Gender, Households, and Society: An Introduction



A critical strength of the discipline of archaeology is that its access to the material record of human history extends well beyond the written record and includes societies and cultures unaffected by Western colonialism and capitalist penetration. Bringing to light the social relations of earlier time periods, archaeology plays a critical role in documenting the full range of human variation, a role that cannot be filled by ethnography, history, or ethnohistory with their shorter temporal spans. By questioning essentialist notions of binary gender systems, gender research in archaeology can lead to a reevaluation of long-standing disciplinary assumptions about the nature of household organization, subsistence and craft production, ritual performance, and the structure of ancient states. The materiality of gender relations and gender identities in the archaeological record allows archaeologists to conduct historical comparisons of ancient, ethnohistoric, and ethnographic time periods to document changes as well as continuities in human social conditions. In so doing, archaeologists are able to expose ancient social scenarios that are distinctive from contemporary arrangements and thus widen the scope of the social sciences.

This volume demonstrates how archaeological data viewed through the lens of gender studies can lead researchers to question and reformulate current models of household organization, subsistence and craft production, ritual performance, and the structure of ancient states. Our central argument is that existing models of many aspects of prehistoric societies often assume the existence of rigidly binary gender systems. After three decades of feminist anthropology, few archaeologists claim that sex/gender roles and identities are fixed by the constants of human biology. And yet, a residue of assumptions from earlier views of male and female roles continues to color archaeologists' understandings of their data (Nelson 1997).

These lingering assumptions include (1) binary and exclusive models of gender roles and identities, (2) an expectation of fixed routines of domestic labor even in the face of significant economic and political change in wider society, and (3) an expectation that gender identity will always constitute a key axis of social organization. These are homogenizing assumptions: they postulate universal features of gender in human society across time and space. Subjecting them to critical scrutiny enhances our ability to recognize variability in the archaeological record and enriches archaeological model-building and hypothesis-testing.

Binary and Exclusive Gender Roles?

Feminist and gender archaeology began as a response to gender stereotypes and gender bias in archaeological interpretations of the past (Conkey and Spector 1984). Without much reflection, archaeologists had portrayed men in ancient societies as strong, aggressive, dominant, and active and women as weak, passive, and dependent. Archaeologists showed considerable interest in activities coded as male in Western culture (e.g., tool-making, hunting, trade, warfare, and power-building), but they ignored activities coded as female (e.g., food-gathering, food-processing, and parenting). Archaeologists had assumed a relatively rigid sexual division of labor in past societies, assigning activities to one sex or the other, with little overlap. These practices resulted in untested and probably inaccurate reconstructions of gender in past societies that were often simple projections of our own gender system onto the past. These uncritical models of the past then had the effect of naturalizing and legitimating gender stereotypes and gender inequality in American society.

Feminist and gender archaeologists responded with a series of remedial studies of gender in ancient societies. They questioned models that attributed human evolution to male hunting and male tool-making without considering women's contributions to foraging subsistence and women's abilities as toolmakers (e.g., Conkey and Williams 1991; Gero 1991; Gifford-Gonzalez 1993; Wadley 1998). They provided examples of women in roles at variance with Western stereotypes (women as hunters and warriors; women as miners, craft specialists, merchants, and long-distance traders; women as political and religious leaders) to demonstrate that gender roles were and are socially constructed and not biologically determined (e.g., Bell 2002; Bruhns and Stothert 1999; Doucette 2001; Hollimon 2001; Marcus 2001; Nelson 2003; Stalsberg 2001). They investigated women's roles in the development of food production and the connections between household production and imperial economies (Brumfiel 1991; Crown 2000; Gero and Scattolin 2002; Hastorf 1991; Jackson 1991; Stahl 2001; Watson and Kennedy 1991; Wright 1991). More recently, feminist and gender archaeologists have explored how women's and men's experiences within a society differed (Martin 1998; Meskell 1998a; Stahl and Cruz 1998), how engendered experience has varied according to class, age, and other differences (Brumfiel 2006a; Deagan 1996; Lightfoot et al. 1998; Meskell 1999; Robin 2004, 2006; Voss 2000; Wilkie 2004), and how genders (female, male, and other genders) have been constructed and perpetuated (Casella 2000; Gilchrist 1999; Hollimon 1997, 2000; Joyce 2001, 2004; Klein 2001; Looper 2002; Meskell 2000; Meskell and Joyce 2003; Prine 2000; Reeder 2000; Sørensen 2000; Weglian 2001).

