Article body


In the winter of 1922-1923, anthropologist Therkel Mathiassen was forced to stay on Southampton Island due to adverse ice conditions. During this period he was able to observe and record in some detail a series of shamanistic seances performed in the community. The most readily available accounts of these seances are published by the leader of the expedition, Knud Rasmussen, who included excerpts from Mathiassen’s diary in the published monographs of the expedition (Rasmussen 1929, 1930). Neither Rasmussen nor Mathiassen provide adequate contextual background for these sessions, primarily because the Inuit blamed many of their problems on Mathiassen and his activities in excavating archaeological sites on the island (Oosten 1984: 378). One of the most striking series of seances focuses around the attempt to cure a woman called Nanuraq from various persistent ailments. During the sessions Nanuraq confessed to a number of transgressions[1] and through these was to be cured of her problems. Despite these efforts Nanuraq persisted in her complaints, but the shamanistic activities ceased, probably because, as Oosten (ibid.) has suggested, the real focus of the activities was on the annual “New Year’s”[2] ceremonies and not particularly on Nanuraq’s problems.

The shaman’s performances highlighted the cause of Nanuraq’s illness in terms of ritual transgressions. Mathiassen notes, however, that the shaman himself attributed her problems to the fact that Nanuraq had beaten her little daughter so much that her arm had turned bad (Rasmussen 1930: 111). Already there are two explanations in place for this illness, both of which have been discussed and elaborated on in subsequent anthropological literature. Oddly enough, Nanuraq’s own explanation seems to have been ignored by the investigators. “Nanoraq says that since the bear was shot she had been afraid more and more each day that passed, and at last she fell ill. The skin, head and legs of a bear must not be taken home at once, but must first be cached; the result of this is that half of the bear’s skin has now been eaten by dogs” (Rasmussen 1930: 108). This statement makes it clear that a polar bear had been shot by members of the camp, however, the proper restrictions had not been observed in bringing in the body parts and as a result the carcass had been desecrated by the dogs. Oosten (1984: 387) is the only investigator to have picked up on this point, explaining that “Nanoraq’s name means ‘polar bear skin’, and it is clear that she felt related to the polar bear. This is not surprising when we consider the great importance of the namesake relationship in Inuit society and in particular the important symbolism of the bear.”

Oosten fails to elaborate any further, but I believe that this small anecdote takes us to the heart of a number of critical issues dealing with the relationship between women and bears among the Inuit, and in a more general sense the puzzling problems of the nature:culture relationship as articulated through the categories of gender. Oosten points in the right direction when he invokes the namesake relationship for, as Saladin d’Anglure (2001: 107) has pointed out, those people with animal names bear a special relationship to the animal and must be very cautious in observing the taboos associated with that animal.

The incident described by Mathiassen and analysed by Oosten points to the intersection of two metaphorical frameworks within Inuit cosmology: bear symbolism and gender categories. The literature on Inuit symbolism is extensive in both of these areas, but only Saladin d’Anglure (1986) has attempted to show the links between the two. In this paper, I will examine the existing discussions within each of these areas and then develop an alternative hypothesis about the relationship between these two metaphorical fields that will further bring into question our understanding of Inuit cultural categories.

The metaphor of the bear

The polar bear (nanuq) (Ursus maritimus Phipps) takes on a symbolic significance among Inuit far beyond its importance in their diet. In one of the earliest (1906), and only, theoretical considerations of Inuit society, Marcel Mauss (1979) pointed to the fundamental division between land and sea in Inuit cosmology, clearly reflected in the taboos which separated products and activities associated with the sea with those from the land. This is most easily seen in the linguistic distinction between caribou fat (tunu) and seal fat (uqsuq) which parallels the distinction between caribou skin (amiq) and seal skin (qisik). The linguistic categories merely reflect a much deeper, broader and more systematic set of rules that prohibit (for example) cooking and eating caribou and seal meat in the same pot or on the same day, or working caribou skins while living on the sea ice. The polar bear has the peculiar property of being an animal that lives and hunts primarily on both the land-fast ice and pack ice, while being able to travel overland. In the fall and spring, bears hunt seals on the land-fast ice by attacking them through their breathing holes while also hunting them from floating pack ice, and in the summer bears move onto the ice flows to continue seal hunting in the open water. Randa (1986a: 77) points out that while bears can run up to 25 km/h, they generally do not hunt caribou. Although polar bear and caribou habitats overlap, and bears will kill caribou opportunistically, they focus most of their hunting activity on the sea ice (ibid.: 90). In water, bears can only travel 6 to 10 km/h and are thus also unable to chase seals. Bears use a number of techniques to hunt seals; from breathing hole hunting to stalking seals sleeping on the ice, their preferred and most efficient hunting technique is from the floe edge (ibid.: 96-107) To follow Mary Douglas’ (1966) line of argument, bears are thus categorically ambivalent and thus logically potentially sources of “pollution.”

