of Chan Nòohol's total charcoal weight is pine, whereas almost 100 percent is from hardwoods. Clearly, the farmers at Chan Nòohol were burning very little pine.

Figure 5.6.

Distribution of hardwood and pine charcoal at Chan Nòohol by ubiquity (a) and weight (b).

The distinction in wood charcoal between the two sites may be related partially to the environment. Two species of pine exist in the Maya Lowlands and in Belize specifically (Balick et al. 2000; Farjon and Styles 1997; Mirov 1967; Morehart et al. 2005; Standley and Steyermark 1958:46–51): Pinus caribaea Morelet and P. oocarpa Schiede. P. caribaea is by far the more common and characterizes the swampy savanna lands of northern, central, and southern Belize and certain areas of Petén, Guatemala, and the Mountain Pine Ridge of western Belize. P. oocarpa principally inhabits higher elevations, including Mesoamerican Highland regions, with a less common distribution in Belize. The Mountain Pine Ridge, located along the northern fringe of the Maya Mountains and north of both Pook's Hill and Chan Nòohol, has populations of both P. caribaea and P. oocarpa.2 Nevertheless, neither species inhabits the semideciduous, subtropical forests in the immediate environment surrounding Pook's Hill and Chan Nòohol; both can be classified as nonlocal resources.

Access to pine was not simply an ecological concern but likely was mediated by the political economy (Lentz et al. 2005; Morehart 2002:261–264; see Hastorf and Johannessen 1991 for a similar Andean example). Pine was a valued item throughout the Lowlands, serving both utilitarian and ritual purposes, though it is locally scarce in many regions (Lentz 1999; Lentz et al. 2005; Morehart 2002; Morehart et al. 2005; Thompson 1970:146). In the Roaring Creek Valley, the political center Cahal Uitz Na may have had developed centralized control over pine and other resources from the Maya Mountains (e.g., slate and granite). No archaeobotanical data are available from Cahal Uitz Na itself. But samples from Actun Nak Beh, a cave connected to Cahal Uitz Na by a causeway, overwhelmingly contained pine (Morehart 2002; Morehart et al. 2005). In this context, the abundant pine charcoal from Pook's Hill may indicate political and economic relationships between the two sites. In contrast, the residents of Chan Nòohol, a site with fewer political ties, likely had less access to pine and instead used a range of locally available trees for their wood needs. That some households procured their own wood supplies from the local environment (i.e., Chan Nòohol) while other households obtained it through trade, gifts, or tribute (i.e., Pook's Hill) would have affected the organization of household labor. The residents of Chan Nòohol had to integrate the collection of wood for their daily needs as part of recurring daily practices, and wood collection potentially involved the entire family—men, women, boys, girls.

The impact of collecting wood on the allocation of household labor is of particular interest when one examines the evidence for other economic activities. Consider an activity long stereotyped as essential to Maya womanhood: weaving. Spindle whorls were recovered at both Pook's Hill and Chan Nòohol, but as in the case of wood charcoal, the sites differ (Figure 5.7). At Chan Nòohol a total of eight spindle whorls were found among the site's seven farmsteads (Robin 1999:272). This low number in relation to households suggests that weavers were producing cloth for local consumption only. The less time spent on activities such as weaving enabled productive diversification. At Pook's Hill eight limestone spindle whorls, one modeled and incised ceramic spindle whorl, 20 perforated ceramic disks (both complete and fragmentary) that may have been used to spin cotton, a charred fragment of a bone sewing needle, and the remains of at least two additional bone weaving needles were found (see Stanchly 2006). These weaving needles are similar to other inscribed needles found at Maya sites (attested epigraphically as puutz' b'aak,“needle bones”), often bearing the names and titles of their original owners. Such inscribed needles are known from Altun Ha, Tikal, and Naranjo, where they record the royal names of men and women of the court (see Houston and Stuart 2001:64–65; Trik 1963:10–18). Since needles have been found inscribed with both male and female names, one cannot make an immediate leap to gender attribution.

Figure 5.7.

Sample of spindle whorls from Pook's Hill (a–o; drawn by Helmke) and Chan Nòohol (p–r; drawn by Robin 1999): (a–h) limestone spindle whorls; (i–l) ceramic spindle whorls and reused, centrally perforated sherds; (m) slate spindle whorl preform; (n–o) slate pebble spindle whorls; (p) ceramic spindle whorl; (q–r) limestone spindle whorls.