But even as they have challenged some gender stereotypes, archaeological studies of gender have left other assumptions unexamined. Specifically, many archaeological interpretations still rest on an implicit assumption of rigidly binary gender systems. The belief in biologically based and unambiguous difference between males and females is a central tenet of Western gender ideology, but it is not a universal feature of human thought and practice (Errington 1990). Drawing on the male/female oppositions that were a prominent feature of structural anthropology (e.g., Lévi-Strauss 1969a, 1969b), some influential early contributions to the anthropology of gender featured such contrasts, for example, Rosaldo's (1974) opposition of female/domestic to male/public spheres and Ortner's (1972) suggestion that female is to male as nature is to culture. These models enabled feminist anthropologists to highlight the importance of the female principle in native thought, but at the same time they reinforced binary conceptions of gender (di Leonardo 1991).

In archaeology, the “remedial” recovery of women in ancient societies perhaps unwittingly reinforced the idea of separate and contrasting male and female spheres and experiences. Given the tendency of most archaeologists to emphasize “male” activities and ignore the presence of women in prehistory, some early efforts to “find women” in prehistory emphasized the complementary roles of men and women. For example, the model of “Man the Hunter” was countered by the model of “Woman the Gatherer” (Dahlberg 1981; Slocum 1975). These challenges to androcentric models of the past were effective in some ways, but they left assumption of rigidly binary gender systems intact. Structuralist analysis in archaeology has continued to generate binary models of gender roles and ideologies in prehistory (Hodder 1990; Huffman 1986).

The universality of rigidly binary gender systems is now challenged by several lines of evidence. First, archaeologists have demonstrated that the range of social roles played by women in the prehistoric past extended far beyond the domestic sphere and often overlapped with roles occupied by men. Women have been hunters, merchants, healers, athletes, warriors, rulers, and religious leaders (Bell 2002; Bruhns and Stothert 1999; Doucette 2001; Hollimon 2001; Marcus 2001; Nelson 2003; Stalsberg 2001). Equally important, archaeologists have documented cases in which women and men very likely collaborated in productive activities instead of adhering to a rigid female-male division of labor (Gero and Scattolin 2002; Robin 2002; Wright 1991).

Second, feminist scholars have critiqued the research that hoped to identify universal male and female roles in the sexual division of labor such as Murdock and Provost's (1973)“Factors in the Division of Labor by Sex: A Cross-Cultural Analysis.” Feminist scholars have found that even in research that seeks to identify sex differences through cross-cultural analysis, the results of research lead to the conclusion that few tasks are carried out by only one sex. Murdock and Provost's study of 185 societies refutes biological determinism: over two-thirds of the 50 tasks included in the study exhibit cross-cultural flexibility in labor organization (Du 2000:521). And this is true even though many of the ethnographies that served as a basis for cross-cultural research were carried out by Western male fieldworkers in societies profoundly affected by the male biases of Western colonial administrations (Pyburn 2004).

Third, historical archaeologists have demonstrated that even in the United States, binary male and female domains have not always existed. Wall (1994) traces the emergence of a division between male/public and female/private domains among middle-class New Yorkers from 1790 to 1840. Prior to 1790, commerce and domestic life were carried out in the same locale in workshops and yards attached to homes; after 1840, homes were removed to the peripheries of commercial districts. Historical archaeologists have investigated an array of plantations, brothels, mining camps, lumber camps, military posts, and boarding houses, all fundamental to the construction of America during the 19th and 20th centuries, where workers were not joined to nuclear family households (e.g., Barile and Brandon 2004; Galle and Young 2004; Mrozowski et al. 1996; Scott 1994; Seifert 1991). In camps where “traditional” families were present, women played critical roles in political activism (Wood 2004). Spencer-Wood (1991, 1999) demonstrates that even for middle-class Americans, the content and boundary of male/public and female/private spheres was continually negotiated and changing. By focusing on times and places where American gender relations were not strictly binary and exclusive, historical archaeologists reveal the artificial and arbitrary nature of American gender ideology (Leone 1999:9).