Graburn (1976) has noted that in Inuit art the nalunaikkutait or distinctive features of the polar bear are the canines, the claws as well as the tongue. In contemporary sculpture and prints, bears are often portrayed either standing alone or hunting seals, but when faced with a non-human adversary they are almost always portrayed battling walrus (ibid.). For the Inuit, walrus are considered the only formidable opponent for polar bears. The distinctive features of the polar bear noted by the Inuit should warn the reader of any ethnocentric bias created by classifying polar bears with other types of bears where, in Western terms, the distinctive feature is the bear’s whiteness. The bear constitutes a distinctive category of its own among Inuit. While other bears are present in some Inuit regions (black or brown bears in the east and grizzly bears in the west), they are marked from the polar bear by their terrestrial habitat, as animals that do not move across the boundary between land and sea.

Most important is the fact that bears exhibit a direct parallel with human activity. Here we begin to note the dichotomization of the bear in terms of the sexual division of labour. Bears, like men, are primarily hunters and more particularly seal hunters. Like humans, bears hunt seals on the ice through breathing hole techniques. Indeed Randa (1986a: 148-149) has suggested that Inuit learned breathing hole hunting techniques through their observations of bear behaviour. In an early analysis of bear symbolism, Bernard Saladin d’Anglure (1980) argued that bears were the archetypal symbol of male power which appeared in three forms:

  1. Bears are the greatest opponent for single male hunters and as such are the most prized prey for men. Bears are such formidable opponents that they were normally hunted with the help of dog teams and with the aid of other men. Exceptional value was placed by men on hunting bears, and in some senses one could not be a “real man” (angutimmarik) without having captured a bear.

  2. As hunters themselves, bears constituted the primary model for male hunting activities. Bears thus represented male hunting prowess at its best.

  3. Bears appeared frequently as shaman helpers and since it was thought that most shamans were men, they were thus primarily men’s helpers.

These conclusions accorded well with the available data and the thinking of the time, despite the obvious androcentrism of their formulation. But this formulation overlooks another very important part of the bear’s annual cycle. Like humans, bears built dens in the winter out of the snow that are similar in shape, size and function to igloos. The dens are primarily the spaces of female bears who give birth to their young in the dens and reappear in March to begin hunting on the ice. In the same way, igloos are primarily women’s spaces and are occupied by the Inuit in the darkest part of the winter. The literature is clear that igloos are women’s spaces and are metaphorically parallel to the uterus and the vagina (Saladin d’Anglure 1977a).

The hunting activities of bears directly parallel those of humans, thus providing a critical juncture between the natural world and the human world that allow humans to think about their own cultural constructions. This can be seen in the taboos surrounding the killing of a bear that indicate that the bear corpse receives death rites comparable to those for humans. A female skin must be hung in the igloo for three days with the appropriate women’s instruments (ulu [woman’s semi-lunar knife], qulliq [lamp], sewing thread and needle), as well as the bear’s bladder and spleen, laid alongside, and a male skin is hung in the igloo for three days with men’s knives (savik) and hunting implements laid alongside (Rasmussen 1929: 188). The Inuit also note a parallel between the flesh of humans and the flesh of bears such that cannibals among Inuit are prohibited from eating polar bear meat (Saladin d’Anglure 1994 [1989]: 184).

