Corps de l’article

Archaeologists tend to ground their interpretations of past social structures in terms of male and female social roles. In particular, Inuit social scientific research has largely been conducted on the basis that the Inuit gender system traditionally comprised the complimentary pairing of men and women, each with distinct but mutually beneficial social roles (Briggs 1974; Guemple 1986, 1995; Hodgetts 2013; McGhee 1977). Archaeologists have frequently used diagnostic men’s and women’s tools, such as hunting implements or ulus (women’s knives) and qulliq (lamps), respectively, to “see” gender in the archaeological record, and to reveal the distribution of men’s and women’s spaces on Inuit sites (Hennebury 1999; Reinhardt 2002; Whitridge 1999, 2000).

While this model is appealing due to its relative simplicity, it is problematic in that it probably does not encompass the range of variation of Inuit gender that existed in pre-contact times and quietly persists today. Archaeologists have not yet effectively addressed a growing body of ethnographic evidence about nonbinary gender roles in Inuit society. This is largely due to the perceived difficulty of understanding complex gender ideologies when one is interpreting archaeological materials. In this article, I discuss the concept of Inuit nonbinary or “third-gender,” pointing toward avenues archaeologists might take to begin to represent a wider variety of gender expressions previously deemed archaeologically invisible. The primary objective of this article is to provide a critique of current approaches to Inuit gender archaeology and to begin to identify potential avenues to explore gender expression beyond a binary framework. It should be noted, however, that the suggested approaches are theoretical and must be tested and developed in future publications. This turn in archaeological research is essential not only to satisfy historical accuracy in our research but also to provide nonbinary, gender-fluid, or Two-Spirit Inuit living today with a concrete sense of past.

This work grows out of a desire to reconstruct a more coherent and culturally specific understanding of gender in the Inuit past on the basis that genders, identities, and concepts of personhood are socially constructed but nonetheless have tangible impacts on lived experiences and their associated materials. Instead of uncritically projecting binary gender onto the past, archaeologists must consult contemporary anthropological literature and critically engage with ethnography, oral history, and folklore in order to formulate ideas about what form past gender systems might have taken. From there, we can begin to consider how material culture conforms to, reflects, or influences gender.

In this article I provide a brief background of gender research in archaeology, outlining the broad trends that have preceded our current approaches to gender, discuss the notion of “third” (or nonbinary) gender in anthropology, and outline the anthropological literature pertaining to nonbinary gender in the context of Inuit culture. I then outline a variety of ways in which we can begin to look at pre-contact Inuit gender archaeologically. These ways should be applied cautiously with assiduous attention to regional variability among and between Inuit cultural identities and traditions throughout the Arctic.

The Archaeology of Gender

It is crucial to recognize that any archaeological research has at least two sets of actors: the subjects of study, as contextualized within the archaeological record, and the researchers, who are also situated historically and temporally. Although we tend to conceive of gender categories as static or inherent, they are actually subject to change over time. Despite the constructed and mutable nature of gender, these concepts have real implications for archaeological interpretation (Wylie 2007, 97).

In the shifting social circumstances underpinning the history of archaeology, the approaches archaeologists take to gender have changed. During the 1950s and 1960s, the Man the Hunter model, which sought to naturalize strict divisions of labour along binary gender lines, dominated archaeological discourse. This model positioned male hunting practices as the major influence on social, cultural, and biological progress. Early proponents of this view include George Bartholomew and Joseph Birdsell (1953), who attributed the evolutionary trajectory of early humans to the nuclear family structure, stating that this trajectory would not have taken place in the absence of stringent, dichotomized gender roles. Man the Hunter became explicitly solidified in 1968, when a symposium of the same name (Lee and DeVore 1968) was organized to examine male hunting behaviour in hunter-gatherer cultures.

In the 1970s reactions to this approach surfaced in the form of Woman the Gatherer. Proponents of this view (Slocum 1975; Zihlman 1978) emphasized the roles of women in hunter-gatherer society. This shift is associated with the emergent second-wave feminist movement, which emphasized female agency but still enforced the view that there were two distinct genders. Although this trend broadened the scope of archaeological literature beyond its previous androcentric focus, it did nothing to challenge the binary interpretive framework within the archaeology of gender.