Finally, gay and lesbian scholars have challenged assumptions of exclusive and unambiguous heterosexuality (e.g., Butler 1990, 1993; Rubin 1984), and archaeologists have simultaneously examined the lives of individuals who represent third and other genders (Casella 2000; Hollimon 1997, 2000; Klein 2001; Looper 2002; Prine 2000; Reeder 2000; Weglian 2001). In addition, the concepts of performance (Butler 1990, 1993) and heterarchy (Ehrenreich et al. 1995) have convinced many archaeologists that all gender roles are labile and context dependent; that is, the expression of gender in behavior and symbolism even within a single culture varies, depending on the social situation (Joyce 1998; Levy 2006; Meskell 1999; see also Cohodas 2002). These developments have led some archaeologists to rethink the nature of sex and gender relations in human societies and to consider nondichotomous models of households, household labor, and social structure (e.g., Stockett 2005; Wiesheu 2006).

Uniformity in Domestic Labor?

The assumption that the duties assigned to women in the division of labor were based on their biologically fixed roles in human reproduction led many archaeologists to assume a tiresome uniformity in domestic labor. Structural change in economic and political organization was thought to occur in the male public sphere, outside the house, leaving production routines inside the house unchanged (see critiques by Brumfiel 1992; Luedke 2004; Silverblatt 1988). This, then, justified the uncritical use of the direct historical approach: the use of ethnographies from historically and culturally related groups to interpret the artifacts and to reconstruct the domestic routines of women and men in prehistory. Ethnographic analogy was applied particularly to commoner households, long regarded as the passive objects of elite political strategizing. While the use of ethnographies forced archaeologists to recognize some differences in domestic routines from one culture area to the next, variability and long-term change within culture areas were not well recognized. A catch-22 was in effect: archaeologists did not search for variability and change in domestic labor because they assumed that domestic labor was a constant, and because archaeologists presented very few examples of variation and change in domestic labor, the assumption that domestic labor was a constant went unchallenged. In fact, the static nature of domestic labor was enshrined in the common distinction drawn between “maintenance” activities inside the house and “production” activities outside the house (e.g., Binford and Binford 1969).

This situation began to change during the 1980s as household archaeologists began to take seriously the material remains of ancient houses that predominate in the archaeological record and to use those materials to understand ancient activities, activity areas, and labor organization (Wilk and Ashmore 1988). Social and economic perspectives on households exposed a rich array of household variability in the archaeological record. Although they were rooted in different issues and followed different trajectories, household archaeologists and feminist and gender archaeologists both came to question the assumed universality of a domestic/private/female/consumption/passive sphere that paralleled a public/male/production/active sphere. Both concluded that the “public” and “private” spheres are closely interrelated. People frequently respond to changes in the “public” sphere such as taxation, trade, and warfare by reorganizing their domestic routines, and people often create and alter the “public” sphere by intensifying household activities such as feasting, craft production, and ceremonial sponsorship (Allison 1999; Ames 1995; Brumfiel 1991, 2006a; Clark and Blake 1994; Costin 1993; Crown and Fish 1996; Dietler and Hayden 2001; Earle 1997; Feinman and Nicholas 2000; Hastorf 1991; Hayden 1990; Hendon 2004; Inomata 2001; Robin 2004; Stahl 2001; Wall 1994).

The responsiveness of the domestic sphere to evolving economic and political conditions suggests that household labor is a flexible, adaptive, and dynamic element of sociocultural systems and that it is best studied in relation to a broad array of environmental and social variables (Brumfiel 1992; Crown 2000). This responsiveness to changing conditions also suggests that ethnographic analogy should be used with great caution: current practices and beliefs that appear to represent unbroken continuity with the past should continually be tested against archaeological evidence (Ardren 2004; Brumfiel 2006a; Chance 1996; Hayashida in press; Robin 2002).

The dynamism of the household economy forces us to recognize that the domestic domain was not simply a passive and devalued version of a male public domain but was an integral part of the public and political life of a society. Houses were places where men and women, young and old, rich and poor came into concert, forming critical arenas for both learning about and contesting social norms. Households were platforms where individuals could articulate with the broader society around them and try to launch innovative behaviors and manipulate social relationships to advance their own ends (Allison 1999; Brumfiel 2004; Hastorf 1991; Hendon 1996; Robin 2003; Wall 1994). Relating gender and household studies, archaeologists can develop a view of the past as it was lived and consider how day-to-day practices and meanings are embedded in larger socioeconomic and political processes.

Gender: A Universal Axis of Social Organization?