There is thus plenty of evidence for articulating metaphorical parallels between human activity and bears, which the Inuit are keen to exploit. We already have begun to indicate that the parallels not only apply to men’s lives, but to women’s as well. In cosmological terms, bears act as the dog team for the sled of the “Man in the Moon” (Aningat) and inhabit his sleeping platform. Aningat represents both male sexual virility and hunting prowess and is also one of the primary sources of shamanic insight. The bears that occupy his sleeping platform become shaman familiars. Bears are thus centrally located in the primordial male symbol among the Inuit. At the same time bears also protect the sleeping platform of the “Mother of Sea Beasts” (Sedna[3]) where she keeps sea mammals when withholding them from humans, usually due to infractions of the taboos surrounding menstruation. One must also note that when Sedna’s father comes to join her at the bottom of the sea, he wraps himself in a polar bear skin and lays down on the sea shore (the fluctuating boundary between land and sea). The combined ambivalence of the sea shore and the bear skin transport her father into the “spirit” world at the bottom of the sea. What is key to note here is that in the primary opposition established in Inuit cosmology, the bear is separately situated on both sides of the boundary (Figure 1).

In the third spiritual realm, that of Narsuk, the bear also appears as an important operator. Parallel to the changes in size from giant, to human, to dwarf that take place in the story the bear appears alternatively as a bear, a fox, and as a lemming depending on visual perspective and relative size. In the sequence of creation stories outlined by Saladin d’Anglure (1991), bear, fox and lemming are among the original animals prior to the creation of caribou and sea mammals. The fox is also a categorically ambivalent creature as it occupies the sea shore and scavenges on the carcasses left by polar bears[4]. To the extent that these three animals can act as symbolic substitutes for each other, the Inuit are clearly exploiting the parallelism in the classificatory ambiguities of these three animals, transforming it into a distinction based on size.

In two publications, Randa (1986a, 1986b) has intensively examined polar bears from both a zoological and cosmological point of view. Randa’s zoological evidence provides a scientific basis for the symbolic expressions of the Inuit. Randa is concerned primarily with the classification of bears within Inuit thought and concludes that bears operate as the mediator between the land:sea, hunter:hunted, and animal:human dichotomies in Inuit thought. He concludes: “Pour revenir à l’ours polaire, s’il a si fortement frappé l’imaginaire des Inuit, c’est en raison de sa tendance naturelle à chevaucher la frontière des catégories contrastives construites socialement[5] (Randa 1986b: 168).

Models of gender


Briggs (1974) has examined the evidence for patriarchy and explored the arguments surrounding the equality and inequality of the sexes among the Inuit. She found substantial evidence to support both positions, but none that is conclusive. Briggs concludes that male-female relations among the Inuit are fundamentally ambivalent due to the flexibility of Inuit social forms that makes Inuit society so adaptable to the variable ecological conditions in the Arctic. While it seems to me that this flexibility is what needs to be explained rather than constituting an explanation, I think her focus on ambivalence is important. At this level the symbolic ambivalence of the bear thus constitutes an excellent metaphor for male-female relations. To see this, one needs to look more closely at how male-female relations are constructed in Inuit society.

An important aspect of Briggs’ discussion is the cases she notes where women perform what are usually understood as men’s activities, and where men perform women’s activities Briggs’ argument points to the fact that the division of labour may be qualified by circumstances. For example, in a family that has had a series of girls, one of them will be chosen by the father to learn male hunting skills and thus brought up as a boy (ibid.). Thus, while there may be a strict division of labour, the people who perform that labour are not so strictly defined. For Briggs, this again speaks to a fundamental flexibility in Inuit society that also forms the core of the gender system as well. People may move between the gender role categories, where circumstances dictate.


Saladin d’Anglure (1977b) took up the challenge set forth in Brigg’s article and argued that violence against women, and consequently patriarchy, was endemic in Inuit thought and practice. As evidence he cited the myth where the original two humans were males and, desiring to reproduce one of the men became pregnant. When the child was ready to be born the other male recited some magic words that caused the penis to split and turn into a vulva whereupon the child was born. In addition, Saladin d’Anglure points to the Arnaqpaktuq myth where the foetus of a battered wife dies and its name soul moves through a series of animals until finally as a seal it is captured by the mother’s own brother and enters into her brother’s wife to be born as a son. For Saladin d’Anglure, this evidence establishes a fundamental patriarchy among the Inuit.