In the 1980s archaeologists began to recognize the need for more nuanced studies of gender. Margaret Conkey and Janet Spector (1984) suggested that archaeological interpretations tend to be based on present norms that might not apply to the cultures we study. They challenged this normative application of gender constructs to past cultures, arguing that gender categories are not the same as biological sex but are cultural constructs administered haphazardly to archaeological data (5). Conversely, they contended that archaeology is used to substantiate a gender mythology (3) that naturalizes Western gender categories. In other words, by projecting our own gender categories onto the past, we create the illusion that they are naturally fixed.

“Third” or Nonbinary Gender

Nonbinary gender, gender fluidity, and blended gender roles are more common and widespread than archaeologists tend to recognize. There have for many years been recorded ethnographic accounts of what has often been referred to as “third” gender people among Indigenous North American groups (Blackwood 1984; Jacobs 1968; Lurie 1953; Williams 1992). Manifestations of third gender have also been recorded in other parts of the world, including South America, South Asia, and Polynesia (Herdt 1994; McMullin 2011; Nanda 1994; Williams 1992). The nature and significance of these gender categories varies from culture to culture, and several include multiple distinct manifestations of nonbinary gender. In fact, many authors have argued that the designation “third gender” should be abandoned due to cross-cultural variation of gender expression. The adjective “third” implies that gender naturally fits into a binary framework and that anything that falls outside that framework can be slotted into a singular “other” category. In light of this, nonbinary gender categories might be understood as a component of a “multiple gender system” (Blackwood 1984, 3), rather than as an extension of a naturalized gender binary.

In the early 1990s, as a reaction against the anthropological tendency to classify diverse Indigenous genders and sexualities as third, LGBTQ Indigenous scholars and activists coined the term Two-Spirit to describe Indigenous gender and sexual diversity in their own terms (Driskill et al. 2011a, 17). There has since emerged a body of literature on queer Indigeneity, whose authors have asserted queer Indigenous identities while simultaneously resisting restrictive white settler gender roles and categorizations (see, for example, Driskill et al. 2011b; Gilley 2006; Rifkin 2010). This literature has also distinguished Indigenous gender and sexual diversity from white settler notions of queerness (Finley 2011; Morgensen 2011).

Beyond a small handful of publications (Hollimon 1997, 2000, 2001; Looper 2002; Preston-Werner 2008), this discourse has rarely carried over into archaeological research. I argue that many archaeologists have fallen victim to naturalization narratives that equate gender with Western concepts of biological sex. We have predominantly been raised with a template that includes only men and women, and many of us have felt the pressure to fit seamlessly into these two categories. Only recently have other genders been introduced overtly into the mix. The effect is that these genders can appear to be new and particular to Western culture. While the specific modes of being trans, nonbinary, gender-fluid, or agender might have been conceived of in a primarily settler sphere and therefore do not capture the full range of Indigenous gender expression, the idea that genders do not exist in an immutable binary system seems to be ubiquitous.

The conceptualization of gender as a cultural system rather than as a rigid binary framework provides an understanding of identity and personhood as flexible and culturally specific (Fowler 2004; Sørensen 2000; Voss 2005). Judith Butler (1990, 179) argued that binary genders are “cultural fictions” that gain credibility through the repeated performance of these roles. Therefore, gender research in archaeology must be carried out with the understanding that different forms of identity exist in different cultures; there is no natural set of gender expressions that exist cross-culturally (Blackmore 2011, 78). Pre-contact Inuit relied on their own set of socially constructed and performed ideas of personhood.

Extensive work on Inuit nonbinary gender comes from the French-born Canadian anthropologist Bernard Saladin d’Anglure, who has focused most of his work on Canadian Inuit social categories and shamanism. Saladin d’Anglure (2005, 1) describes the Inuit nonbinary gender as a “third element which straddled the boundary between the two others,” contending that it mediates between male and female social roles. This conception of nonbinary gender is intimately linked with shamanism. Cross-culturally, shamans often fulfill a mediatory function in society, straddling the line between the earthly and the supernatural, the human and the animal, and, in this case, the male and the female (Williams 1992, 33; Hollimon 2001). Shamans usually do not gain their social position through heredity, but rather on the basis of their statuses as special people. It is therefore unsurprising that nonbinary individuals, both among the Inuit and across multiple Indigenous North American cultures, often act as shamans (Saladin d’Anglure 1986, 1992, 2005, 2006).

