The Anthropology of Northwest Coast Oral Traditions Bibliographic Essay, Brian Thom, PhD Student, McGill University, Department of Anthropology, March 2000

Introduction: Discourse & Ethno-literary Categories

The verbal, textual, performed and artistic discourses of the people that we study as anthropologists working in Native North America are the central, if not the defining elements of how we come to understand and describe 'other' cultural systems. These discourses are the means by which the people we work with communicate the ideas and meanings to us and to each other. To gain a subtle understanding of these discourses, it is clearly not sufficient to make a cursory glance at them, and try to assign some particular and wholly encompassing historical or cultural significance. Discourses must be engaged and contextualized. They must be viewed and re-viewed, spoken and rearticulated. From our position in these discourses, and our continuing effort to reposition and recontexualize ourselves within them, the anthropology of Native North America has built an academic tradition. We have written, presented, testified, exhibited and sometimes even acted out our ever-changing understandings of the discourses of the communities in which we work. The message we have tried to express has changed since Boas collected his first native text , but these objects of our study, Native discourses of myth, legend, folklore, history, life history, (auto-) biography, have remained a core focus for the production of meaning and understanding within our own discipline.

In this paper, I will present a critical history of anthropological writing about native Northwest Coast⁽¹⁾ oral traditions. The Northwest Coast is a highly interesting and useful place to discuss how anthropology has viewed oral traditions in part because so much influential work has been done following the vast text collections of Boas and his collaborators. In the generations following Boas' work, a increasing attention influences of the theoretical interests of British social anthropology brought a new attention to the social function of oral traditions (in the realms of ritual and property) and in describing how oral traditions formed structured accounts of local cosmologies. With the influence of Lévi- Strauss, attention turned from function to meaning of oral traditions, in explicating how symbols operated within traditions such as mythology. Recently, the Northwest Coast has also been the centre of much discussion and debate surrounding the ethnography of speaking, which builds largely on the work of Melville Jacobs and Dell Hymes, both of whom studied Northwest Coast oral narratives.⁽²⁾ In this literature, attention changes again from structure to agency, placing an emphasis on how narrators use particular forms of speaking to embed meanings (both personal and cultural) in the active process of telling of oral traditions. Thus, stories come to have ' social lives' which through different tellings, can come to have a wide array of powerful meanings.

In critically reviewing the anthropological literature on Northwest Coast oral tradition, I must outline my own theoretical and methodological perspectives. Like many Americanist anthropologists (as reviewed by Valentine & Darnell 1999:6), I subscribe to the theoretical position that culture is a system of symbols and there is an inseparability of language, thought and reality. By focussing on one kind of native discourse - oral traditions - a rich set of texts is generated which can be studied to learn something of that culture's communicative systems and norms, and the way these norms are put into action through the practice of telling the stories. It is this interplay between structured cultural/linguistic practice and the agency of individual story tellers is where culturally-situated meanings lie. To be able to navigate this conceptual territory between structure of oral traditions and the agency of narrating them, we need to explore the connections between form and meaning. In such an examination, we would grapple with the interplay of literature, linguistics and ethnography in the realms of cosmology, marriage, descent, succession, political inequality, and the nature of human relations to non-humans. By laying out these communicative practices, we can begin to see how they are used by individuals to make meaning in their own social context.

It is clear, however, that this academic discourse is not centred on the object of native discourse (Hill 1999:181). Native oral traditions on the Northwest Coast is, as we shall see, often concerned with negotiating power relations between individuals in that society. In being told and retold, they are often concerned with very particular and highly local social processes and situations. Each narrative in Native discourse is "connected to every other [narrative] and to a highly contextualized discourse that assumes familiarity with biography and shared experience"(Ridington 1999:22). Thus, native discourse is "in conversation and dialogue" (Ridington 1999:19; see also Sarris 1993:5), and it is a dialogue in which anthropologists, by and large, are not engaged. They become particularly removed from this dialogue when stories are abstracted from their social setting and presented as 'the mythology of such-and-such a tribe'. So in this review, we must keep in mind that the anthropological discourse is very seldom about the real object of the native discourse on this fine level. It tends to operate at a level of abstraction, trying to narrow the gap in meaning between the culture of the discourse makers and that of their non-local (often unintended) audience.

Before narrowing down to a discussion of the particular analysis of Northwest Coast oral traditions, some attention must be paid to the definitions. Terms like 'oral traditions', 'mythology', or 'legend' often be used loosely, laying groundwork for potential misunderstandings. This problem is even more acute when considering native ethno-literary categories. These categories are often 'understood' by a native audience who know conventions of narrative form and content, but not made explicit in the telling of the story to outsiders such as anthropologists.

Oral tradition refers to the cannon of narratives (which include myths, histories, folklore, proverbs, riddles, some songs, deeds/charters) that are more widely known. It would be the canonical nature of their content and form which make them oral 'traditions'.

In their review of native discourse for the Handbook of North American Indians volume on Languages, Kinkade & Mattina suggest that there are eight main kinds of Native American discourses: myths and tales, sacred texts, historical narratives, speeches, poetry and songs, life histories (Kinkade & Mattina 1996:244). They also observe that the Northwest Coast may be unique in Native North America where "special tales are told of the origin of a specific family; such tales were used to validate the status of the family within the community and in relation to other families"(ibid. 270).