Guérin (1982) challenged this argument by questioning the textual evidence of the myth of first humans, and asking if the text said inuk (‘human’) rather than anguti (‘male’) would this not suggest an androgynous origin for humans thus qualifying the strict patriarchy of the text? Saladin d’Anglure (1986) has replied by pointing out that the original Inuktitut text does indeed say anguti (male) and thus the argument holds. Guérin also challenged the interpretation of the Arnaqpaktuq myth by pointing to the variations of the myth across the arctic, only some of which begin with a battered wife while many others begin with women, children or men. Furthermore, she questioned the arguments based on the prevalence of female infanticide (especially among the Netsilik Inuit) as indicating a fundamental preference for boys as opposed to girls. She examined the material surrounding the spirits of children who either die through abortion, are miscarried or are killed shortly after birth. In Greenland, these spirits are especially dangerous to hunters and can bring bad weather. Shamans must struggle to capture and kill them. On the other hand in Alaska, if a hunter can capture the spirit of the dead child and use the dried corpse as a hunting charm, then these spirits can be very beneficial to the hunters. Guérin thus provides evidence that the rigid patriarchy proposed by Saladin d’Anglure might be qualified, and might at least be ambivalent in the terms suggested by Briggs.

Ambivalence through birth and names

Saladin d’Anglure (1986) has re-evaluated the evidence by asking what the contexts for such ambivalence might be, and in so doing has shifted the argument to different grounds. The basis for such gender role ambivalence may not be based on pragmatic circumstances but on fundamental properties of the Inuit understanding of gender. This is well illustrated in the cover picture of the 1982 edition of Equinox magazine which shows a woman hunter from Igloolik. His point of departure has been the naming system whereby children, through the inheritance of names, take on the social gender of the namesake despite their evident biological attributes. Each person is made up of four elements: the blood from the mother’s blood, the bones from the father’s sperm, the flesh from the game animals, and the name soul from a recently deceased person.

When a person dies, the name soul rises up out of the body and searches out a pregnant women, into whom they enter and form a new human being. The foetus then has a certain amount of volition as to what sex it may choose to be when born. Although the data is difficult to assess, it appears that foetuses are generally thought of as male, but that as the child leaves the uterus (understood metaphorically as an igloo), it can choose either male (knife or harpoon) or female (ulu or qulliq) implements and thus determine its sex. During the birth process, a child may change its sex and the penis is absorbed internally to become a vulva and the foetus becomes a girl (sipiniq). While it is logically possible for a female foetus’ clitoris to grow into a penis, this seems to happen so infrequently that none of my informants could recall a case. Saladin d’Anglure concludes that approximately 2% of the population are sipiniq. At birth the child is socialised according to the gender of the person who last held the name and not according the their biological sexual characteristics. This socialisation is indicated through the clothes the child wears (there are distinctive male and female outfits), the length and style of hair, and the tasks the child is trained to perform. Thus approximately 15% of biological girls are acculturated as boys and similarly for biological boys.

This is further complicated by the fact that a child may receive multiple names (usually four or five in the Igloolik/North Baffin district, but this may go as high as 11 names) from ancestors of different genders and thus have multiple genders according to which name is in use. Furthermore, more than one child may share the same name, thus creating a system of avvariit or “halves”[6]. Saladin d’Anglure has pointed to the fact that a combination of circumstances is usually required before a complete gender transference takes place: multiple names of the opposite gender and circumstances such as those outlined by Briggs. In other cases, partial transformations may occur: the child will be dressed as a boy one day, a girl the next, the child may be split so that half of their hair is long and the other short, and their clothes are designed with a male and female side.

At puberty[7], the child’s gender roles were realigned to their biological sex[8]. While I was in Arctic Bay in 1981, I had asked a young 9 years old girl to behave more appropriately (as a girl) like her 11 years old female cousin. I was promptly informed that the cousin was not a girl but a boy. With great disdain, the cousin replied that she had not been a boy for two years. For girls socialised as boys, their first menstruation was usually marked by them visiting each house where the occupants declared what a great hunter the child was and how s/he had captured such large game as walrus or whales. After this, the child’s clothes were changed and she was taught to sew. For boys the transformation of gender occurred when they captured and distributed their first game. As Saladin d’Anglure has noted, the first game need not be large game, but small game such as rabbits, ground squirrels and birds may be substituted, and praised by the senior people as if they were large game.

This cursory summary of the evidence suggests that Inuit understandings of gender are constructed in a markedly different way from those in the West. Firstly, people can and do change their gender over the course of their lives. Secondly, a single person may in fact have multiple genders (given that they have a number of names). Thirdly, gender may be situationally defined. One of the consequences of gender derived from names is that the children are in fact the ancestors returned to the living. As an example, I will point to a situation outside of Arctic Bay where I was hunting in the spring of 1981 with a 55 years old man. This man was a well-established hunter in the community, had raised and provided for a family and as far as I was concerned was clearly male. Three younger men in their late 20s came to our camp one evening and immediately began making ribald sexual jokes with this man, asking to touch his genitals. I soon discovered that the older man had received his name from a woman and the younger men had received their names from that woman’s husband, and thus they were treating him (jokingly) as their wife. Situationally, he had become a woman.

