Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Introduction

The integration of gender into archaeology as a “structuring principle fundamental to interpreting past societies” (Gilchrist, 1991:499) is still incredibly new to the field. Archaeologists may have become more attentive to their assumptions regarding gender in the years since Conkey and Spector’s (1984) important treatise, “Archaeology and the Study of Gender,” but only 18 years ago, Alison Wylie (1991) still had cause to ask, “Why is there no archaeology of gender?” While historical archaeologists had a timely response to her plea with the publication of Gender in Historical Archaeology (Seifert, 1991a), the newness of the consideration of gender made this response varied with regard to what constituted appropriate goals, stemming from differences in the conceptualization of gender and resulting in differences in the methodological approaches to its archaeological study. This variation can be broken down into three basic types of studies: (1) those seeking to establish the material correlates of one or more gender-based groups or constructions, (2) those interpreting archaeological assemblages through gender as a principle that structures society, and (3) those drawing upon the role of gender in constructing identity.

Following a brief synthesis of the conceptual issues that have challenged and motivated the study of gender in archaeology, this chapter provides a survey of each of the three types of studies and highlights relevant theoretical and methodological issues within the context of the continuing development of gender studies in the field of historical archaeology. This survey, primarily drawing upon North American examples, should quickly make clear that the three types of studies have not been mutually exclusive, in that the emphasis of each is necessarily present, whether explicitly or implicitly, in the background of the others. One cannot assess how material culture reflects gender without recognizing that gender imposes some type of order upon human beings and, in that sense, constitutes a structuring principle, nor without recognizing that structuring is related to the identity, constructed by the self or by others, of men as men, women as women, or other gender categories as other gender categories. Because these studies are archaeological in nature, examinations of gender as a structuring principle or its relation to the construction of identity necessarily address at some level the material manifestations of these functions. Recognition, however, of the overlap between and integration of these emphases has occurred to differing degrees. Increases in both have had a positive effect on the evolution of the study of gender in historical archaeology.

Gender Is as Gender Does: Concepts and Reconceptualization

In responding to her own question regarding the absence of an archaeology of gender, Wylie (1991:34) maintained it was due partially to the “commitment to theories which focus on other classes of variables as the primary determinants of cultural behavior” and which often are assumed, unnecessarily and incorrectly, to be incompatible with the consideration of gender. Primarily, however, she cited the projection of recent Western gender structures onto the past, resulting in and reinforcing the perception that gender roles were biologically determined, therefore stable, and therefore requiring no further development of resources for their study. The assumption that gender roles are biologically determined is linked to what were, historically, difficulties fundamental to gender studies in a variety of fields: the conflation of gender with sex, a biological given, or with sexuality, a personal, projected, or perceived identification with sexual preference, practice, or relations. Gender is neither of these, but it may invoke both.

Gender is the cultural construction of “man,” “woman,” or other related categories made up of social and cultural ideas about what each category means (gender ideology) and the appropriate roles for each category (gender roles). These constructions are incorporated into the ways individuals perceive themselves or are perceived by others to fit into a category (gender identity) (Conkey and Spector, 1984:15; Scott, 1994a:10). Because gender is sociocultural and not biological, it is not universal or inherently stable, and it does, therefore, require the development of resources for its study. Understanding gender as sociocultural construction means that an archaeology of gender must engage the culturally specific processes and effects of that construction, moving well beyond biological ties. Despite, then, the major impetus provided by feminist voices (e.g., Conkey and Spector, 1984; Engelstad, 2007; Gilchrist, 1991; Spencer-Wood, 1991, 2001; Voss, 2006; Wylie, 1991, 1994, 2007) to an archaeology of gender, reference to a woman or women, finding women in the archaeological record, or even studying an individual woman or group of women as the primary subject of an archaeological study does not automatically render it gender archaeology. Only when “woman” and other gender constructions are examined in relation to the processes and effects of creating them can a study be defined as such.