Much of the anthropological discourse about Northwest Coast oral traditions has been framed in terms of the analysis of the first two of these categories: 'myths' and 'tales' or 'legends'. Myth is a slippery word in that it has been used by people studying native oral traditions to describe a wide array of narrative forms with little precision in its use. In general, oral traditions which are of 'mythical' form can be taken to have a number of key elements: they are accepted as 'fact' on faith, often being sacred narratives; they take place in a remote time and in a 'different' world; they have 'non-humans' as main characters. Myths often explore themes of 'creation' of both people and nature. This definition is often put into contrast with 'legends' which are also taken as factual and sometimes sacred, but differ in myth in that they take place more recently in a world like that of today, and largely deal with the affairs of human characters (Finnegan 1992:147).

Starting with Boas (cf. Jacobs 1959a:132), anthropologists writing about Northwest Coast oral traditions have generally held to this distinction between 'myth' and 'legend'. Authors usually support making this distinction by referencing words elicited from native languages which share the semantic content of myth and legend in English (see Table 1).

[unfortunately, the native words in this table do not display properly without the Halqemeylem Font installed]

However, as Jacobs (ibid.) and more recently Kinkade and Mattina (1996:270) have pointed out, there are important categories of oral tradition which must be discerned. Some oral traditions have mythical elements, yet are more than (or other-than) 'myth'. Gitksan⁽³⁾ adaawk (as discussed by Sterritt et al 1998), have clear mythical elements, with people marrying animals, voyages to sky-world, and the transformations of human ancestors into stone. However, they might better be thought of as a kind of oral history, with temporally-stable content about genealogies and ownership of land and access to resources. Linguist John Enrico supports such a tri-partite division from Skidegate Haida, which divides stories into q'aygaang ('myth'), q'ayaagaang ('lineage history'), and gi7ahlralaang ('real history, news') (Enrico 1995:4). In Haida:

the boundary between 'myth' and 'lineage history' is blurred somewhat by the incorporation of some episodes taken from myths into lineage histories, and by the semi-mythical nature of early events and individuals they describe. For the most part, however, lineage histories can be said to begin where the major creation myths leave off (ibid.)

Cove (1987) suggests a different tripartite division of Northwest Coast oral traditions into 'myth', 'legend' and 'folktale' with 'myth' further dividing into 'house narrative', 'shamanic narrative', and 'secret society narrative'. The sub-divisions of the myth category are all privately held knowledge that are 'owned' by individuals or local kin groups. Though no native terms are given for folktales, he distinguishes these from legends and myths in that they are clearly thought of as fiction, often with a moral purpose.

Berman has taken an linguistic approach to the analysis of ethno-literary categories in her presentation of a "partial taxonomy of Kwagul ethno-literary categories"(Berman 1991:118, see also Berman 1992:148). In a branching hierarchical diagram (which I have represented as boxes in Table 2) Berman takes the native categorical distinction between myth and tale further by suggesting Kwagul people make further distinctions about the kinds of myth. Berman notes that Boas himself did not recognize this branching hierarchical structure, and imposed his own translations when glossing native ethno-literary terms: myth, story, legend or tradition for nuyºm, where Hunt gave the gloss 'historie' (Berman 1992:127).

[unfortunately, this table does not display the Kwakwala words correctly without the Halqemeylem Font installed]

Berman's branching model is quite interesting as it reflects quite closely how Kwakw'ala speakers themselves envision the differences between the kinds of stories they tell. Berman recognizes the classificatory scheme as partial in that it does not include a break-down of the different kinds of 'tales', and that it fails to explore the nuances of 'animal stories' which can seem like 'tales' when involving recounts of interactions with guardian spirits, and at other times are more clearly mythological. Pre-ordered knowledge about the kinds of plots, episodes and characters in the stories are clearly needed for the listener to be able to sort out what kind of story it is they are being told.

Skilled story tellers can play at the margins of these 'myths' and 'tales', drawing on characters and plots from one category and placing them in the theme and incidents of another. In a story told in Kwakw'ala to Boas by Qumgilºs, the usually serious mythical Transformer figure meets the rancorous 'Oolachan-Woman' from the Animal stories, with comical consequences (Berman 1992:149). Berman points out that Boas was not aware of the 'play' on ethno-literary categories that was being made in the telling of this story, and completely missed the (obscenely) humorous point of the story. 'Missing-the-point' of metaphors made in the contexts of inverted narrative forms is tricky business, and not all stories will challenge their anthropological audience in this way.

Such a misinterpretation does provide an excellent cautionary tale, which leads us back to a point made earlier. The objects of anthropological study - our implicit or explicit theoretical goals - are seldom those of the narrators of the oral traditions from which the anthropological goals are met. The gap between the meanings of oral traditions told by native narrators and those meanings ascribed by anthropologists becomes greater as the goals of anthropological theory move further away from understanding highly local, culturally and historically situated contexts. Of course, particular studies of the local (which Boas, as we shall see excelled at), are not the only kinds of questions which can be asked of oral traditions. But when these other questions arise, driven by our theoretical interests in, we must recognize their position (just like the telling of oral traditions) as another kind of social act. And all social acts, as we know, are imbued with relations of power and the accompanying potential for dominance, hegemony, resistance. Thus, even the seemingly innocuous categorization of native oral traditions as 'myth' or 'legend', 'house story' or 'tale' can take on a highly potent social life in arenas where myth is transformed into 'common law' and 'house stories' become legal codes.