Further evidence for ambivalent gender relations can be found in Inuit cosmology. In the myth of human origins and the Arnaqpaktuq myths, already presented, we see further evidence of changing genders. In addition, a careful analysis of shamanism has shown firstly that not only is the proportion of shamans to the general population much higher than anticipated but that there were also many more female shamans than had been predicted (Saladin d’Anglure 1988). Saladin d’Anglure has further shown that shamanism is linked to the naming complex and that those who become shamans are more likely to have been transgendered. Much more important has been the analysis of ritual activity where significant transvestite transformations take place culminating in the exchange of spouses. Shamanism thus provides a further field for gender transformation among the Inuit.

A “third sex”

The overlapping and variable gender relations in Inuit society have led Saladin d’Anglure (1986, 1988, 1992, 1994 [1989], 2005) to conclude that these man-woman/woman-man figures at birth, in childhood, and throughout shamanism constitute a separate sociological category, the “third sex,” which is neither male nor female, and both male and female at the same time. As adults, Inuit men and women fulfil the gender roles assigned to their sex, although within specified contexts those who have had differential upbringings will also perform the roles of the opposite sex. It is mainly within the context of kinship references that the ambiguities will remain, and I have argued elsewhere (Trott 1995) that the significance of this lies in the fact that it creates a structurally ordered set of ongoing relations between people and land rather than creating an intermediary gender.

Fulton and Anderson (1992) provide a succinct review of the debates that have occurred around the presence of alternate gender formations among North American Aboriginal societies. While much of the discussion has been hampered by religious and cultural biases from the West, and heavily implicated in psychoanalytic understandings of dysfunction and perversion, they propose a different approach that would place people performing alternate gender roles in a sacerdotal position, working through the liminal situation of birth, death and other transitions in societies. This approach comes very close to the more recent formulation of the notion of a “third gender.” Saladin d’Anglure’s work, in particular, has been used as evidence in the much wider debates over the construction of a third gender in social thought, especially by Herdt (1994) and Roscoe (1998) (although Roscoe complicates the issue by introducing a fourth gender distinguishing biological males who take on the roles of women and biological women who take on the roles of men). Herdt’s book compiles extensive historic and ethnographic evidence to demonstrate that even where there may have been no conscious distinctive categories as such, there were people who fulfilled and combined male and female roles in a wide variety of ways. The argument is compelling and the evidence is strong that there are in fact many cases where an additional category of gender is required beyond male and female. But this does not seem to entirely address the issue among Inuit where it is not so much a question of categories, but the fact that gender is situational and contextual ; the same persons can move between genders at different points in their lives and in different social situations.

Linking bears and gender

In 1982, I was presented an ethnographic puzzle in the form of a small carving with a bear’s head placed centrally within a stylised representation of a vulva. My informants had told me that the carving was “all about women,” and given Saladin d’Anglure’s male oriented interpretation of bears I sought to elaborate in what sense bears could also “stand for” women. At that time, I pointed out that while men stood in a relation of opposition to bears, women stood in an homologous relation to the animal (Trott 1982). While men stood in danger of being hunted and killed by bears through the instability of the predator-prey relation, women stood in danger of transforming into bears through the relationship between their reproductive cycle and the life cycle of the bear. Insofar as women’s bodies act as passages through which human reproduction is effected, they become transformational operators in the same way that bears, through their ability to cross the land/sea boundary become operators in the symbolic realm.

In the first paper where he applied the concept of the “third sex” to the Inuit, Saladin d’Anglure (1986) proposed that an axis mundi emerges along the connections between the changeable sex of the foetus (sipiniq), the sexual variability among children and the transformations that take place at puberty, and the sexual ambiguity of shamans. Within the Lévi-Straussian structuralist framework that forms the basis of Saladin d’Anglure’s theory, this is a logical conclusion. The primary opposition between male and female is mediated and transcended by a third category that resolves the opposition given in nature. The mediation of the third sex within the gender system appears coded in animal terms as the bear. In the same way that the bear transverses the opposition between land and sea animals, the third sex mediates gender relations, and thus the two appear as homologies of one another. From this perspective, bears are then the most appropriate animal to be used as shaman familiars. Graphically one can represent this position as in Figure 2.