Historical archaeology, in this respect, benefited from its relatively late entrée into gender studies. When the authors of the papers in Gender and Historical Archaeology were asked to contribute to the volume, “they were asked to move beyond looking for women in the archaeological record through identifying artifacts associated with women; they were asked to stretch further and to examine the relationship between their data and gender as a structural principle of culture” (Seifert, 1991b:2), indicating that even this early foray into gender in historical archaeology was well informed by gender theory. Even so, historical archaeologists have struggled to make operational its salient points, most commonly by enlisting the notion of gender-exclusive material culture (see below). The idea that such material culture exists stems largely from “separate spheres” gender ideologies that assign men the qualities of public, political, production, active, and culture and women those of private, domestic, consumption, passive, and nature (reproduction). These ideologies were historically constructed as binary oppositions ultimately based on biology/sex, thereby reducing the real-world complexity of gender and disguising the shared usage of material culture, albeit sometimes in different ways (Gilchrist, 1994; Kerber, 1988; Wurst, 2003). Reliance on binary oppositions has colored even those studies that identify the crossing of members of one gender group into the realm of another; these studies universalize those realms by supposing that infiltration occurs instead of recognizing “infiltration” as an indication that gender categories are not stable. The result is the “[piling up of] one cautionary tale after another, hoping to smother the flawed logic beneath its weight” (Wurst, 2003:229).

In 2003, Wurst noted, “To date, most of our attention has focused on defining what gender is, and the corollary, where or how gender can be linked to the material world. I find it useful in this context to raise the question of what gender does” (Wurst, 2003:231). By reconceptualizing what gender is as what gender does, the processes and effects of gender construction come to the forefront of archaeological studies. Processes and effects, of course, are not tangible and therefore cannot be recovered as artifacts from the archaeological record. Material culture, however, that conveys these processes and effects is tangible; thus in regard to those who cannot speak, material culture is necessary to understanding these processes and effects, as well as adherence to or manifestations of gender roles and ideology and how these might have been or not been expressed individually or within groups. Despite, therefore, any shortcomings that have been evinced in the material-correlate class of study, such studies and the cautionary tales they generated represent important steps toward an appropriate methodology for studying gender through historical archaeology.

Presence over Process: Material-Correlate Studies

Studies that can be classified under this heading are those that have emphasized the identification of the archaeological signatures of gender groups in the context of a specific historical situation. These material culture studies are typically couched within a discussion of pertinent gender ideology, roles, patterns, and/or interactions, which alludes to gender as a structuring principle. Frequently, though, the material culture study and the discussion have not been successfully linked, so that in the end, the material culture points to presence and not process.

In her study of women in the Spanish New World, for example, McEwan (1991) establishes late-fifteenth-century and sixteenth-century Spanish gender ideology as one in which women were expected to be ordered, restrained, pious, and chaste and to invest their time in domestic and religious endeavors. She addresses these ideological expectations as a background for her contention that Spanish women were powerfully influential in their homes and thus in disseminating Spanish culture among their Native American and African-American domestic workers. Additionally, she notes that Native American and African-American wives “assumed the roles traditionally held by Spanish women with regard to food preparation, child rearing, and homemaking” (McEwan, 1991:39). Based on this information, she establishes the material correlates of Spanish women as those of a domestic nature, identifies the material correlates of Native American women who married Spanish men also as largely domestic and including ceramic cooking pots, storage vessels, stone manos, metates, and manioc griddles, and discusses the potential for pottery to represent African-American women who married Spanish men.

The material correlates, though associated with Spanish gender ideology, are not considered as a function of what gender does. Concluding, for example, that Native American and African-American women were agents of acculturation without assessing how gender structured their cultures prior to involvement with the Spanish insinuates that Spanish gender constructions mirrored their own, missing the ways in which Spanish gender constructions may have restructured the lifeways of Native and African-American women and how material culture might reflect, for example, the extent to which such restructuring occurred. Similarly, interpreting intermarriage as a means of “stabilizing and converting the Native element” (McEwan, 1991:36) to Christianity misses the possibility that intermarriage was a means of maintaining existing gender structures within Spanish society and how material culture might shed light on the importance of such maintenance, for example, in support of a particular mode of production.

Like McEwan, Jackson (1994) searches for the means to identify women in the archaeological record, in this case, of Russian America during the mid- to late 1800s in relation to the fur trade. Though much background information is provided that points to the gender roles of the Tlingit and other Native Alaskans at that time, the primary goal of the study is to demonstrate how food preparation and garment making, activities that “belong to the female domain” (Jackson, 1994:30), appear archaeologically in general and in relation to stylistic attributes. Though she notes that Tlingit women participated in the fur trade and that a sexual division of labor existed in at least one Native Alaskan society, Jackson uses these only to support the arguments that women engaged in fur-trading activities and that their participation can be evidenced by cloth, clothing, and related paraphernalia, with the implication that eventually archaeologists can determine gender roles within the fur trade. That this study is geared toward finding women is indicated by Jackson’s argument that if women are not identified through their association with specific artifacts, then “artifacts remain in a disconcertingly gender-free environment” (Jackson, 1994:30). If gender is considered a principle structuring society, no artifact can remain in a gender-free environment.