Franz Boas' Native Texts - Oral Traditions as History and Culture

Any analysis of the anthropological discourse on Northwest Coast oral traditions must consider the foundational work of Franz Boas who personally did more towards the collection and publication of native texts than any other scholar (Jacobs 1959a). For Boas, his colleagues and students, 'native texts' (as they called them) were first seen as important kind of data for obtaining information about the history and distribution of culture. In his early work for the British Association for the Advancement of Science and the Jesup North Pacific Expedition, Boas studied the distribution of elements in oral traditions in order to show relative historical movements of cultural groups - resolving historical questions about the prehistory of native America, and the contacts between the Northwest Coast of North America and the northeastern regions of Asia. In his later work at Columbia University, Boas struggled with the problem of how cultural outsiders - with all their biases and preconceptions - could gain insight into how native people understood their world and how they ascribe meaning to the stories. Documenting oral traditions provided a means to discover what was important to native people, in their own words, untroubled by the pre-selected responses of the question-answer methodology of participant observation. The legacy of this work holds major importance for the development of anthropology, in general, and to the current shape of anthropological studies on the Northwest Coast.

During his career, Boas made collections of texts in a great many of the languages spoken on the Northwest Coast.⁽⁴⁾ Boas had an acute ear for phonetic dictation and was able to record (and publish) a great deal of material in their original language. For some of the languages he was completely unfamiliar with, or which he did not have much time to work on taking dictations, Boas was forced to work in Chinook Jargon and English. In these cases, the published versions are almost always given in English only.⁽⁵⁾ During his tenure as coordinator of the Jesup North Pacific Expedition, Boas directed his colleagues to also record texts - resulting in important volumes such as Swanton's Haida Texts (Swanton 1905, 1908) and the work of James Teit with the Nlakapamux (Thompson) (Teit 1898, 1912, Boas (ed.) 1917) and less voluminous works such as Farrand's Quinault and Quileute texts (Farrand 1902; Farrand & Mayer 1919 respectively). While working at Columbia University, Boas inspired a number of students to collect and publish myths and oral traditions in areas he had not touched himself, including the important work of Edward Sapir with the Nuu-chal-nulth (Sapir 1924; Sapir & Swadish 1939), and some minor works in Coast Salish communities which he had not worked in himself (ie: Haeberlin 1924, Gunther 1925; Andrade 1931).

However, of greater importance in terms of sheer volume are the two major collections of native language texts which came out of Boas' first-hand work with collaborators who lived in the communities they were documenting. The most well known are the Kwakiutl texts, which George Hunt produced for Boas (Boas 1910; 1935; Boas & Hunt 1905; 1906). Boas also had a long, active correspondence with Henry Tate, which ultimately produced a series of Tsimshian texts (Boas 1912; 1916). Given that these are among Boas' most thorough and well-known works, I have picked these to focus on from his massive corpus of work on Northwest Coast oral tradition.

Most of the Kwakiutl texts were taken down phonetically by Hunt in the Kwakw'ala, a translation provided, and then sent in notebooks and letters back to Boas in New York (though a few were collected by Boas himself (ie: Boas 1910:1-243)). Often accompanying these texts were objects collected or purchased for the American Museum of Natural History, and ethnographic notes that were collected in response to Boas' questioning in previous letters. While the objects were put on display or in storage and ethnographic notes filed and organized for future publication, the transcriptions and translations of Hunt were reviewed and corrected by Boas from his office. These were subsequently published, normally with the native language version and a morpheme-by-morpheme translation and a summary gloss in English provided on the facing page, at the end or sometimes even in an accompanying volume. These texts were presented as an object in and of themselves, removed from their social context (as recorded in Hunt's ethnographic notes) and their material context as the objects which originally accompanied them were now on display or in storage at American Museum of Natural History. This separation of the texts from their context make them a daunting 'object' to approach. The highly technical phonetic transcription and often awkward, literal translation make the texts seem like oddities to the student with little prior knowledge in Kwakiutl culture. Boas was criticized for this at the time by Sapir who argued that he had not provided adequate annotations of these texts which could help "the student of Kwakiutl mythology and culture towards the understanding of the tales" (Sapir 1912:197).

For Tate's Tsimshian texts, the route to publication took a more complicated route. Tate took down versions of the stories he knew or heard in English, and then proceeded to translate them into Tsimshian, which he sent off to Boas (Barbeau 1917:562; Maud 1989:158). Boas was never satisfied with Tate's transcriptions of Tsimshian, so he had Archie Dundas, a Tsimshian-speaker who was studying on the East coast, to read Tate's Tsimshian out loud, which Boas then re-transcribed. Not wanting to be biased by Tate's English, Boas drew on his own limited knowledge of Tsimshian to re-translate the Dundas transcriptions in the same fashion he had done with the Kwakiutl material. The final results were published in the style of Hunt's Kwakiutl work. The Tsimshian texts are subject to the same critique as the Kwakiutl material, though the long path of translation makes it more difficult to be satisfied with this material as representing the narratives of a living community. They represent more the work of a unique anthropological collaboration. Boas' method of publishing the Kwakiutl and Tsimshian texts is important to review because they form the basis of several subsequent generations of anthropological study of these communities, who frequently took the texts as authoritative, original sources which stand for 19^th century Northwest Coast cultures.