The bear is therefore also the third sex, both male and female, and neither male nor female. This is a substantial argument that we need to take very seriously. Nevertheless, it overlooks the rather important point that men and women each stand in a fundamentally different relationship to bears. One of the striking elements is that when one discusses bears with men and women, one encounters very different responses. Women inevitably say that bears are iqsinaqtuq, (‘frightening’), and give a shudder to express the fear. One would not be surprised to find such a response as indeed bears are very frightening and can be dangerous, but the similarity of answers from different women suggests a standardised cultural response from them. On the other hand, men almost inevitably became animated and excited when discussing bears and would enter into stories of various risky adventures in capturing bears (usually other hunters adventures as it is inappropriate to talk about one’s own prowess). While men recognise and articulate the dangers of hunting bears, they will quickly refer to successful hunts especially by elders. For example, there is a well known story in Arctic Bay about Ujukuluk, who, when a bear got between him and the dog sled (where his rifle was located), jumped on the back of the bear and garrotted it with his harpoon line. In addition, the men would talk about the polar bears ability to hunt seals, killing them with a single blow of the paw, and then eating only the blubber and leaving the remainder of the carcass. At least on the emotional level, there were very different responses.

Mythologically, Randa (1986a) has shown that bears are very prominent and figure primarily in contexts where social conflicts cannot be resolved. In these cases, one of the partners will transform into a bear and seek vengeance on the other partner or on the other members of the social group. It must be noted that in myth it is almost always women who transform into bears while men will marry bears who fluctuate on the border between bears and women. One of the well known stories concerns a man who had three daughters and when the eldest reached the age to be married, he shut her up in an igloo alone. The girl gradually transformed into a bear and broke out of the hut. The father with his wife and two younger daughters gave chase but the bear turned on them and ate the father. The mother and two sisters protected themselves by laying out skin stretchers in a circle and sitting within. The bear sniffed at the ring of skin stretchers and then wandered off. It is thus the women’s ability to construct a (circular) boundary out of distinctively women’s tools that protects the family. The story explores both the reality that women can transform into bears while at the same time maintaining that it is the women who must construct and maintain the boundary between humans and bears. There is a clear association here between a women’s reproductive cycle, eligibility for marriage, and the creation and manipulation of boundaries that permeates all the stories about bears and women.

Randa (1986a: 275-280) notes further that bears are also frequently adopted by women to replace children that have been lost. Not only does this appear in the corpus of myths, but stories from my informants in Arctic Bay indicated that bear cubs were often adopted in practice. This would suggest that women can also stand in a mothering, nurturing, relationship to bears as well.

Bears do not therefore collapse the categories of maleness and femaleness but maintain the distinctiveness of the relationship between the two. There is no logic of synthesis or mediation here, but rather the attempt to hold in place the terms of the opposition while seeing them as related. The same argument can be applied to the categories of gender. I am not convinced therefore that a third sex exists among the Inuit and the symbolism of the bear parallels such a conclusion. Indeed, I would argue that given the ambiguities of gender in Inuit society, the gender of each person only becomes evident as men and women construct and demonstrate their gender through the practical activities they undertake. Thus in the social division of labour, one does not hunt because one is a man, rather one’s hunting ability demonstrates that one is indeed a man.

The significance of marking a boy’s first catch is that the boy has demonstrated his maleness and is thus in the position to give away his first game. For a parallel with girls, first menstruation is clearly marked by isolation (as seen in the myth of the girl transforming into a bear above) and by tattooing. As the myth shows, though, first menstruation is the process of transformation, possibly into a woman, possibly into a bear. Tattooing tries to hold the girl onto the human side of the boundary through the parallels between the girl and the woman in the sun (Saladin d’Anglure 1991) and the use of soot to mark boundaries (see Fienup-Riordan 1994). Only when a girl produces her first child does she become “fixed” as woman, and, parallel to the boy renouncing his first game, the girl must renounce her first child to the community through adoption[9].