Studies focusing on finding specific groups by establishing their material correlates have often failed to meet the challenge of avoiding circular reasoning in interpreting gender in the archaeological record. If, for example, a known gender ideology dictated that women should do housework, based on that ideology, one might assume that a woman would leave behind an iron, then use the iron as the basis for suggesting a woman was present at the site and adhered to that ideology. What that might mask is that the man who actually used the iron operated in opposition to that ideology. The challenge is then posed of identifying who contributed what to a given assemblage. This difficulty is highlighted by Starbuck’s (1994) study of gender at eighteenth-century U.S. military sites. He discusses the inability to locate women’s artifacts, noting that “in military settings, men and women were apparently often using and sharing virtually the same material culture, so their identities are not easily distinguishable archaeologically” (Starbuck, 1994:124). Starbuck’s study, in noting the potential for shared material culture between gendered groups, reinforces what has been problematic in many archaeological studies considering gender, which is that artifacts are interpreted by default as related to men’s activities, unless they relate specifically to women’s fashion, hygiene, or medical conditions or to domestic activities. Interpreting gender in this way eliminates not only women who might have smoked pipes or participated in agriculture but also men who might have sewn or prepared food.

The assumption that artifacts are “male” or “female” conceals the fact that many artifacts are “both” and, therefore, suggestive of the interactions between and activities shared by men and women. This problem was most recently tackled by Spude (2005) with regard to prostitution-related assemblages, where difficulties in interpretation are compounded by the knowledge that both women and men were involved in the creation of these assemblages and sharing several activities, including eating, drinking, smoking, and sex. She suggests that the archaeological identification of gender categories should be limited to those artifacts, such as earrings, cosmetic containers, douching paraphernalia, suspender buckles, and jeans rivets that overwhelmingly point to “male” or “female.” Though Spude’s approach may prove useful for identifying contributors to a site and the site type, ultimately the use of material correlates to identify the presence of a member of a given gender category, i.e., to find women or men, in order to analyze the relationships of an assemblage to the resistance to or maintenance of gender roles and ideology has tended to reify gender roles within society into the stable and universal entities that they are not. It was the recognition of the potential for incorrectly assuming and reifying gender roles that led many historical archaeologists to refocus their study of gender by considering gender as a structuring principle.

Process over Presence: Gender as a Structuring Principle

The dangers of assuming and reifying gender roles, as has often occurred in material-correlate studies, were elucidated in one of the earlier historical-archaeological studies to address gender as a structuring principle of society. Purser’s (1991) study of gendered patterns of mobility between mining camps along Port Wine Ridge in California led to the discovery that the pattern of women’s visiting was more than just a domestic-ideologically approved activity. These reciprocal visits served social and practical functions in maintaining community. Unlike men’s visits, which occurred over short time frames in saloons, general stores, and other public venues, women’s longer visits to relatives and friends “created and maintained sets of social ties that provided some continuity to the community as a whole. Some of these networks linked and reinforced kinship ties; others mirrored, and in some instances presaged, sets of small business partnerships seen among the men of the ridge community” (Purser, 1991:11). Additionally, Purser found, during the summer and early fall, that women residents of one of the mountain mining camps, Grass Flats, would visit relatives or friends in the agriculturally based foothill settlements with an older child, who she would leave there until just before winter. In this way, the “mountain household fed one less mouth during the lean, non-mining season and possibly provided additional labor to the valley household during summer harvests. It also linked the mining-camp household with a supply of fresh vegetables, fruits, and meat” (Purser, 1991:12). In the spring, valley residents would visit their mountain counterparts to take advantage of the recreational opportunities afforded by the mountain setting.

Perhaps more important in this study than the elucidation of these community-related functions, however, was the exposure of several originally made assumptions that, without further analysis of the data, would have rendered Grass Flats women as acting out fixed roles associated with Victorian gender ideology. This caused Purser to reevaluate her earlier study of the women of Paradise Valley, Nevada, in which several of these assumptions had been carried through the examination of relevant historical documents. These assumptions were that Grass Flats women were simply participating in activities approved by Victorian domestic ideology, an assumption that Purser (1991:13) attributes to the “danger inherent in the rich body of comparative historical literature available to historical archaeologists and the seeming familiarity of the subject matter.” It was assumed that gendered archaeology had to be considered at the household level and that gender was a specialized topic instead of a structuring principle. Regarding the latter, Purser notes, importantly:

The intellectual shift described [in this study] only begins to move from excavating “women” in “households” to perceiving gender relations as an historically constituted structuring principle inherent throughout society. It raises questions about what is being excavated, and why, in specific contexts, but cannot yet answer those questions fully. These questions are not about “doing” versus “not doing” gender in historical archaeology, or how to “find” women or men in the archaeological record. Rather, they examine the extent to which recognizing the gendered character of social life problematizes archaeological concepts like household, community, or human mobility.