Role swapping, wherein male children are raised to perform traditional female roles and vice versa, is another common, more pragmatic manifestation of nonbinary gender. This usually occurs when gendered labour encounters a sex-ratio imbalance, which creates a need for individuals to assume roles not traditionally assigned to their biological sex (Saladin d’Anglure 2005). As Barbara Crass (2001, 111) stated, “a family with several daughters may decide to raise the next female infant as a male or vice versa.”

Harry Stewart (2002) similarly discussed the kipijuituq category of gender among contemporary Netsilik. While relatively little is understood about this social category, it is understood that kipijuituq are usually biological males raised as social females until they undergo a rite of passage, usually the killing of a seal or a polar bear, and become a social man (15). Kipijuituq are designated as such by their grandparents during infancy, on the basis of the child’s reactions to their grandparents’ speech (14). Alternatively, a person is automatically kipijuituq if their namesake is of another gender. Children are usually assigned names that belonged to recently deceased ancestors (Saladin d’Anglure 2005; Jenness [1922] 1970). According to Saladin d’Anglure, the child is sometimes socialized with the gender of the deceased rather than with the one that more traditionally corresponds to their biological sex. Diamond Jenness ([1922] 1970, 167) contended that name and gender are not closely connected and that most names can be applied to either men or women, thus exemplifying the fluidity of Inuit gender categories. Names are also not strongly gendered, and use of kinship terms is often based on the nature of relationships with a person’s namesake rather than on their biological sex (Crass 2001, 108). This naming system might reflect the link between Inuit gender and social identity; past nonbinary gender might therefore be effectively explored through linguistics.

Various Inuit stories reflect nonbinary gender categories or carry related themes that express fluidity and transformation of identities. Using mythology, we can incorporate Inuit cosmologies into archaeological analysis and provide a much-needed antidote to southern perspectives. Myths can also articulate ideas inherent in a culture that cannot be accessed through interviews. On the other hand, myths cannot be taken at face value; nor can we assume that they are unchanging. Inuit mythology is regionally varied and cannot be applied across all northern cultures. Therefore, the insights we reap from mythology must be taken with a grain of salt.

Grace Slwooko (in Saladin d’Anglure 2005, 135–37) recounted the story of a biological male who identifies primarily as a woman, wearing women’s clothing and carrying out traditional women’s work. When they are ostracized by some of the hunters, the Maker of All gives them a child: a baby whale. This whale brings other sea mammals to shore for the humans to hunt but is killed by hunters, who are punished for their carelessness. Saladin d’Anglure (2005, 137) contends that the whale represents this individual’s fluidity; their female aspect gives life while their male aspect brings a prosperous hunt.

Another story tells of a woman named Itijjuaq, who cannot scrape skins, sew, or have children but has an “understanding of things” (Saladin d’Anglure 2005, 139) and the ability to heal and who therefore becomes an angakok (shaman). In a creation myth recounted by Knud Rasmussen (1929, 252–53), two men emerge from the earth, wanting to populate the land. The penis of one man is split apart and becomes a vulva and they are thus able to procreate. These stories all suggest that pre-contact Inuit may have valued or accepted gender and sexual diversity.

For archaeologists who wish to reconstruct the past, some of the most valuable records of Inuit culture come from European and American explorers and ethnographers at the time of early contact. These accounts are particularly useful, since they capture some aspects of Arctic life ways that had not yet been altered dramatically via contact and might therefore (cautiously) be projected into the past. Unfortunately, those who penned them did so through settler lenses. Early accounts of northern cultures are therefore heavily impacted by European social norms, which render many identities invisible.

One example comes from Jenness ([1922] 1970), who largely skipped over manifestations of gender that he could not fit into his framework of binary gender. However, several things do stand out in his accounts. Jenness suggested that the Innuinnait kinship system was “more concerned with the nature of the relationship than with the sexes of the individuals” (84). He also noted that some Innuinnait men could sew and that women sometimes acted as hunters and sealers (88). He mentioned, for example, a woman named Milukkattak who often went sealing with the men; another, unnamed woman who was a prolific caribou hunter; and an adolescent girl who received hunting lessons from her stepfather. Through Jenness’s account, we see evidence of fluidity between gender categories.