Ritually, symbolically, and metaphorically each child, whatever their gender as a child, is made into either a man or a woman. Because of the strict division of labour among Inuit, I would agree with Briggs that there are only two genders. What remains in question is how those two genders are internally constituted and how they then relate to one another. I would propose that a quite different symbolic logic is at work that allows the analyst to account for the flexibility Briggs has seen within the Inuit data.

Not all authors agree that gender does form the essential line of symbolic differentiation among the Inuit. Fienup-Riordan (1994), on the basis of her data from the Yup’ik Eskimo of Alaska, has argued that the human/animal distinction, the maintenance of boundaries between these two through injunctions (taboos), and the ritual passage between these two realms appear as the basic symbolic modality. Fienup-Riordan’s and Saladin d’Anglure’s positions can be articulated through understanding that the production of animal products and the reproduction of human and animal life form the nexus of Inuit and Yup’ik ritual. Inuit discourse can be seen to run along two axes, male-female and animal-human at the same time (Oosten 1989).

Towards an Inuit epistemology

A different logic of the concrete needs to be articulated for Inuit gender relations. Rather than a logic of opposition and mediation, or of thesis-antithesis-synthesis (to put it in its philosophical frame), we need to see a logic of embeddedness and complementarity. What I mean by this is that in any pair of oppositions the distinction between the two terms will be maintained, but each term will always invoke the other by the fact that a piece of one term is already embedded in the other term. The classical yin-yang model of the Chinese is the most evident example of such a position (Figure 3). In this model, binary aspects of reality (male:female, animal:human) oppose and complement each other to form a unified whole (represented by the circle). Within each aspect lies a portion of its own opposite, thus overcoming the apparent mutual exclusion of the two terms. This model seems to work well for the understanding of gender among the Inuit where male and female, especially in practice, complement each other while at the same time each man and each woman contains within them an aspect of the opposite (given through their names) that becomes situationally and momentarily evident.

Figure 1

Bears within Inuit cosmological structures.

Bears within Inuit cosmological structures.

-> See the list of figures

Figure 2

Homological relations between gender and animals.

Homological relations between gender and animals.

-> See the list of figures

Figure 3

Gender relations according to the Chinese model.

Gender relations according to the Chinese model.

-> See the list of figures

In the summer of 2006, I acquired a set of pen and ink drawings by William Kilabak of Pangnirtung. The drawings represented a man beating a drum and a woman holding a qulliq over the forms of two harpoon heads (Figure 4). The lines for the harpoon heads formed the yin-yang symbol. The gender asymmetry of having the male symbol of the harpoon head on both sides of the drawing (one would expect that one side would contain something like a worked seal skin—and it would have been most interesting had Kilabak placed the man on a seal skin and the woman on a harpoon head) poses a set of analytical questions beyond the scope of this paper. I would merely suggest that it reflects a level of patriarchy in Inuit society, that has not been approached in this discussion. Even more interesting is the fact that the harpoon lines that form the yin-yang do not close into a single circle, but are separated and leave open a space between the two curves. The space between the curves suggests the possibility of a passage, a space for movement and transformation. Kilabak artistically generated the gender model I have proposed, but has modified the classical Chinese symbol in an important way, pointing towards three problems with this model: 1) it requires one to assume that ultimately everything comes together into a united whole; 2) it cannot account for homological symbol of the bear; and 3) it does not allow for movement between the categories.

The philosophical drive towards unity is probably a Western intellectual imperative, and has been brought into question not only by post-modern thought but also by Turner’s (1985) reflections on Australian Aboriginal thought. Earlier work on Inuit social organisation (Trott 1995, 2000) has suggested that the Inuit maintain a balance between what Turner has called an “incorporative and federative logic of organisation” that never allows one to collapse into the other. The “federative” links created by the history of names over an area divide the domestic groups, as both the husband and wife maintain the ties given to them by the name. At the same time, residential associations at the level of the household and the local hunting group create incorporative relations that set local task groups in competition for resources (Trott 2000). In the origin myth discussed earlier, it must be noted that two (i.e. there is no original unity) men emerge from the earth, and from their activities further distinctions are created in the universe. Further Graburn (1972: 168) has noted, “There is an inherent symmetry in carvings of most single-creature figures and there is very often bilateral symmetry (illuriik = two equals as a pair) or complementary symmetry (aippariik = two complements as a natural pair) in the multi-figure sculptures, the type of symmetry being appropriate for the creatures modelled.” Interestingly, illuriik is also the term for “cross-cousins” and “song partners,” while aippariik is frequently used for husband/wife couples. This linguistic and artistic model suggests that Inuit recognise at least two modes of relating pairs. Thus, it is not possible to reach back to a single monadic existence in Inuit thought. If one postulates that Inuit thought departs from an unmediated dualism, the question then becomes how do Inuit create relationships between such distinct terms?