In raising such questions, the personal transition from looking for women to looking through gender also begins to articulate the critical potential of gendered research with other, parallel arguments for critical, self-reflexive approaches in historical archaeology (Purser, 1991:13; italics in original).

Several studies have since reflected positively on the points raised in Purser’s study: that gender functions as a structuring principle at multiple levels of society, that recognizing gender as such creates a significant shift from material-correlate studies in the goals of archaeological excavation and the interpretation of material culture, and that looking through gender allows gendered research to be articulated with that of class and ethnicity, principles also considered in self-reflexive approaches in historical archaeology.

Multilevel Approaches

The need to consider gender as a structuring principle of society at multiple levels was aptly demonstrated by Hardesty (1994) in his prospectus for an engendered archaeology of the nineteenth-century American mining West. Within this prospectus, Hardesty presented historical data to illustrate the organization by gender of households, communities, and mining districts and archaeological data to suggest how such organization might be reflected in the material record. At the household level in the mining West, gender frequently structured household activities in ways that are visible archaeologically. Citing Blee (1991), Hardesty notes, for example, that all-male residences tend to have a low percentage of liquor-related artifacts because Victorian gender ideology abided public drinking by men, who did not therefore have to indulge at home. He additionally suggests that archaeology may be used to study the degree to which households were organized by that ideology. Victorian gender ideology further organized community geography, which also can be examined archaeologically. Hardesty cites, for example, the segregation of women who operated outside of Victorian gender ideology through prostitution into red-light districts. Additionally, he presents evidence for the structuring by gender at the community level in the correlation between (1) the absence of special-purpose buildings, (2) the presence of special-purpose buildings mostly for men’s activities, and (3) the presence of special-purpose buildings used for both women’s and men’s activities and (1) mining towns that experienced a quick boom and bust, (2) working-class mining towns of greater duration with a well-defined class structure but predominated by men, and (3) mining towns of yet greater duration with a well-defined class structure and a more balanced sex ratio, respectively. Hardesty cites the above-mentioned study by Purser (1991) to support the idea that gender structures mining districts, but also notes regional gendered patterns, such as those “centered around a town with a variety of outlying settlements…Most of the women, for example, lived either in the town, with families at outlying ranches, or at toll stations managed by families. Men occupied the working-class satellite settlements clustering around the mines or mills scattered throughout the district” (Hardesty, 1994:141). The evidence for the organization by gender at each level supports Purser’s contention that an archaeology of gender should not be limited to household studies.

While Hardesty’s and Purser’s studies make clear the effectiveness of a multilevel approach to gender organization in the mining West, the necessity of considering multiple levels in other contexts has been corroborated elsewhere. Spencer-Wood’s (2006) archaeological approach to American Utopian communities indicates that the structuring of society by gender is not only evident where Victorianism constituted the dominant narrative. Though her approach uses the separate spheres dichotomy as a basis for comparison, because some of these communities promoted gender equality, this study is able to tread a relatively new path. It should also be noted that in doing so, it bolsters the argument against the dichotomy as stable or natural. Spencer-Wood explains the ways in which the landscapes, intra- and interspatial building layouts, their functions, and associated artifacts in sites associated with Utopian communities are reflective of practices associated with Utopian ideologies, particularly gender ideologies. The presence of one or more community-scale buildings, for example, as might be indicated by large foundations, could support, with the appropriate suite of artifacts, the practice of cooperative housekeeping, which occurred in several Utopian communities promoting gender equality because it freed women to pursue nondomestic work. To interpret the archaeological data, then, requires viewing them through gender, but it also requires doing so at multiple levels; in this case, examination of a single household instead of the community landscape would miss this indication of gendered processes.