Historical factors have complicated the visibility of nonbinary gender in contemporary research. One such factor was European colonialism and Christianization. With the arrival of Europeans, new values were imposed, and often forced, upon the Indigenous Peoples of North America. This process had different effects on different cultural groups but often involved the demonization of Indigenous religious beliefs and expressions of identity that were not in keeping with the accepted European social order.

White settler values dictated that it was sinful to deviate from prescribed male-female roles and treated nonbinary people as holy, or under the sway of the devil (Williams 1992, 31). Government officials and missionaries actively suppressed what they saw as deviant, uncivilized, and unnatural behaviours (177). Over time, through the introduction of religion and the implementation of new social values, nonbinary gender fell from its respected position in Indigenous cultures and was eventually viewed as a “social disorder” (185). While the impacts of white settler colonialism were differentially felt between different Indigenous North American cultures, suppression of Indigenous expressions of gender and sexuality was deeply entrenched in the colonial process, a reality often noted in queer Indigenous literature (Driskill et al. 2011b; Gilley 2006; Rifkin 2010; Smith 2010).

Frédéric Laugrand and Jarich Oosten (2010) challenge the perspective that settler colonialism had the power to destroy Inuit ideology, arguing that testimonies of Inuit Elders affirm that shamanic ideology still plays an integral role in Inuit culture. It is important to recognize that Inuit agency and culture were not wholly decimated by the influx of white settler values, but it would be naïve to argue that no loss occurred. In either case, material manifestations of angakkurniq (shamanism) and the angakok as a central figure within Inuit communities have all but disappeared as a result of the introduction of Christianity.

The Materiality of Nonbinary Gender

Nonbinary gender has been successfully investigated by various archaeologists in recent years across a number of cultural and temporal contexts (Hollimon 1997; Looper 2002; Preston-Werner 2008), but nevertheless continues to be an underexplored research area. While many archaeologists are familiar with ethnographic work that points toward more complex and varied gender systems, few are actually willing to bring this evidence into their research. To an extent, this reluctance is attributable to the archaeologists’ personal biases, particularly the internalized normalization of binary gender categories in Western culture, which permeates our interpretations and makes the presence of two distinct gender categories feel more natural and therefore more objective than the possibility of multiple categories, despite abundant evidence to the contrary. This view is largely based on the idea of biological reductionism, which frames people as concrete biological bodies (Butler 1990) from which the social aspects of personhood emerge. Because we tend to view biological sex as naturally binary (despite a significant portion of the population being born intersex and medically “corrected”), it follows that gender, which is often taken to be a sociocultural reflection of sex, should also take on a binary structure.

The second reason is more pragmatic in scope. Personal behaviours tied to identity can be difficult to interpret in the archaeological record because of their low visibility and high degree of variation. It is challenging to identify and interpret archaeological manifestations of gender because gender is complex and has varying manifestations in different contexts.

Traditional approaches to gender often involve “gendered” tools being uncritically slotted into singular functional categories that are taken to represent gender. While some tools can be generally understood as proxies for gendered activities, this type of approach must be applied carefully. When applied haphazardly, several problems emerge. First, tools are often understood as having a singular function, when they might actually serve multiple purposes. These tools might be passed between people and used by different genders. Furthermore, tools used for the activities of one gender might actually be produced by another, so their meanings might vary in different contexts. This type of artifact analysis also does not take variation in the performance of gender into account. As discussed, male and female roles were often swapped out of necessity.

It can be misleading for archaeologists to restrict their analyses to binary categories represented by two opposing sets of artifacts. If we approach the archaeological record with the notion that we are looking for two distinctive and homogenous genders, we might come to the conclusion that gender is not strongly represented, that it is only represented in specific contexts, or that archaeologically represented gender is confined to a highly restricted set of roles. However, if we approach material remains with the knowledge that we are examining a social system comprising individuals whose identity is formulated along multiple axes, and whose social categories are not necessarily analogous to dominant Western social categories, we can start to see identities played out in multiple aspects of domestic, economic, social, and ceremonial life.