Apart from presenting the texts themselves, Boas' analyses are limited. Boas was loath to see these oral traditions as forming a cultural 'system', unlike his colleagues overseas where were influenced by Durkheim and Malinowski, or the older Americanist school of Holmes and Dorsey from the Bureau of American Ethnology who saw oral traditions as an indicator of general processes of unilinear cultural evolution. For Boas, oral traditions reflected particular historical processes of each 'tribe' from which they were told. Thus, by collecting a wide range of texts from many Northwest Coast communities, Boas attempted to address questions concerning the history of the distribution of Northwest Coast cultural groups. Some of his investigations along these lines have become useful contributions to an overall historical picture, while others have not so well stood the tests of time. Two examples of his numerous observations along these lines will illustrate the point.

Boas proposed a late migration of the Tsimshian into the area they now live on the basis of the similarity of their myths to interior and southern North American native peoples (Boas 1916:872). Though this movement occurs at a much longer time-scale than Boas had projected, his hypothesis is further substantiated today with linguistic and archaeological evidence (Sterritt et al. 1998:15-24). Boas similarly proposed that the Coast Salish were recent migrants to the Coast, having come originally from the interior and then having been influenced by their more complex neighbours to the Kwakiutl. Boas' argument rested on the idea that the Kwakiutl were necessarily there first, because of their more complex social organizations and myths:

All the privileges of the [Kwakiutl] clans are explained by the clan traditions, which, for this reason, are considered a most valuable property... When the Salish tribes began to be thrown into contact with the Coast tribes, the lack of family traditions must have been felt as a great disadvantage ... [b]ut their tribe was organized on a different basis from that of the Coast people. While the [Coast tribes] were divided into clans, the idea that was present to the minds of the Salish people was that of village community; and it is clear therefore, that the traditions which developed would be of such character that each village would have one mythical ancestor. (Boas 1898c:17-18)

The hypothesis of a late Coast Salish migration to the coast has been rejected by archaeological and linguistic evidence (Matson & Coupland 1995:242-6; Ames & Maschner 1999:86; Thompson 1979). This puts into doubt Boas' methodology of using comparative analysis of oral tradition to understand these historical processes. Snyder has commented on how difficult it is to know where these stories really originally come from, as so many narrative elements are widely shared throughout western North America (Snyder 1964:56-7), and thus the logical fallacy in using myth as a marker of historical distributions of cultural groups.

Boas' second theoretical position for understanding oral traditions drew on his position as a cultural relativist. By the time the Jesup Expedition had run its course, Boas had become more interested in trying to understand native culture on its own terms. For Boas, an observer from outside the culture being studied would not necessarily know the significance of what had been said and done, or even what was significant and what was not. To counter this problem, Boas worked with native informants, who collected material in the language spoken by the community they were studying. He felt that these texts:

...probably contain all that is interesting to the narrators and that in this way a picture of their way of thinking and feeling will appear that renders their ideas as free from the bias of the European observer as possible. (Boas 1935:v)

This would be presented in its entirety so that all significant material was present, whether the researcher recognized it as so or not. Making 'complete' collections of native oral traditions from a particular community provided a methodology to achieve fairly rich cultural descriptions. This perspective was developed in explicit opposition to the evolutionary ideology of his American counterparts, which viewed mythology as being 'primitive rationalization' which survived as irrational custom. For Boas, mythology had unconscious origins and was "central to the maintenance of society through its rationalization of traditional forms of behavior" (Stocking 1968:225). Given that the content of the texts is unconscious in origin, Boas set out to use these texts as a basis for ethnographic observations on the culture of the peoples from who the stories were taken.

In his work with Hunt and Tate, Boas felt he had a sufficient body of material to make authoritative statements on the cultures as wholes. Boas' two major analyses which explored this idea, Tsimshian Mythology (1916) and Kwakiutl Culture as Reflected in Mythology (1935), were lengthy projects which drew solely on mythology to state 'ethnographic facts' about the culture of the societies from which the myths were recorded. These works read like long, detailed lists of cultural elements. They provide an annotated index to the myths on topics of: material culture, personal and family life, tribal organization, emotional life and ethics, ceremonial objects and procedures, supernatural power and objects, numbers, the world, supernatural beings, animals and plants, and the origin of local geographic features. This range of topics covered was constructed inductively from the content of the myths themselves, and thus for Boas were 'the things important to the Indian himself', unbiased by the interests of the observer. The conclusions of these works return to his previous interests in the connections and differences between the groups on the Northwest Coast which could be seen from patterns of similarities and differences in the content of the myths.