Perhaps one of the more interesting and instructive studies of gender to move beyond the household level is Kryder-Reid’s (1994) examination of the nineteenth-century, all-male religious community of the Redemptorists, a study that clearly shows gender not to be equated with biological sex through a case in which gender structured society even where members of only one sex were present. The Redemptorist community associated with the St. Mary’s site in Annapolis, Maryland, comprised ordained priests and lay brothers. It was a cloistered community in which the priests trained students for ordination, whereas the brothers committed to a life of “service and bodily labours… constant and careful in discharging the various domestic employments, always mindful that they have come to serve” (quoted in Kryder-Reid, 1994:104).

Despite the Redemptorist promotion of equality between priests and between brothers, the Redemptorist Rule made clear that a hierarchy should be present between the priests and brothers, with the latter ranking below even those studying to become priests. Using historical documents from the Redemptorist community and drawing on the Victorian ideal of True Womanhood, which promoted piety, purity, submissiveness, and domesticity, Kryder-Reid shows that the community was organized through the engendering of the priests and brothers with masculine and feminine roles and attributes, respectively. Though both groups were expected to be pious, chaste, and submissive to the Rule, the latter meant that the brothers had the added charge of being submissive (reverent and servile) to the priests. Furthermore, the duties of the brothers were largely domestic in nature: infirmarian, or nurse, gardener, porter, refectarian, cook, tailor, sacristan, caller, and Brother Procurator (Kryder-Reid, 1994:105–106, 110).

Kryder-Reid concludes the study by discussing the challenges of identifying the material culture associated with each gender group. The challenge of avoiding circular reasoning, as discussed above, is implied in her statement that it was impossible to know whether a collar button recovered from the St. Mary’s site, from a collar that was only supposed to be donned by priests, represented “an obedient priest or a renegade brother” (Kryder-Reid, 1994:108). Furthermore, the communal nature of the deposits from features such as cisterns prevented association with either one gender group or another, making it difficult to test adherence by the brothers to domestic duties. Though she uses the example of minimally processed meat to suggest that food preparation was conducted in the cloistered community by the brothers as opposed to the public butcher and finds evidence in archaeological and historical landscape data that point to the domestic duties that would have been performed by the brothers, she acknowledges that the material-cultural challenges are steep. Hence this study, like Purser’s, Hardesty’s, and Spencer-Wood’s, is largely theoretical.

Social Structure and Material Culture

A number of studies, however, have innovatively interpreted material culture through the lens of gender as a structuring principle. One of the earliest and most well known of these is Wall’s (1994) analysis of ceramics from late eighteenth- through mid-nineteenth-century households in New York City. This period encompasses the transition to a dominant (Victorian) ideology of separate spheres marked by the removal of in-home businesses to separate locations, the separation of residential districts from business districts, and more pronounced differences in prescribed gender roles. It had often been suggested that middle- and upper-class women crystallized the domestic sphere only after their husbands began to leave them for the public sphere on a daily basis. One result of this crystallization was the ritualization of family meals, including their presentation, and was supposedly catalyzed by the separation of the family during the day. By comparing the stylistic attributes and composition of chronologically separate ceramic assemblages from middle-class and wealthy households spanning the decades between the 1780s and 1830s, Wall (1994) was able to demonstrate that women were active participants in constructing their gender roles, which in turn structured social practices. Changes in decorative motifs toward the religiously influenced Gothic style, the increased cost of the dishes used for family meals, and a move toward matched sets of dishes prior to the physical and conceptual separation of home and workplace and throughout the period of study indicate that women were not simply responding to the departure of men to the public sphere. Rather, they were actively involved in the structural transformation of society by participating in the creation of their gender roles.

Goodwin (1999), similarly, assigns an active quality to the gender roles of merchant-elite women in seventeenth- and eighteenth-century Massachusetts. In her study of merchant society in this spatiotemporal context, she contends that gender roles associated with mannerly behaviors, as outlined in contemporary prescriptive literature, complemented each other to reproduce merchant society and to provide the merchant class with a means for maintaining the desired separation from what they considered lower classes. Goodwin suggests that while it was typically men who engaged directly in the trade and business negotiations of merchant society, women, by putting on and presiding over events such as teas and dinners where alliances, financial, marital, or otherwise, could be made, negotiated the social networks that upheld and reproduced it. She proposes that the material culture of merchant society should be interpreted through the lens of mannerly behaviors and their broader role in maintaining the lifeways of that class.