Seeking legitimacy for women's and gender studies, feminists in the 1970s declared the primacy of gender as an axis of social organization. This claim for the importance of gender continues in much recent literature in which the culturally constituted nature of gender and sexual identity and experience are thoroughly explored (Gilchrist 1999:109–145; Meskell 2002). However, societies in which gender difference is not highly marked have received less attention (but see Joyce 2001:19–53). Today we recognize that while sex/gender systems are universally present in human cultures, they are not equally salient in all times and places. Some cultures have a gendered division of labor, but they do not elaborate upon the ideological implications of this division; others play it up (Conkey and Spector 1984:16). Some societies are deeply cleaved by gender; other societies have social categories that cut across gender lines such as kin groups, houses, ethnic groups, classes, and religious statuses.

Engels (1972 [1884]) asserted that gender oppression accompanied the evolution of private property and the state, and many have followed Engels in proposing that accentuated gender difference dates from the epoch when women's labor was harnessed to centralized political economies (e.g., Crown and Fish 1996; Gailey 1987; Joyce 2001). However, even this more limited proposal, that strong gender difference and inequality is associated with states, has been challenged (Pyburn 2004; Silverblatt 1988). The conditions under which gender difference is emphasized and gender inequality develops require further study. In our current state of knowledge, it cannot be assumed that salient gender difference is a universal feature of human culture or even an attribute of particular types of social formations. This issue is open to investigation.

Archaeology as Long-Term History

Research from contemporary and recent historical time periods indicates that Western and capitalist penetration in non-Western societies was earlier, more widespread, and more profound than once thought (Etienne and Leacock 1980; Leacock 1954; Wolf 1982). This implies that many of the ethnographic and historical data that are traditionally cited in discussions of gender and social relations are not representative of earlier time periods or the range of human societies that existed in the past. With this recognition, the analysis of prehistoric societies becomes increasingly important for scholars interested in understanding human social organization in general and gender systems in particular. However, archaeology can make this contribution only as it develops methods for differentiating the past from the present. Most scholars now recognize that the unexamined projection of ethnography or ethnohistory into prehistory erases the record of culture change and promotes essentialist views of both gender and culture. The result is an image of cross-cultural uniformity that does not reflect the full range of variability in ancient societies that archaeologists now painstakingly document.

Questioning the overextended use of ethnographic analogy in archaeology has a long history and broad implications for all archaeologists and scholars. Critiques have come from a range of theoretical perspectives, including processual and postprocessual approaches as well as feminist and household studies (Gould and Watson 1982; Johnson 1989; Stahl 1993; Trigger 1981; Wobst 1978; Wylie 1985, 1992). We suggest that a feminist perspective is uniquely placed to push archaeology forward on this critical issue because feminist scholars bring a critical perspective to the binary gender models that shape our understanding of so many institutions.

If we accept that all human thought is necessarily analogical, how do we move beyond the “tyranny” of ethnographic analogy (Wobst 1978)? The materiality of the archaeological record enables us to identify differences (as well as similarities) between the past and the present. Tacking back and forth between multiple lines of material evidence (e.g., artifacts, ecofacts, soil chemistry, and architecture) and the ethnographic record, archaeologists can identify dissonances that provide a footing for departures from ethnography. When the ethnographic record is used to identify differences between past and present, archaeologists are forced to consider the causes of these differences. Like a pebble in a pond, the initial recognition of difference between ethnography and the archaeological data can produce wider transformations of our models of subsistence and craft production, household organization, state structure, and artistic expression. The ethnographic record tyrannizes archaeology only when single-source ethnographic analogies obstruct archaeologists' recognition of difference in their own data. While our data exist in the present and even what qualifies as data is decided in the present, our data are also of the past. This past materiality provides a constraint upon and a mechanism for advancing our interpretive possibilities (Wylie 1992).

Gender, Farming, and Cloth Production: Moving from the Present to a Very Different Past

Two case studies drawing on Mesoamerican archaeology, ethnohistory, and ethnography illustrate the preceding points. Both studies started from quite traditional questions of gender and labor organization. For Robin (2002, 2006) the question was how to engender Late Classic Maya farm work (660–780 c.e.) at the farmsteads of Chan Nòohol in Belize. For Brumfiel (1991, 2006a) the question was recognizing and understanding the variation in cloth production across time and space in Mesoamerica.