One of the most basic units of archaeological analysis is the artifact, which has often been used as a stand-in for gender in archaeological research. However, the face-value assessment of artifacts as proxies for gendered individuals is problematic, since gender is complex and culturally varied, as outlined above. While it can be problematic to equate gender categories uncritically with categories of artifact, the study and comparative analysis of certain objects can be useful if approached with caution. One approach that could be used to transcend the male-female binary comes from Louise Senior (2000), who explored how craft production is affected when one gender starts producing materials that were formerly socially allocated to another gender. She argues that the form of the producer or the product will necessarily change (71), drawing a link between material culture, the body, and manifestations of gender. This idea that stylistic shifts can occur without extra-cultural changes might be useful for the problem of identifying nonbinary gender categories.

While Senior’s discussion is broad in scope and does not specifically focus on Inuit culture, it does suggest the possibility that anomalies in the material culture of a site might hint at social difference. Senior argues that we need detailed studies of gender and craft production in order to apply these models (81). Because a strong argument has been made that there is a complementarity of gender roles in Inuit culture and that certain tool types are generally associated with women and men, respectively, such as women’s ulus, or men’s knives, we should focus on these artifact types. When, ethnographically, do these gendered associations fall apart? Archaeologically, how much regional and temporal variation should we expect to see in these artifact types? And most importantly, does the form of a tool with a gendered association say anything about the gender of the user or the maker? Obviously, these are fraught considerations. If we take into account ethnographic evidence that tools were often shared by all genders as needed, it becomes clear that Senior’s argument can be applied directly to Inuit contexts. We might begin to ask to what extent these artifacts carry gendered meaning, how they were understood as gendered at their time of use, and whether finer-grained gendered meanings are associated with aspects of artifacts.

As a starting point, Robert McGhee (1977) famously argued that the materiality of pre-contact Inuit artifacts holds a symbolic significance. He discussed these aspects of pre-contact Inuit technology and material, arguing for strong conceptual associations between women, ivory, and sea, and between men, antler, and land. Although McGhee presented an oversimplified understanding of Inuit materials and belief that may not apply to all sites and assemblages as seamlessly as he suggested, this influential piece provides a good point from which to start forming associations between objects, materials, and gendered ideology. If certain materials hold gendered associations, it might be possible to identify artifacts with unusual combinations of material, artifact type, and gendered context. Christopher Trott (2006) argued that the polar bear, or nanuq, is a symbol of gender mediation as well as shamanism in Inuit culture. Pointing toward folklore, oral histories, and anthropological evidence, Trott contended that nanuuk are powerful mediators because they straddle the line between land and sea and are associated with a variety of men’s and women’s activities. Building upon McGhee’s approach, we see that this mediating role points to the potential for complex gender systematics to be expressed materially through artifactual evidence and symbolism.

Common within pre-contact Inuit collections are functional items that incorporate polar bear imagery into their forms. One example is an ivory polar bear effigy, which is likely a fragment of an ivory seal drag handle, found at Qariaraqyuk, a pre-contact winter village site in Nunavut excavated by Peter Whitridge in the 1990s (Figure 1). The polar bear is conceptually tied to angakkurniq (shamanism) and gender mediation but is applied to a sealing implement, which is associated with men’s activities. Further, if we accept McGhee’s hypothesis, the ivory from which the bear had been carved can be related back to Inuit concepts of femininity. Inscribed into the object is a V-shaped motif, which is also common in Inuit women’s forehead tattoos, providing an additional reference to Inuit femininity.

Figure 1

Ivory polar bear effigy and possible seal drag handle fragment with incised V-motif, found at Qariaraqyuk (PaJs-2)

Ivory polar bear effigy and possible seal drag handle fragment with incised V-motif, found at Qariaraqyuk (PaJs-2)

-> Voir la liste des figures

The mixture of gendered function and features hints that the gendered meaning of artifacts is not always simple or singular. The notion that complex ideas about gender could be expressed through artifacts is consistent with ethnographically documented attitudes toward Inuit social roles, which were seen as fluid complimentary parts of the same system (Crass 2000). In artifacts that exhibit mixing of gendered function and/or symbolism, we can see the merging of gender roles, and the mediatory spaces between them. These artifacts demonstrate that functional categories associated with economic activities ethnographically associated with men and women, respectively, are not enough to convey the breadth of gendered meaning that is actually written into the artifact.