Boas' peers critiqued him for not providing adequate context to interpret if the myth contains 'borrowed' events or themes and thus non-local cultural practices (Barbeau 1917:551). This critique did not pose a problem for Boas, who saw each culture as being constituted by their unique historical development in interaction with neighbouring groups. Thus, a 'borrowed' element was just as important to the native view as a local one, in a culture which at any given moment is the product of a dynamic history.

Boas' also ran into trouble in his assertion that the oral traditions told or collected by his individual informants represented notions held by the culture as a whole. His final cross-cultural comparisons illustrate the difficulty with this view. Boas concluded that there is a direct correlation between the number of occurrences of particular kinds of historical events mentioned in the myths, like starvation and natural disasters, and the actual relative frequency of those events in the history of the societies which the myths are taken to describe (Boas 1935:173). In making further distinctions between the 'interests' of the cultures being described, Boas distinguishes "the selection of preponderant themes, ... the style of plots, and ... their literary development" (Boas 1916:878). The Kwakiutl, Boas lamented, are interested mainly in rank and privileges and are otherwise "lacking in variety of subject matter and in skill in composition" (Boas 1935:190). The Tsimshian, on the other hand, were said by Boas to be mainly interested in Raven's voraciousness, Mink' amorousness, and the marriage between humans and animals (Boas 1916:877). Since Tate did not provide stories of ownership, Boas concluded that they were missing from Tsimshian mythology (Boas 1935:176). To substantiate his extrapolation, Boas claimed that he had published all there was to collect of Kwakiutl and Tsimshian mythology (cf. Barbeau 1917:552).

All of these conclusions are based on the central notion that the oral traditions recalled or recorded by Hunt and Tate represent a 'complete' body of narratives from their respective cultures. This is clearly a suspect position to take if one considers the possibility of personal agency. However, Boas did not consider the relationship between individual style and variation in narrative form. Nor did he see oral traditions as being part of an active social context. For Boas, the content of the stories stood as the substance of what there was to be analysed, regardless of individual style or 'use' of story as a social tool. Thus, as ethnographies of living communities, these works are disappointing.

If we reject Boas' practice of having the texts narrated almost exclusively by a single collaborator stand for the society as a whole, we can make sense of the differences in the interests of the Kwakiutl and Tsimshian. Berman points out that in Boas' correspondence, he certainly asks for a range of texts, but rather, that his requests were met with silence from the collaborators who were either unwilling (Hunt) or unable (Tate) to produce texts from outside their relative social sphere (Berman 1991:45; Maud 1989:161). It seems clear that these emphases in myth are a reflection of the individual teller's interests and abilities (those of Hunt and Tate), and cannot necessarily be extrapolated to the whole societies from where they come. George Hunt, for instance, was a potlatching community member, very much worried abut the same issues he recorded, and thus was apt to pay special attention to recording, for Boas, stories which were concerned with property (Berman 1991; Suttles 1991).

Tate's own narratives were guided firstly by his status in the community. Tate was not a high-standing community member like Hunt (Barbeau 1917:553), and likely did not have access to the range of stories recorded by Hunt. This is confirmed by Voila Garfield who worked with the Tsimshian in the 30-40s. She noted that Boas had presented the mythology "as if it were general community property. Only a few stories in these [Boas'] published collections were identified as property or as explanations of historical details of lineages and clans"(Garfield 1966:52). Garfield saw that not a single Raven story told to her by Gitksan people bore any similarity to those collected by Tate from the Tsimshian (ibid. 53). This demonstrates the highly limited range of stories Tate had access to because of the proprietary nature of many Northwest Coast myths. Barbeau observed further that Tate had often simply restated Nisga'a stories that Boas had previously collected and published. Tate's restating the Nisga'a stories arose in part from a misunderstanding of Boas' instructions (Maud 1989:16).

More careful attention to the range of narratives from any particular ethno-literary category might have drew Boas' attention to these problems. Seguin notes that Tate's texts "include few of the adaox [adaawk] and more of what might be called 'stories'... They Beynon texts [which were collected in collaboration with Boas' critic Barbeau] are much richer in the adaox but have figured little in the published literature." (Seguin 1985:24). Hunt, on the other hand "consulted a wide range of Kwagul in producing these texts and the variability the texts display is probably a more accurate picture of the state of cosmological opinion in many societies than what anthropologists show us" (Berman 1991:132). Boas, however "does not label or categorize the narratives except by provenience or by subject matter" (ibid. 117), thus failing to see general patterns across Northwest Coast oral traditions, and assuming incorrectly that his collections represented perspectives on a cultures entire oral tradition.

Hymes has challenged the long-standing idea in the study of native discourses that the oral traditions told by individuals can stand as authoritative representatives of the traditions of the entire culture (Hymes 1996). Hymes has looked at different tellings of 'Coyote' stories and has shown that very different meanings emerge from the various tellings of the same stories. This points to the real short-comings of the analyses of Boas who was so interested in objective content of the stories that they virtually ignored the structure of particular tellings and the meanings that individual tellers may have intended. For Hymes, Boas' legacy of presenting oral traditions disconnected from the social process of telling them has had an impact on more mainstream (non-academic) publications of 'Native Mythology'. Hymes suggests that we would best honour the thought of native storytellers "not by extravagant generalization, but by close attention to the detail of their narrative skill"(ibid. 131).