Class, as Wall and Goodwin indicate, was a component of the ability to create the gender roles identified in their case studies. This point is reinforced by Gibb and King (1991) and Bell (1995), who used the spatial distribution of material culture to assess the adherence by English colonists to seventeenth- and eighteenth-century English gender ideologies and associated roles. In analyzing three archaeological homelot sites in the Chesapeake Tidewater region, Gibb and King found that specialized activity areas, which they interpret as “evidence of a sexual division of labor expressed in the distinction between home- and commodity-production” (Gibb and King, 1991:128), were most, though not strongly, pronounced at the wealthiest of the homelots. They concluded that gender organized the homelots in relation to labor and space but noted that the way this organization occurred was based on what the wealth of the landowner could support. In wealthier households, landowners could afford to sustain the personnel required for specialization of tasks, who in turn produced sufficient quantities of goods to maintain the wealth of the landowner. Part of this arrangement, they note, was the involvement of women in home production and men in commodity production. Gender, therefore, structured social organization by promoting the accrual and maintenance of wealth in households that could afford the gendering of labor. Such specialization could not be supported by those of lesser means. Bell (1995), working in the same region on a site dating to the early- to mid-eighteenth century, found that at Flowerfew Hundred Farm, many of the artifacts were reflective of the middle class, but the spatial distribution of the artifacts and lack of specialized activity areas were more consistent with a lack of middle-class ideals. Her conclusion implies that although English gender ideology may have been influential, it was not fully accepted by all, including members of the middle class.

The notion of resistance to the structuring of society by gender per the dominant ideology, as implied by Bell’s conclusion, has gained ground in the historical archaeology of gender as it moves away from simple dichotomies. Casella’s (1995) study of nineteenth-century Tasmanian Female Factories has its background in the Victorian ideal of True Womanhood, but only to show how women in these reform institutions were constructed in opposition to that ideal. Importantly, her study demonstrates that gender is not a binary category, as these women did not cross over into the male sphere but into an alternative one (there is nothing ideologically male about crime or prison), that these women, like the Redemptorists of Kryder-Reid’s study, constructed gender in a primarily single-sex setting, in this case to express power instead of submissiveness, and that such construction had an effect on the adherers to the dominant ideology, structuring their actions and responses as well.

Despite the fact that most of the women in the Tasmanian Female Factories had committed the crime of thievery, they were sexualized through being conceived of as and termed whores, cementing their non-womanhood in the eyes of prison officials. The prisoners built on this image through sexual displays, though Casella (1995) notes that it is unclear whether these displays included sexual acts. The implication of such, however, was enough for prison officials, who noted alternative gender identities (and possibly sexuality), referring to some of the prisoners as “men–women” and noting the propensity of other prisoners to primp for and fight over their attentions. As a result, during the mid-nineteenth century, prison architecture at the female factories was revised to incorporate double-room solitary cells that allowed prisoners to sleep and light to work by while giving officials the means for inmate surveillance and separating the women.

Archaeologically speaking, gender as a structuring principle over multiple levels is manifest in the prison landscape. The dominant gender ideology structured greater society and demanded that these women, who as criminals could not be “Women,” be separated from it, which is physically manifest in the presence of the prison. The construction of gender by the prison population eventually structured the internal space of the prison. At both levels, the architectural components of the prison are indicative of resistance to the dominant gender ideology. As Casella (1995:38) notes, “Why would prison officials bother to fortify a fenceline, brick over a window or erect a lamp post in any particular location? If institutional containment is understood to be a constant, fluid and partial negotiation of power, then the architectural features were responses or challenges to subordination as much as they were methods or statements of domination.”

Landscape and architecture have been similarly used by other historical archaeologists to examine the relations between gender, power, and space. In Delle’s (2000) study of late eighteenth- and early nineteenth-century Jamaican coffee plantations, for example, he maintains that despite clearly unequal class relations between the slaves and enslavers, the engendering of spaces, resulting in enslaved women’s control of provision grounds where they grew food to sustain their families and to sell at market, afforded them some sense of autonomy and empowerment. Gilchrist (1994), in her analysis of England’s medieval nunneries, examines the ways in which the surrounding landscape and the architecture of the nunneries were structured by gender, particularly in comparison with male monasteries. She argues that the ways these elements were manifested were intertwined with the gender constructions of the “social group to which [the nunneries] were predominantly linked: the local gentry” (Gilchrist, 1994:188), and with power relations between gender groups and within the nunneries during the period in which they operated. Archaeological gendered landscapes in the United States are the subject of a recent edited volume (Rotman and Savulis, 2003) that explores these resources in numerous and varied contexts, from the Shaker communities of the 1830s and 1840s (Savulis, 2003) to the creation of the Vietnam Veterans Memorial in the early 1980s (McGirr, 2003), and firmly establishes the role of gender in structuring the material and the spatial along with realms of activity.