When Robin set out to explore how previous researchers had engendered Classic Maya farm work, the predominant activity in this agrarian society, she was initially surprised by the absence of discussion in both research articles and textbooks. Textbooks might include an unreflective assertion such as “[a]s he still does today, the Maya farmer raised the native stingless bee” (Coe 1991:156). Researchers who attempted to engender ancient Maya farm work felt similarly compelled to assume a parallelism between pre-Columbian times and the ethnographic present: “It is interesting to note that Classic period Maya men are not depicted either farming or flint knapping, two productive activities thought to be male-associated based on ethnographic and ethnohistoric evidence and which I assume for this study” (Sweely 1999:161; also see McAnany and Plank 2001; Neff 2002). Sweely takes the position that given the apparent absence of Classic Maya archaeological data on gender in Maya farming, researchers must assume a concordance between the past and the present. As Fish (2000) had previously observed in reference to the lack of attention to questions of gender and agriculture in the American Southwest, Robin suggested that the ephemeral nature of farm settings and the absence of representations of subsistence production in prehistoric art had hampered initial efforts to engender ancient Maya farmers and led researchers to rely on ethnographic scenarios.

Reviewing the ethnographic, ethnohistorical, and archaeological evidence for gender and farming, Robin concluded, perhaps unsurprisingly, that the organization and social relations of farming were not static across the centuries of post-Columbian and pre-Columbian Maya society. Differences and discontinuities as much as continuities characterize Maya gender and farming relations across time and space. Throughout Maya history, farming was not simply a way of doing things but a set of social relations, and these relations were never predetermined but were deeply entwined with other social, economic, and political relations such as conquest, conversion, and changes in marketing, settlement, land tenure, and schooling.

While the initial impetus behind Robin's study was a critique of an essentialist gender stereotype, “man the farmer,” what became evident throughout the course of her study was how questioning an implicit gender assumption opened up new ways of thinking about the interrelationships of social relations, technologies, gender identities, and ideologies among the ancient Maya. Ultimately the significance of Robin's research was not to determine who (man, woman, child) did what in terms of farm work at Late Classic Chan Nòohol but to understand how the internal organization of day-to-day activities and labor production on farmsteads related to broader social, political, and economic issues in Maya society. From the artifacts, ecofacts, soil chemistry, and architecture of Chan Nòohol's farmsteads, Robin was able to document an intensive hill-slope terrace agricultural system that employed local understandings of the landscape and required the variable labor of all ages and genders of household members across different stages of terrace construction, use, and maintenance.

Diversified agricultural production was the central economic activity of Chan Nòohol's residents and the primary means, beyond labor itself, through which they could engage with the broader political world around them. Given the absence of formalized schooling, children were able to partake in the daily life of the farmstead. Fields were located short distances from homes. Women predominantly made soups and gruels for meals, meals that were less easily transported over long distances but also were less time-consuming daily cooking tasks for women than other forms of foods (such as tortillas), which gave them more time to carry out other tasks around the farmstead. Given the open nature of outdoor work spaces and pole-and-thatch buildings across farmsteads, as well as the proximity of homes and fields, men, women, and children who were collaborating on the same or different farmstead tasks would have been able to view and interact with one another throughout their day. Unlike much domestic work today, work around the house at Chan Nòohol would have combined domestic and nondomestic tasks performed by men and women, old and young.

Brumfiel's (2006a) research began by confronting the homogenizing “woman the weaver” stereotype. Women have been identified as weavers in many ancient and contemporary societies (Barber 1994), and throughout Mesoamerica women have often been associated with weaving (Hendon 2006). Comparing women weavers in three cultural and temporal contexts (ancient Maya, ancient Aztec, and 20th-century Mesoamerica), Brumfiel showed that deceptively similar images of women weaving on backstrap looms concealed important historical variability in the social, economic, and political roles of women weavers in their societies.

For the Classic Maya, cloth was an important prestige good. The association of elite women and cloth on monuments and figurines suggests that cloth-making was a high-status activity. Household excavations across the Maya area have revealed that tools used in cloth production are not uniformly distributed across Maya households; instead, they are much more frequent in higher-status households within a community (Beaudry-Corbett and McCafferty 2002; Halperin in press; Hendon 1997; Robin 1999). In terms of both the ideology of production and the locus of production, cloth-making in Classic Maya times was a high-status craft.