Ethnographic examples of clothing styles might also provide insights into how we can gender the past beyond binary categories. Ethnographically, women’s parkas differ from men’s in a variety of ways, including design, trim, and cut, and the amaat, or hood, which, in a woman’s parka, is designed to hold an infant (Crass 2001, 109; Hall, Oakes, and Qimmiu’naaq 1995). However, gendered clothing and biological sex do not always align. Judy Hall and colleagues (1995, 52) note that children who are named after a deceased relative wear the clothing of that relative’s gender identity. Because angakkuit often had shifting gender identity, or because nonbinary individuals sometimes became angakkuit, shamanic dress often combined attributes of male and female clothing (Hall, Oakes, and Qimmiu’naaq 1995).

Although clothing usually does not preserve well in archaeological contexts (Crass 2001, 114), the merging of aspects of men’s and women’s clothing points to the possibility of artifacts that are an amalgamation of gendered features. In particular, with this question in mind, it would be interesting to analyze collections that have known associations with shamanic practice in order to determine whether ceremonial collections show any evidence of gender fluidity. This question might be answered through an examination of symbolism, motifs, and figurative art, with attention to the mixture of gendered elements and anything that might communicate themes of transformation.

This approach seems promising, given the fairly common mixing of gendered elements in contemporary Inuit art that deals with shamanism. Because Inuit art has changed and developed significantly over the last several hundred years, it would be inaccurate to say it directly or literally communicates anything about the pre-contact gender system. Nonetheless, it might hold kernels of the past in its portrayal of gendered themes. For example, both oral folklore and contemporary art incorporate strong themes of fluidity and transformation when dealing with angakkuit, and we should test whether these associations held true in the past.

In addition to artifact analysis, spatial analysis has proven to be a powerful means to examine gender in the northern archaeological record, thereby producing tremendous insight into past experiences and life histories. Places can be used as physical links to conceptual realities, thus providing us with hints of “the physical activities, meanings, and associations that people connect with them” (Hodgetts 2013, 8). Material culture reflects social production, and its depositional patterning is therefore intrinsically tied to social processes (Tilley 1989). By rigorously examining the spatiality of archaeological sites, we can see material manifestations of cultural meaning, identity-based perceptions, and social difference.

These studies exist at a variety of resolutions, ranging from fine-grained analysis of housing structures to course-grained regional analysis. Many of these approaches have involved the simple mapping of male-versus-female artifacts onto a site and taking for granted that the results reproduce discrete categories of gender. More nuanced approaches have creatively combined statistical approaches with community-based understandings of gender to produce more culturally accurate results. While much work examines manifestations of gender in households through spatial analysis, no studies in the Inuit context have attempted to transcend the male-female binary. The fluidity of space, as well as the fluidity of gender and an understanding of non-southern constructs of identity, must be factored into our approach to the spatial patterning of material remains.

Conkey (1991) suggested that the examination of gender should begin at sites of gender performance and interaction. According to this logic, households are exemplary sites of analysis, due to their centrality to life functions (Hennebury 1999, 21). There are multiple examples of household distribution being used to understand gender in the Arctic record. For example, Christine Hennebury (1999) examined gendered spaces in pre-contact Inuit housing structures, using k-means cluster analysis to reveal areas of gendered activity. She concluded that distinct gendered spaces do not exist and that there is considerable overlap and mixing of materials, while nonetheless arguing that this type of analysis could be useful for future assessment of gendered spaces in pre-contact Inuit housing. Similarly, Peter Whitridge (2000) used artifact correspondence analysis of a pre-contact winter house to reveal patterned redundancies in gendered movements within the structure.