Though these are serious problems for Boas' attempt to understand culture from the native point of view, there are a number of important points that Boas makes which are useful today. The methodological implications of Boas' program of collecting texts in native languages in order to say something emic about native culture has become a very important idea in Americanist anthropology. Edward Sapir noted that these seemingly massive collections of texts are the barest minimum necessary for making a proper study of the language and culture of Native groups (Sapir 1912:194). The theoretical implications of Boas' methodology was the starting place for Sapir's vision of the relationship between language and culture, which in turn has had a impact on the development of the study of discourse in structural linguistics (Sherzer 1987), and anthropologists who study discursive practice as social action (Valentine & Darnell 1999:6).

Boas' work has laid the foundation for much future research on the Northwest Coast. No future study has shared his ambition to record and publish so many oral traditions in their native language, though many research projects have used his material (largely the English translations) to pursue new theoretical goals. Before examining the work which followed Boas, I would first like to note the unusual position Boas has in contemporary native communities today. Gloria Cranmer-Webster, great grand-daughter of George Hunt and curator of the U'mista Cultural Centre has written about how important Boas' work is now in her community. In a paper which details many of the contemporary significance of the Boas material to her community, she mentions the importance of the Kwakw'ala texts:

We are re-claiming what was lost during the 'dark years', as our old people say... The missionaries almost succeeded in their efforts to wipe out our language, so that we utilize the Boas/Hunt texts in developing language materials to teach Kwakw'ala to both adults and children. (Cranmer-Webster 1997:6).

Though the process is very different than in Boas' time, his contributions in recording oral traditions continue to be felt by the decedents of the native peoples he worked with.

The Social Function of Northwest Coast Oral Traditions

British Functionalism after Boas

In the decades after Boas and his colleagues had done their work, new questions arose concerning Northwest Coast oral traditions. In this new generation of scholars, oral traditions were not so interesting for what they could tell about the large-scale questions about the ancient movements and contacts between populations on the Northwest Coast. Nor were they considered particularly interesting for what they could reveal about the 'mentality' of the cultures from which they emerged. This generation of anthropologists were more keenly interested in looking at how oral traditions functioned in society. What were their role as law, ritual guidance, cosmological explanation? Oral traditions, like other institutions such as the potlatch, filled a social role which could be described in relation to the functioning of traditional Northwest Coast societies. Many of these emerged under the influence of Malinowski, who saw 'myth' as a social charter which validated the cultural institutions and customs of a culture, providing a common bond of social solidarity (Malinowski 1948 [1926]). Finnegan (199:34) reviews this literature, noting that emphasis was placed on the stability and homogeneity of the systems of oral traditions, and little emphasis was placed on local meanings or individual creativity. All oral traditions with mythical qualities tended to be viewed as 'myth', and most other forms more-or-less ignored, as they had little relevance for gaining an understanding of social structure.

This perspective was largely dismissed by Boas, who repeatedly argued that there was too much local variability to see oral traditions as forming a social system to explain cosmology. This view comes out forcefully in his condemning review of Locher's The Serpent in Kwakiutl Religion which badly mis-read Boas' own data on Kwakiutl mythology and religion (Boas 1940 [1933]) by suggesting that Kwakiutl myths form a systematic, integrated social system. Given Boas' commanding presence in American Anthropology in general and on the Northwest Coast in particular, very few people attempted to take this perspective while Boas was alive.

One exception to this was Boas' own reading of Bella Coola oral traditions, where Boas did posit a connection between the particular social organization of an individual society and the content of their myths (Boas 1935:173). This connection was evident for Boas in the seemingly conflicting traditions of the Bella Coola, where each village has their own tradition of the origin of the world. Boas' explanation for the great variation in narratives is that it reflects how Bella Coola communities conceive of property; each community owns a distinct myth of origin and accompanying it are unique rights and privileges over resource areas (Boas 1898a:48). Thus what seems on the surface as conflicting cosmologies are in fact an integrated set of traditions around property, each held privately by the property-owning social unit.

The other exception to this comes from the work of T.F. McIlwraith, who also worked with the Bella Colla. McIlwraith, ironically, corrected Boas' above observation, noting that the 'systematic' account was only shared by two families (McIlwraith 1948:25, 294). In fact, Boas had only worked with a single man for ten days (Maud 1982:88-9), reminding us again of the problems of making generalizations about whole cultures from single individuals. Nonetheless, McIlwraith (who was trained in Cambridge) saw the 'mythology' of the Bella Colla as having a social function in integrating attitudes towards animals and supernatural beings (McIlwraith 1948:30), and as an integrating force in their social organization, with "myths now accepted as facts, [and] hav[ing] had a great influence on the lives of the people. In fact the social structure of the tribe has tended to conform to the myths" (ibid. 118). McIlwraith also viewed Bella Coola origin myths giving authority to the rituals performed (ibid. 292) McIlwraith, as Maud (1982:139) points out, sees oral tradition from the stance of its actual use in society, presenting the material firmly from the stance of having 'been there' and heard oral traditions in use. While it is difficult to argue with McIlwraith's analysis on one level, his emphasis on cultural homogeneity still alludes a sense of oral traditions being used in dynamic ways for a multiplicity of social purposes by different people at different times. For the time, however, McIlwraith's two volume ethnography of the Bella Coola stands in stark contract from Boas' work.