In terms of gender, every male in some sense contains an aspect of the female and every female contains an aspect of the male. This is most evident in those children who have been socialised cross-gender and as adults perform the tasks assigned to both sexes. In a further dimension, Saladin d’Anglure (1997) has noted that male shamans always have at least one female helping spirit which he equates with the sex linked naming system. For them, this cross-gender linking provides the shaman with the ability to move back and forth between the various realms, male and female, living and dead. But one could equally well argue that the opposition is not so much transcended as maintained and that each person has some ability to enter into both realms (Figure 5).

Figure 4

Men and women (pen and ink drawing by William Kilabak of Pangnirtung, 2006, used by permission of the artist).

Men and women (pen and ink drawing by William Kilabak of Pangnirtung, 2006, used by permission of the artist).

-> See the list of figures

The bear is not so much a cosmological mediator (in terms of transcending and combining the land/sea dichotomy), as a cosmological operator. The bear (and the shaman) provides the logical term that makes up the linking passage or path between the two terms. The bear is not an internally unified figure, but an internally divided figure: there are two bears, male and female, and within each of them is contained the seed of its own opposite. It is precisely this property of an animal containing within it two figures that makes the bear such a powerful object of thought. This argument links to Fienup-Riordan’s (1994) analysis that Inuit symbolic thought consists of the construction and maintenance of boundaries, with the ritual opening of a passage to move outside of those boundaries. I would further argue that in Inuit terms any enclosed boundary already contains within it a piece of its own opposite—which opens the way for a passage.

Epilogue on the Nanuraq case

Let us to return to the initial ethnographic problem: what was Nanuraq’s illness? In the transcription of the seance, the shaman reviews the various offences committed by Nanuraq. Most of these have to do with eating certain foods or working materials while she was menstruating, while other offences include combing her hair shortly after the birth of a child, hiding a miscarriage, and having sexual intercourse while menstruating. In other words, nearly all of Nanuraq’s offences relate to her reproductive cycle. At the same time one must take into consideration that the seance was part of a much longer series of events that leads Oosten (1984) to suspect that Mathiassen was observing the tivajuut ceremony that takes place at the shift from summer land hunting to winter sea ice hunting. During this sequence the entire universe is recreated through confronting the productive and reproductive activities of the entire group. At this point of cosmological crisis and renewal, Nanuraq stood in real danger of transforming into a bear herself because of the offence against the slain bear and because she embodied in her reproductive capacity the transformative power of the bear. The shamanistic seance effectively resolved this problem, despite the fact that it did nothing to cure her physical ailments and indeed may never have intended to do as much.

Figure 5

An Inuit model of gender relations (The plan of the double snow house is taken from Boas [1964 (1888): 138, fig, 106]).

An Inuit model of gender relations (The plan of the double snow house is taken from Boas [1964 (1888): 138, fig, 106]).

-> See the list of figures


In this paper, I have argued that there is an homological relationship between the symbolism surrounding the polar bear and the categorisation of gender among Inuit. The homological relationship between the bear as a metaphor and gender does not exist at the level of a series of one-to-one correspondences. I cannot postulate a direct relationship between male bears and men nor one between female bears and women. Instead, bears act as a transformational operator in Inuit thought that allows movement between two bounded, but open categories of thought, categories such as gender. Unlike other models of gender which work with either a fixed binary, or, in the case of the third gender argument, a third mediating category, the Inuit model of gender operates with two clearly defined categories, which are open to each other through passages that allow individuals to move between the categories. The strict division of labour in Inuit society maintains the boundaries of the two categories, but the construction of the person among Inuit, in which gender is assigned through names (or more significantly multiple names) allows individuals to transform themselves. It is the transformational possibilities of bears and gender that create the homology between animal categories as represented by the bear and humans.

Inuit ethnography shows that there are clear boundaries between the animal and the human, the living and the dead, male and female, but also that each of these terms contains the other within it while simultaneously providing a passage between the two. It is this conceptual framework that leads to the series of cycles and transformations between each of these realms that are acted out in human and animal bodies and through rituals.