Seven centuries later, at the height of the Aztec Empire, cloth circulated widely as a market commodity and an item of tribute. Both nobles and commoners in Aztec society were involved in the production and distribution of cloth. Just as weaving had clear class associations in Maya iconography, weaving in Aztec ideology served as a strong marker of gender. Baby girls were presented with spindles as symbols of their womanhood, and female deities were depicted in association with spinning and weaving tools (Klein 1982; McCafferty and McCafferty 1991; Sullivan 1982). However, this cultural identification of women with cloth production seems not to have reconciled women to their fates as producers of large quantities of tribute cloth in the Aztec Empire. At Xaltocan, the alienation of women from tribute cloth production is suggested by the popularity of plain small spindle whorls, used to spin cotton for tribute cloth and market sale, during the Aztec period (Brumfiel 2001). Thus, cloth production was first a source of income for commoner and noble female weavers in Postclassic Mesoamerica but later an instrument of their oppression.

If weaving among the Classic Maya defined class and weaving in Aztec Mexico defined gender, then weaving in 20th-century Mesoamerica defined ethnicity. Weaving was used to produce distinctive community styles of clothing signaling a willingness to participate in local forms of reciprocity (Pozas 1977; Sandstrom 1991; Watanabe 1992). As the tourist economy expanded in Mesoamerica, handwoven cloth became a prominent craft produced for tourists that highlighted indigenous culture (García Canclini 1993; Morris 1991; van den Berghe 1994). In the emerging pan-Maya political movement, indigenous activists often wear pieces of clothing from other communities to express the solidarity that the movement hopes to achieve (Otzoy 1996).

With close attention to variation in the distribution of artifacts, images, burial goods, and production locales, what at first seemed to be a clear-cut example of static gender and labor relations through Mesoamerican time and space emerges as a testament to the diverse lives and meanings of Mesoamerican backstrap-loom weavers. The apparent paradox here is that the cultural elements that persist the longest are those with the greatest capacity for change. Ultimately, the importance of this study was not to determine that women (or men) at different times used a backstrap loom but to demonstrate that gendered cloth production within local settings was intricately tied to wider social and cultural processes and changes. Approaching the analysis of archaeological and ethnographic data through comparative historical analysis (rather than relying on ethnographic data to fill in for the archaeological past) revealed how gender was implicated with (and sometimes overshadowed by) other dimensions of social difference such as class and ethnicity at various times and places in Mesoamerica and responded to factors as varied as ritual practice, market trade, tribute assessments, land shortages, highway construction, free-trade agreements, and political violence.

Recovering the Diversity of the Past

The individual studies in this volume are unified by a common effort to question lingering assumptions concerning (1) the binary and exclusive organization of gender roles and identities, (2) the uniformity of domestic labor across time and changing political contexts, and (3) the expectation that gender identity will always constitute a significant dimension of social organization. This critical stance enables the contributors to propose new models of social life in a range of archaeological contexts. Probing the differences between archaeological data sets and ethnographic models, the contributors expose ancient social scenarios that are distinct from contemporary arrangements. Thus, questioning homogenizing assumptions about gender enables these archaeologists to reconceive basic human institutions, to recover the diversity of past societies and lived experience across time and space, and, finally, to free themselves from the tyranny of the ethnographic record.

This volume grew out of a graduate seminar at Northwestern University in the winter of 2005. The following fall, seminar participants presented their papers as an invited symposium at the 104th Annual Meeting of the American Anthropological Association in Washington, D.C. The symposium, “Unraveling the Threads of the Past and the Present: Archaeology, Ethnography, and Gender,” was invited by the Association for Feminist Anthropology and the National Association of Student Anthropologists.

On a theoretical level seminar participants were emboldened by Anne Pyburn's introduction to Ungendering Civilization (2004), which argues that when we uncritically accept deeply embedded gendered assumptions (for example, that women have always been confined to a narrow domestic sphere of activity because they are tied to childcare), we rule out the ability of our archaeological data to produce new understandings. The contributors to Pyburn's volume (also members of a graduate seminar) subjected to critical scrutiny archaeological data that had been thought to confirm Engels's (1972 [1884]) proposition that the rise of economic and political spheres beyond the household universally produced a decline of women's status. Examining eight non-European case studies, the contributors concluded that only in Mesopotamia does the archaeological record suggest a correspondence between the subjugation of women and the emergence of the state and social classes. And even in this case, societal subjugation of women seems to have existed before the institution of the state (Pyburn 2004:31).

In a similar vein Joan Gero and Cristina Scattolin (2002) question the implicit binary behind archaeological investigations that seek to identify gender relations as either hierarchical or complementary. Their research into early formative Yutopian gender relations in northwestern Argentina identifies how configurations of gender and work could be simultaneously hierarchical and complementary. Pyburn's volume and Gero and Scattolin's article lead archaeologists to question not only how binary thinking pervades our understanding of gender relations but also how it pervades our classifications of all aspects of ancient societies.