As noted above, however, and as was apparent from many of the studies described thus far, gender does not work alone in structuring society, and historical archaeology continues to undergo the process of understanding how gender interacts with other constructions, most commonly class and ethnicity/race. Attention to this combination of principles is evident in the titles of such edited volumes as Those of Little Note: Gender, Race, and Class in Historical Archaeology (Scott, 1994b) and Lines That Divide: Historical Archaeologies of Race, Class, and Gender (Delle et al., 2000), which show that historical archaeologists have engaged these topics for some time. A shift, however, has occurred toward an understanding of gender as an isolated principle to one that is inextricable from class and ethnicity in the process of structuring society.

Gender, Class, and Ethnicity

In earlier years, the trend toward understanding the interactions of gender, class, and ethnicity manifested itself in the viewing of one construction through one or both of the others. In Hardesty’s article on engendering archaeology in the mining West, for example, he says that the ways in which “mining towns were organized by gender [have] to be interpreted within the context of class divisions” (Hardesty, 1994:131) and that “emigrants carrying quite different cultural and social traditions [affected] the way in which gender structured the community” (Hardesty, 1994:134–135). Similarly, Wall (1999:102) “explor[es] the extent to which class and ethnicity…structured the construction of gender in the [mid] nineteenth-century metropolis [of New York City]” by comparing data from her 1994 study and another middle-class household with material culture from a working-class tenement. Griggs (2001:83), alternatively, considers how “household makeup and economic strategy were influenced by ethnicity and gender” in the working-class neighborhood of New York City’s Five Points in the middle to later part of the century. Using the types and qualities of sewing tools in combination with the quality of textiles recovered from archaeological deposits, Griggs found that working-class Irish families on Block 160, often headed by widows or other single women, needed to recycle and reuse textiles and remnants to supplement household income and meet their needs. On the other hand, Jewish households, which were often more standard nuclear families in which men and women could pool their incomes and divide their labor among them and their children, were financially better off.

These early studies defined gender, class, and ethnicity as interacting structuring principles but are characterized by a unidirectional approach of looking at one principle through others. Recognizing that this approach may give the illusion of primacy to one principle over the others in structuring society, a recent framework developed for the study of these principles (Vermeer, 2006) has called for a multidirectional approach that recognizes the complexity of these interactions in light of works that have called out, for example, class as a function of gender (e.g., Hill, 1993), or the inseparability of gender from race (e.g., Glenn, 1992). Additionally, many of the early studies continued “to identify all of these aspects as objective traits or attributes that characterize individual identity” (Wurst, 1999:8). As a result, more recently, some historical archaeologists have begun to promote a relational view of these principles (e.g., McGuire and Reckner, 2002; Vermeer, 2006; Wurst, 1999), in which each individual principle “is not an entity that changes or reacts to history, but a set of relations that are historically constituted, fluid and constantly changing” (Wurst, 1999:9).

Kruczek-Aaron (2002), employing a relational view of gender, demonstrated that the highly decorated ceramics used in the Gerrit Smith household during the nineteenth century may represent a gendered struggle over self-presentation there. Smith, a politician involved in the abolitionist and temperance movements, believed that any type of display, whether a garden, wedding, or home interior design, should be simple because “sacrifice in terms of consumption set the example of pious living” (Kruczek-Aaron, 2002:179). The household’s transfer-printed pearlware vessels and dishes, however, bearing elaborate designs and “often aristocratic subject matter…may indicate that the Smith family women [Gerrit’s wife, Ann, and his daughter, Elizabeth] were asserting themselves in the struggle over material culture” (Kruczek-Aaron, 2002:180). By examining the ceramics from a relational framework, instead of taking the household as a cohesive and single-minded unit, Kruczek-Aaron was able to elucidate potential gendered differences in the construction of identity.

Process and Presentation: Gender and Identity Construction

Some of the studies discussed herein draw upon the role of gender in the construction of identity. As just noted, Kruczek-Aaron’s study of the Smith household suggests that the women of the family may have been attempting to construct a different identity to outside observers through their ceramic choices than Gerrit Smith was in agreement with. His wife, Ann, having come from a family that instilled her with “fashionable taste” (Kruczek-Aaron, 2002:179), may have desired to construct a more upper-class identity that involved showing others one’s refinement and purchasing power, an identity more in line with her upbringing in one of Rochester’s founding families. Wall’s middle-class women of New York City, similarly, were trying to construct an identity in line with the middle-class values of piousness and domesticity, and the Redemptorist brothers were trying to construct a feminine identity that would reinforce their submissiveness, domesticity, and other values in line with their service to the church.