The Function of 'Myth' as a Charter for Property Rights

The students who were trained by the students of Boas in American universities set about a project of publishing a fairly standard range of descriptive ethnographies, focusing on the 'memory culture' of the older generation of native people living in Northwest Coast communities to say something about 'traditional culture'. These researchers generally operated only in English and were unconcerned with collecting native texts. They brought with them some of the critiques of the Boas school from British structural-functionalism, and set about to describe the operation of various social systems like the potlatch, winter ceremonials, and cosmology. The ethnographies were written in a farily standard formula, and did not tend to include a section on 'mythology' or 'oral traditions'. However, their native informants continued to bring up their stories during their interviews with ethnographers. These oral traditions were integrated into the 'salvage' ethnographies by being summarized for the purpose of a description of their roll in native explanations for social institutions or beliefs. Nowhere was this more common than in descriptions of the social institutions of property and land tenure on the northern Northwest Coast and in an exploration of Coast Salish cosmologies on the Central Northwest Coast. These issues have continued to receive critical attention since the presentation of oral histories as evidence for land ownership and native jurisdiction over the territory in the Delgamuukw case (Thom 1998a).

Myths and Property - Tsimshian, Gitksan and Nisga'a adaawk

Oral traditions for the Tsimshian, Gitksan and Nisga'a have frequently been presented as being important in establishing hereditary rights and privileges. Though these oral traditions are almost always referred to as 'myths', these stories are more clearly from that intermediate class of myths and legends which is concerned with property rights, which may be unique to the Northwest Coast (as reviewed above). Three sources are useful to review in this regard, Viola Garfield's short but well regarded ethnography of the Tsimshian (1966), Wilson Duff's publication of the oral traditions associated with the Tsimshian poles he collected for the British Columbia Provincial Museum (1959) and the recent presentation of Gitksan and Nisga'a oral traditions in support of a Gitksan land claim (Sterritt et al 1998).

For Tsimshian 'myths', Garfield noted that the greatest number of texts had animals marrying humans to punish or instruct, giving power, crests and gifts (Garfield 1966:49). She concluded that the motivation behind many of these myths is the deep interest of the Tsimshian in lineages, territories and possessions (ibid. 52). This plays out in knowledge of myths, where not everybody knows all the stories to the same degree, nor do they have rights to tell those they may know. Raven cycle myths, for example, everybody knows. However some particular episodes in the Raven cycle are family property, which is publicly recognized on totem poles in front of the house of the owners.

Wilson Duff was also involved expressing the relationship between pole owners, territories and myths in his publication of Histories, Territories and Laws of the Kitwancool (1959). The Kitwancool, now called the Gitanyow, and who are a politically independent subgroup of the Gitksan, agreed with the British Columbia Provincial Museum to exchange their decaying totem poles for having their territories and laws recorded and published. Duff had a non-Native who had long taught in the community and who spoke the language of the Kitwancool record the stories verbatim. He accompanied the stories with a short ethnography which described the ownership of the stories and poles by each house, each with a chief who is ranked according to the respect and authority he commands from others:

When a pole is erected or changed, it is erected at the same place. A feast is always given and the territories are discussed. They tell the people the size of their village, the mountains they own, their hunting and fishing grounds... they tell each clan which mountains they can have and what areas the can hunt and fish in (Duff 1959:18).

Personal names and place names explicitly mentioned in the stories are the link of ownership between people and places. Thus, when chiefs tell myths, or rather have an expert 'speaker' tell them at a potlatch feast, story telling becomes an explicitly political act, authenticating claims to land and opening up to challenges from people with other stories.

The Gitanyow (who were formerly called the Kitwancool) have made this argument again, using oral traditions, in a recent publication countering the overlapping land claim of their Nisga'a neighbours. Sterritt et al (1998) have argued that the Nisga'a have made a major breach of indigenous law in making such a large claim over territory that they should know well is not theirs. This indigenous law arises from recalling the oral traditions that document land ownership (adaawk), marking household ownership of land on totem poles, and validating the land tenure at potlatch feasts (yukw). This is the same 'law' that was recognized by the Supreme Court in Delgamuukw as being valid and absorbed into Canadian common law. The book is an upfront challenge to the Nisga'a claim, with comprehensive evidence from oral traditions about land ownership (adaawk) forming the core of the argument. The adaawk are backed up by a detailed review of the written and cartographic record for land ownership and tribal boundaries on the Nass River. Thus, this becomes an extremely important text in suggesting precisely how oral traditions might be presented to say something concrete about aboriginal title and rights in the post-Delgamuukw era in Canada.