Bringing theory and method into concert, participants in our seminar questioned an “artifact-gender-attribution” approach that uncritically links types of artifacts with genders—for example, pots with women/celts with men—imposing a binary gender scheme on ancient material culture (see critiques by Conkey and Tringham 1995; Pyburn 1999). Spatializing material culture and gender studies emerges as a means to move beyond this problematic practice (e.g., Gilchrist 1999; Hastorf 1991; Meskell 1998b; Robin 2006). Thus the chapters in this volume attend to the spatial dimensions of activities and gender relations as a means of exploring social configurations in the past that may not have replicated a binary system. This allows the authors to make inferences about ancient gender relations without resorting to unwarranted gender attribution (Brumfiel 2006b).

A number of the contributors critique notions of a rigidly gendered universal division of labor, derived from ethnographic descriptions, especially those enshrined in the Human Relations Area Files. Morehart and Helmke demonstrate how essentialist notions of who collected firewood based on modern Maya ethnographies have discouraged archaeological investigations into the gendered work of wood exploitation in the Classic Maya past. Comparing rural Chan Nòohol with the better-connected site of Pook's Hill, Morehart and Helmke analyze how ancient gender relations, wood exploitation, and the organization of household production relate to the household's role in the broader political economy. Their study calls attention to the clear difference between the knowledge of local environments and wood use held by commoners who labored daily in their fields and that of high-status individuals who relied on imported wood even for everyday use. By relating people's situated practices in the world with their knowledge and understanding of the world, especially as these differed by class in Maya society, Morehart and Helmke move us toward an understanding of the past as experienced by ancient persons.

Coleman Goldstein challenges static ethnographic notions of a gendered division of labor in the Andes. She critiques the unchanging model of “lo Andino” derived from ethnographic accounts that enshrine a static division of domestic labor between a man and a woman living in a single coresidential unit performing different sets of repetitive tasks in noncollaborative work settings. Instead, the archaeological data from Inka and pre-Inka sites in the Upper Mantaro Valley of central Peru suggest that ancient domestic units were organized quite differently: multiple residential units engaged in communal activities of a domestic nature. Critiquing timeless notions of gender relations, Coleman Goldstein proposes that household organization was fundamentally different in the Andean past. She looks at hearths and grinding stones as items that were actively used by engendered persons within particular contexts in the past. By exploring variation in the contextual associations of hearths and grinding stones—fundamental elements of Andean cooking technologies—Coleman Goldstein is able to tease apart the identity of coresidential units and households in the Andean past, two social units that archaeologists have often considered impossible to differentiate from the scrutiny of archaeological data.

Miller examines essentialist notions of a strictly gendered division of labor in contemporary pastoralist societies that have obscured our understanding of the economic transition from hunter-gatherers to the first pastoralist societies in North Africa. Comparing rock-art, household, and subsistence data, Miller demonstrates the existence of ancient household variability and flexibility that are inconsistent with highly gender-segregated accounts of gender relations among contemporary pastoralists. Thinking about the archaeological record as inhabited by people, Miller uses archaeological data sets from Site E-75–6 at Nabta Playa in Egypt to determine how people, families, and communities would have organized the many economic tasks in which they were involved. As North African society moved from a hunting-and-gathering economy to an agro-pastoralist economy, these two economic strategies would have coexisted and changing economic needs would have necessitated a flexible division of labor. Miller's questioning of essentialist notions of gender allows her to better define the variable strategies that led some households to traverse the critical “economic transition point” between foraging and food production.

Along similar lines, Tejeda examines the important transition from local to regional polities in southern Guatemala during the Middle Formative. She argues that functionalist and macroscale interpretations of polity formation fail to reveal the changing social relationships that underlie the economic and political processes that lead to political development. The formation of a regional polity generates new economic demands at a time when most labor is still tied to household units; thus, static gender roles across this critical political transition are a practical impossibility. Tejeda documents changes in subsistence and craft production and interregional exchange that occurred at La Blanca as the site emerged as a regional polity and ceremonial center, and she concludes that such activities could occur only through change in household organization and the gendered division of labor. She defines the types of research strategies and data that would enable archaeologists to take a microscale approach to political change, one that would foreground the diverse social and economic accommodations that must have been made by Middle Formative households. As Tejeda points out, the microscale and materiality of domestic remains provide archaeologists a lens with which to view the day-to-day practices, meanings, and experiences through which societies were transformed.