Gregory Reinhardt (2002) rigorously examined gendered space in a well-preserved prehistoric house in Alaska. He argued that previous studies, which divided the house into two discrete gendered sides, had problematically reduced the visibility of females, despite the actual presence of at least two female bodies in the house (127). He furthermore stated that we need to consider “what we mean by sex-based ascriptions and…whether those ascriptions have any emic validity” (148), reiterating Conkey and Spector’s (1984) point that we need to be cautious about projecting our own constructs of gender into the past. Reinhardt’s assessment incorporates both statistical analysis and cultural understandings to provide a three-dimensional view of the space, positing that, on the basis of ethnographic accounts, men probably did most of their work outside the house while women worked inside, producing clothing and tools near their lamp and throwing scraps to the far side of the shelter (144). By incorporating ethnographic data, he reached a different, more fleshed-out conclusion. He concluded by observing that the presence of male-associated artifacts does not necessarily imply the presence of males, nor do they exclude the possibility of use by women (148).

This evidence suggests that, until we understand more about gender, we cannot understand the spatial patterning that human activities might yield. Nonbinary gender must be explored in more depth to develop sound hypotheses that also reflect Inuit gender ontology. To this end, we can look at spaces with the knowledge that we might not only be looking at men’s and women’s activities. In keeping an open mind with our investigations, we might start to see new patterns of gender performance in our data, patterns that might otherwise go undetected.

Of course, throughout this process, we must also bear in mind the pitfalls of applying ethnography in these contexts. As mentioned, accounts of gender in early ethnography are often dubious due to the lenses of the ethnographers themselves, who seem to have ignored or misunderstood gender expressions that did not fit into a European binary framework. Also problematic are later ethnographies that do account for gender variation; they are much further removed from pre-contact Inuit culture and therefore might say little about how gender functioned prior to contact. Instead of applying these written works uncritically, we must test their claims against archaeological data and consult Inuit, whose testimony remains an invaluable if underutilized resource for interpreting the past.

To explore nonbinary gender spatially, one might examine kariyit, or ritual houses. These structures were common throughout the pre-contact period, serving as spaces for feasting, community games and activities, and many shamanic rituals and performances. Because the anthropological literature ties Inuit shamanism to gender fluidity, the karigi might be conceived of as a site for gender performance.

Ethnographically, kariyit have been widely documented throughout the Arctic (Hawkes 1916, 59; Rasmussen 1929, 227). Although a few archaeologists recognized these structures beginning in the late 1960s (Lutz 1973; VanStone 1968), their identification did not become commonplace until the late 1970s, when Allen P. McCartney identified a number of possible kariyit on the basis of their paved floors, lack of sleeping platforms, and artifact assemblages (1977, 167; 1979, 288). Since then, multiple kariyit have been identified on the basis of symbolic attributes, such as whale bone construction, central pits, lack of sleeping platforms or kitchens, and distinct artifact assemblages (Patton and Savelle 2006; Savelle 1997, 2002; Savelle and Habu 2004; Sheehan 1997; Taylor 1990). By re-examining these ceremonial spaces, while taking into account the ethnographic literature on Inuit nonbinary gender, archaeologists might begin to develop a better understanding of how kariyit were used and how their spatial patterning and artifact assemblages might reflect aspects of the gender identities of angakkuit.

Of course, this work must be done cautiously for several reasons: these spaces are complex and often represent the coming together of community members of all genders; their functions varied throughout the Arctic; and ethnographic records of ceremonies are often inaccurate or piecemeal because white ethnographers were often not invited to see or participate in the ceremonies or, if they were, they likely misunderstood aspects of what they were seeing. While many aspects of pre-contact Inuit ceremonialism will likely never be understood by archaeologists, more detailed examination of these spaces is warranted, in large part because they were part of daily life for many people living in the Arctic prior to contact. While it is difficult to say what these studies might tell us about gender, they hold intriguing potential for examination of gendered performance and identities.

Burials and mortuary practice can also provide insight into gender through archaeological analysis. Because artifacts in burial contexts are placed in association with human remains that can often be sexed, it is possible to examine ways in which gender is enacted through the body, as well as the complex relationship between sex and gender (Crass 2001, 108). The utility of burials for examining nonbinary gender has already been demonstrated by Sandra Hollimon (1997), who explored archaeological evidence of nonbinary gender among the Chumash in California. She cited two individuals who were biologically sexed as male but had spinal wear associated with traditional women’s work and were buried with women’s artifacts. She concluded that these individuals probably belonged to the social category Aqi. Aqi were members of an undertaking guild that included biological men who took on feminine roles, men who had sex with other men, men without children, celibates, and postmenopausal women.