Pook's Hill's weavers possibly were producing cloth beyond the household level to participate in the political economy via tribute or gifting (Beaudry-Corbett and McCafferty 2002; Brumfiel 1991; Hendon 1997; Inomata 2001) or the production of ritual offerings (Morehart, Awe, et al. 2004). In contrast to Chan Nòohol, individuals at Pook's Hill were able to reinvest their time into weaving rather than collecting firewood or engaging in other agroforestry practices. The presence of such a diversity of weaving implements may indicate technological diversification in this elite craft (Brumfiel 2006).

Conclusions and Future Implications

Although no simple solution, thinking about wood use provides a window into complicating notions of gender. Wood procurement and consumption was part of every household's activities, elite and commoner. But the mechanisms of wood acquisition depended on how household labor was invested in broader processes (Hastorf and Johannessen 1991). The human ecology perspective on household dynamics (i.e., Netting 1993; Wilk 1983, 1997) would seem to contain the recipe for breaking down essentialist notions of gender: if households are adaptive units, they must adapt. Inflexible production, whether attributed to stagnant subsistence strategies or to gender, would be maladaptive. In this regard, it is surprising that ecological perspectives on the past that developed within processual archaeology failed to break apart the “black box” of the household and avoid gender stereotypes.

The issue is not whether women, men, boys, or girls collected firewood or performed any task exclusively. Rather, productive labor is situated in variable political economic contingencies (Pyburn 2004). The manner in which these households were articulated to larger networks of political and economic power influenced the tasks that household members did or did not perform. In this historical context, members of commoner households whose primary labor for regional political economies was through agricultural production and periodic labor projects differed from members of elite houses who participated in political economic networks via the production of prestige goods such as textiles.

Forms of activities and the knowledge derived from them are part of a broader political economic balance of give and take. Although weaving may have been a source of cultural capital for elite houses (Inomata 2001), farming and agroforestry constituted a complex repertoire of knowledge that many elites, male and female, young and old, may have lacked. As elites organized their daily rounds on productive activities that reproduced long-distance social, political, and economic networks, they invested proportionately less time in practices that enhanced knowledge of the local environmental milieu, such as collecting wood (Morehart 2006).

Figure 5.8 presents a schematic model of the relationship between access to woods and access to ecological knowledge. With more use of locally available hardwoods at Chan Nòohol and with a focus on agricultural activities, residents potentially would have had wider and more complex forms of locally based knowledge. With more use of pine at Pook's Hill and a greater investment in prestige good production, residents may have had greater access to and knowledge of more distant social and political relationships (cf. Helms 1998), but their locally based ecological knowledge may have been more restricted and less complex. More taxonomically specific identifications of wood species at both sites will be needed to develop this model more fully. Nevertheless, the major point is that knowledge is not a monolithic and normative entity. Social relationships, productive practices, and systems of knowledge are constituted by one's position in a historically contingent political economy (Morehart 2006).

Figure 5.8.

Schematic representation of the relationship between wood utilization and the complexity of knowledge.

Recognizing the contingency of ecological knowledge and practice reinforces and promotes a social view of the past (Hastorf and Johannessen 1991). Archaeological study of the micro-variability in commoner agro-technologies has been growing (e.g., Dunning 1996, 2004; Fedick 1996). But the vast majority of studies on knowledge remain centered on elite esoterica and ritual life, topics that are consumed easily by the media and popular audiences. By dehomogenizing the past and embedding knowledge in historically material social realities, archaeologists can address enduring, romanticized stereotypes of the Maya, both men and women, such as Ix Chel, the Maya goddess of weaving, medicine, and childbirth—the essential Maya woman—or the timeless maize farmer and native ecologist—the essential Maya man (see Ardren 2006; Klein 1988). A paleoethnobotanical approach that examines variation in a seemingly insignificant form of data—wood charcoal—complicates such stereotypes and may provide an innovative entry into understanding heterogeneity not only in social relationships, such as gender, but also in the systems of knowledge that pervaded such relationships during the Maya past.


  • 1

    It should be noted that the label of “Maya” itself has been imposed upon modern and, by extension, ancient speakers of “Mayan” languages. With the exception of the Yukatek, who term their language Maya‘t’aan (from which the term Maya derives), no evidence exists that other linguistic or ethnic groups placed under the heading of “Maya” originally recognized this classification.

  • 2

    Attempts to make species-level identifications of pine charcoal specimens by comparing archaeological and modern materials in transverse, tangential, and longitudinal sections were unsuccessful given the fragmentary nature of the archaeobotanical material.