Pankonien's critique of static notions of a gendered division of labor takes the form of exploring the implications of a timeless ethnographic notion that men do all the fishing in Huatulco, Mexico. Moving back and forth between archaeological data on shell use in Huatulco and the “man the fisherman” essentialism, she is able to show how women could not have been confined to domestic labor. From what might seem at first a minimal archaeological data set, the shell sample recovered from the excavations of one pre-Columbian site in Huatulco, Pankonien is able to explore how shell collection and use relates to questions of production, consumption, ritual, and trade in ancient Huatulco. By envisioning shells as items that were collected and used by people for particular purposes, people who were also engaged in other activities and with one another in their daily lives, she is able to explore the archaeological record of Huatulco shell work as one that would have been engendered and experienced. She proposes that women would have played a critical role in ancient shell exploitation and interregional trade.

Preston-Werner confronts static notions of strict gender dualism by critiquing the binaries of female/male, utilitarian/ceremonial, and tool/art. Examining a single line of archaeological evidence, ceremonial and utilitarian metates from ancient Costa Rica, investigators have assumed that women were associated with utilitarian metates and men were associated with ceremonial metates. Preston-Werner evaluates these associations against other lines of evidence, particularly Costa Rican figurines, which, as Preston-Werner demonstrates, depict women as well as men seated on elaborate metates. Thus, both men and women were associated with ritual and ceremonial practices that featured the use of elaborate metates as ritual seats. Contrary to a public/male/ceremonial/art versus private/female/domestic/utilitarian model proposed for pre-Columbian Costa Rican culture, the role of shaman cut across gender lines, and women were not limited to the use of simple metates for utilitarian tasks.

De Lucia takes as her point of departure the assumption of an innate relationship between gender inequality and states. Looking at art, burials, and the organization of household space, she shows how gender hierarchy was largely absent within the apartment complexes that were the basic residential unit at Teotihuacan. By documenting the absence of gender hierarchy as an organizing principle, De Lucia is able to show how other principles such as collaborative labor and compound-wide identities were fostered within the apartment complex. Compound identities superceded gender identities, and this was probably important in integrating newly arrived immigrants into the city. Common economic specialties, ritual spaces that were accessible to all compound members, and burials located below living spaces were the means for forging a common compound identity. Thus, the archaeological evidence of craft specialization, ritual, and burial enable De Lucia to understand the way people experienced compound identity in the contexts of their lives. By examining the assumed association of gender inequality and states, De Lucia is able to elucidate some of the distinctive features of state organization at Teotihuacan. De Lucia and Preston-Werner both show how the use of multiple lines of archaeological evidence can be a powerful tool for interpreting ancient social scenarios that diverge from contemporary expectations.

Philosophers and social theorists such as Henri Lefebvre (1991 [1947/1958]), Pierre Bourdieu (1977), and Michel de Certeau (1984) remind us that the most pervasive aspects of our lives are often those that are most overlooked and misunderstood. This volume argues that implicitly held assumptions about gender limit our understanding of the past. Even after three decades of feminist anthropology, hidden assumptions about gender persist in the face of research that ought to undermine them. These assumptions go unchallenged precisely because they are so central, naturalized, and ingrained in our society. Unrecognized, these assumptions constrain archaeological interpretation in numerous archaeological contexts.

While the case studies in this volume derive from the Americas, with the exception of Miller's chapter on agro-pastoralism in North Africa, the volume draws attention to issues and approaches with universal significance. Challenging assumptions about gender is at the heart of a broad range of archaeological debates surrounding questions as varied as the nature of human origins (Conkey and Williams 1991; Gero 1994), the origins of agriculture (Fish 2000; Peterson 2002; Watson and Kennedy 1991), and the rise of the states (Joyce 2001; Pyburn 2004; Silverblatt 1988; Stahl 2001).

The authors in this volume demonstrate how the assessment of long-standing assumptions about gender activities, roles, and identities provides new insights into other traditional topics of interest in archaeology. These assessments are rooted in recognition of the differences between the material record of the past recovered by archaeologists and the models of the past built on the uncritical use of ethnographic analogy. In making these assessments, the authors in this volume demonstrate that gender archaeology can play the central role in creating an archaeology that is an important independent source of understanding of human nature and human variation.