While these studies focused on extra-somatic means of identity construction, others have emphasized gendered identity construction through the body and personal adornment (see White and Beaudry, this volume). In a sense, such construction represents another scale at which gender organized society: the level of the individual. Thomas and Thomas (2004), for example, identify four layers of personal appearance—the body, against the body, clothing, and accessories—within which identity may be constructed. “These different layers of appearance can express different aspects of identity, and each layer may be more or less visible and comprehensible to different audiences. Furthermore, each of these layers functions somewhat differently in terms of the aspects of social identity that they communicate or reinforce” (Thomas and Thomas, 2004:107). Thomas and Thomas contend that three of these layers—the body, clothing, and accessories—are better suited to the construction of gender identity, the body, because gender assignments are often based on sexual differences (though for exceptions, cf. Whelan, 1991), and clothing and accessories because they can be worn in conformance or non-conformance with gender-based notions of appropriate dress and ornamentation. The effects of manipulating these layers in the construction of gender identity are shown through their study of African-American slaves at the Hermitage plantation in Tennessee during the first half of the nineteenth century (Thomas and Thomas, 2004). The work clothing provided to child slaves by their owners was uniform and of one piece, and thus served to mask gender differences. Even the gender-specific clothing worn by the adults was basic and shabby and therefore diminished individual identity, including gender identity (though certainly brought out class identity), which could have been emphasized through more individualistic clothing and accessories. Historical documentation of the former and archaeological evidence for the latter (such as buttons, pipe fragments, brooches, cane tips, parasol parts, and purse clips) indicate that these more individual expressions of self were used to personalize the self and reject the homogenized identity projected by their owners’ choices (Thomas and Thomas, 2004). The manipulation of these layers supports the notion of gender as an active and fluid construction, with part of its fluidity due to the fact that it can be imposed by oneself or by others.

Synthesis and Conclusion

With the growing understanding by the majority of the field that “social scientific research must take gender and women into account if it is to be academically credible” (Wylie, 1991:33), the historical archaeology of gender has found its footing in several quality studies. As would be expected, however, with any recently developed topic of study, it has been wrought with challenges. With regard to theory, the progress in overcoming these has been exponential in a brief amount of time. By quickly moving beyond finding men and women and predicting their material correlates to an understanding that gender structures society in several different ways at many different levels, historical archaeologists have been able to explore the means by which this process occurred historically, its interactions with other processes, such as class, ethnicity, and race, as well as the effects of and responses to these processes. What has become increasingly clear, however, is that the process of gender, its interactions, its effects, and responses to it are exceedingly complex and that our body of theory for such exploration, therefore, is probably still in its early stages. The field has only recently, for example, begun to conceive of gender as relational, to acknowledge its inextricability from other constructs, or to understand it can change or be changed as a facet of someone’s identity. Recognition of these ideas, however, demonstrates that historical archaeology has made significant theoretical advances in the study of gender and suggests that with consistent attention to gender, it will continue to do so.

Methodologically, the field has struggled to keep pace but continues to evolve. Even in moving past the attempt to establish material correlates, which it largely has, the interpretation of discovered artifacts still frequently reflects assignments of separate spheres instead of addressing, for example, who bought them versus who used them (and did more than one individual use them) and how. How were they used explicitly versus implicitly and why? How does the archaeological record reflect the way gender structured the individual, the household, and the community? Clearly these questions are more difficult to assess archaeologically, especially where base documentary evidence is sparse.

As the studies described herein have shown, though, ways to address them are beginning to be developed. By comparing the structuring by gender at multiple levels of society, variation between the levels can be observed, providing a more accurate picture of what gender does and how what it does is affected by, say, private or public settings or power relations. By employing a relational view of the household, discontinuities in the quality and cost between purchased material classes may indicate a conflict over consumptive display. In another example, by looking at differences in personal artifacts, such as the owner-provided versus self-purchased clothing of slaves, the struggle to impose (between self and other) particular class and gender identities may be seen. As complex as gender is with regard to processes, interactions, effects, and responses, the methods for its study will likely need to be equally complex and thus should prove frustrating for some time. In the end, however, the benefits of attending to gender as a structuring principle of society will outweigh methodological frustrations as it will elucidate broader social processes and how these work to impact societies, perhaps in a way that might be useful for effecting social change.