Several problems are inherent in the use of mortuary contexts to reconstruct social structures. First, Inuit burials are rarely complete, having been stripped for utilitarian purposes over time (Crass 2001, 113). An analysis of grave goods might therefore exclude a large portion of the grave’s original contents. On the other hand, because items might have been added to graves during prosperous times, portions of burial assemblages might not actually be associated with the original burial (Crass 2001, 115). Additionally, collections of artifacts found in burials cannot be taken at face value to denote gender. As stated above, artifacts cannot always be assigned to gender in a clear-cut way. Variations in the use of artifacts and in cultural performances of gender complicate the gendered categorization of artifacts. Moreover, graves are usually highly symbolic, ritualized spaces, and grave goods often take on meanings separate from their utilitarian purposes and assumed gender associations. It is therefore important to gain an understanding of the meanings of grave goods before we can start to reconstruct social difference on that basis. To do so practically, it is necessary to have strong cultural understanding and large sample sizes.

The question of ethics is the most glaring issue in the use of mortuary analysis to understand gender systems. It should go without saying that archaeologists are generally not permitted to excavate Inuit graves, and the past exhumation of human remains for the sole purpose of publishing academic papers was a gross example of archaeological malpractice. Because laws protecting Indigenous graves were not put into place until long after Europeans arrived in Arctic Canada, many Inuit graves have unfortunately been excavated and stripped by archaeologists, with much of the burial data remaining available for study, though variable in quality.

Crass collected data from multiple sources and compiled an extensive body of information on Inuit burials. She took a sample of burials that were complete enough for analysis and categorized them on the basis of their gender associations. She then cross-referenced the data with the biological sex of the associated individual when possible. Crass (1998) concluded that gender vaguely correlates with biological sex in Inuit burials, but the correlation is weak. On the basis of these results, Crass (2001, 115) argued for the existence of “a very complex gender ideology that is best described as fluid.” While these results do not speak specifically to manifestations of gender, they open doors to new and tangible areas of gender research. At least Crass (1998) has produced a large body of data that can be worked and reworked as new ethnographic understandings of gender emerge. Of course, the availability of this information does not mean that archaeologists should use it without extensive community consultation.

Conclusion

In archaeological investigation of nonbinary gender, we must keep in mind that the Inuit gender system was probably fluid, and that material evidence cannot be uncritically used as gendered proxies. Instead, our goal should be to reconstruct the complexity of pre-contact Inuit gender ideology. Through the use of multiple lines of evidence, I maintain that complex gender ontologies could become accessible to archaeologists in the future. To bring about that future, archaeologists must work to gain a deeper understanding of gender in its own cultural context and slowly build an understanding of past gender systems, rather than projecting their own constructs of gender onto the material record.

What I have provided thus far is an overview, which defines a complex problem of identity in the northern archaeological record, a set of constraints and challenges, and some approaches that might be taken to grapple with these problems. While the above approaches to nonbinary gender by no means present a fully developed archaeology of nonbinary gender in pre-contact Inuit contexts, they poke holes in the way archaeologists currently explore Inuit gender, while presenting potential starting points for future research. This research would have implications for broader questions of identity in the archaeological record; by examining multifarious datasets, archaeologists can transcend taken-for-granted constructs that are more reflective of our cultural norms than of the groups we study.

By recognizing nonbinary genders in archaeological research, it will be possible to improve not only the cultural and historical accuracy of such research but also the circumstances of nonbinary, gender-fluid, and gender-nonconforming Inuit living today. Archaeological studies of nonbinary gender allow people who do not identify with binary gender roles to see themselves as grounded in an inclusive history and find support through a sense of past. While non-normative gender identity can be isolating and cause people to feel out of place, misunderstood, or stigmatized, recognizing that these roles have a deep continuity can give people a connection with history and a sense of connection to their cultural past. This is especially important for people living in small communities, who might not have immediate support systems in place and who often end up feeling more isolated as a result. By including a greater breadth of actors and roles in our interpretations of the past, archaeologists may be able to contribute to a sense of belonging for individuals who lack access to a living community that can relate to their identity and experiences.