Sterritt et al (1998) claim that in Gitksan/Nisga'a law, the existence of an adaawk proves ownership to the specific lands named in the story. Toponomy is critical in telling adaawk to connect mythological elements of stories to specific, owned areas on the ground which can be recognized today. Potlatch feasts (yukw) are held to formalize the recognition of these ownership rights and to allow them to be contested. In the book, the authors present the adaawk as forming unambiguous statements of ownership of land. Legendary episodes in the stories are correlated with geological history to suggest a relative chronology for the time when the events were supposed to have occurred. From this they build a picture of the ancestors of the present-day clans, households and villages having settled the unoccupied territories in post-glacial times. The initial settlement was followed by three periods of movements and wars which resulted in the present-day structure of land ownership. New territories could be claimed in one of two ways. First, lands that were seen as abandoned during the migration were thrown open for new ownership. Otherwise, lands could be ceded to another tribe as retribution for lost wars or as compensation for services provided within a clan. Such a cession of land must be formally recognized in a special land-cession ceremony (xsiisxw) given at a potlatch. The transfer of land could not occur without a widely recognized xsiisxw. Lands that were owned by a house required permission to use (in principle), and were actively defended from trespass from both unwelcome neighbours and unwelcome Europeans. Marriage alliances brought access to land, but not title.

In assessing their adaawk, the authors have done a remarkable job of working through a very large, diverse body of these stories, from old anthropological texts to court transcripts to researchers field-notes, pulling out concise statements about land ownership and boundaries. Where there are seeming inconsistencies, the authors provide ample historical and ethnographic context to understand their place within the indigenous system of land tenure. The authors further support the position of the Gitksan-Gitanyow by presenting records of historic statements about their respective boundaries that had been made by Gitanyow, Gitksan and Nisga'a people to European and Canadian colonial authorities in their long-standing efforts to maintain ownership and control of their lands. In a series of remarkable early 20th century maps which were produced by Gitksan people, the boundaries claimed in the adaawk are made concrete on the terms of the colonizers, outside the context of the potlatch feast. Other important records include some of the testimony presented by Gitksan Elders at the Delgamuukw trial. For this contemporary evidence, genealogical links to the people who taught them the histories is carefully noted in a kind of cultural footnoting which establishes their authority.

The authors' careful consideration of this wide range of evidence shows that early statements made by Nisga'a people do not disagree with those made by historic and contemporary Gitanyow and Gitksan, including the evidence presented by the Nisga'a in the important Calder land claims case of the early 1970s. However, the recent claims made by the Nisga'a Tribal Council have increased twice since 1979 to their current overlapping status of the entire Nass watershed. The authors then suggest that the conundrum of the overlapping claim may have arisen because the Nisga'a (1) based their current treaty claim on the ambiguous wording of a 1913 Nisga'a petition; (2) mistakenly argued that a few Gitanyow family heads brought their title to the land with them when they settled in Nisga'a villages in the early 20th century; and (3) failed to recognize the land given to the Gitanyow in a 1861 xsiisxw (war reparation) by their mutual neighbours the Tsetaut.

In making these arguments outside the context of a potlatch the Gitksan scholars have given serious consideration to the foundations and content of aboriginal common law. The publishing of the adaawk in a peer-reviewed academic press is in keeping with potlatch tradition which requires witnesses (readers) to claims being made and authorities (reviewers) to validate or challenge the claims. In reality, however, an academic publication is no potlatch. Unlike an oral recollection of an adaawk at a potlatch feast, no one is able to stand up in the text and contradict what is being written. Also not considered are the implications of codifying aboriginal traditions as 'law'. For instance, it was strongly stated that land ownership could not be passed on at marriage, only rights to use land. This was stated as fact with little further support or testimony to support it. However, in seeming contradiction to this, the authors ceded to the Nisga'a a small, plausible area of overlap (a berry-ground called wilbaxt'aahlgibuu) which on the basis of intermarriage is not clearly the property of one group or the other (ibid. 168). This leads me to wonder further about how land tenure might be gendered and how this has been silenced in this text (and largely overlooked in Northwest Coast scholarship).

Despite Boas' clearly mistaken claim that the Tsimshian were uninterested in property (Boas 1935), it is clear from these works that sorting out territories, boundaries, and property rights is one of the central functions of at least one type (adaawk) of Tsimshian oral traditions. We will see below that this is not the only interpretation that Tsimshian oral traditions has had since Boas, as these and other 'myths' have also been the subject of structural and post-structural analysis by Lévi-Strauss and others. There can be little doubt, however, of the importance of this functional reading of Tsimshian oral traditions in the context of contemporary land claims. Given that the 'land question' is not settled in the rest of British Columbia, this kind of analysis has important implications for future research from this perspective.

Coast Salish Cosmology, Power & Property

The dominant themes in describing and interpreting Coast Salish mythology in the post-Boasian ethnographies has been their importance in understanding cosmology and spirit power. Given the emphasis on inherited privileges and property in the Tsimshian (much of the other northern Northwest Coast) literature since Boas, this difference is striking. This difference may partially be explained by the influence of the then-unresolved Boasian hypothesis that the Coast Salish were relative newcomers to the Coast and their property systems were a mere shadow of envy of those held by Boas' Kwakiutl. Anthropologists working in the salvage mode and interested primarily in 'traditional culture' may not have been tuned in to listening for concerns about property. However, given that recording texts is supposed to alleviate ethnographers from the burdens of their interests, it does not seem like this would account for such a different ethnographic emphasis.

Some of the difference may be accounted by the particular historic situation of the timing of the field work. By the time trained anthropologists started to work in Coast Salish communities (the 1890s for the likes of Hill-Tout (1978) and Boas, the 30s to 50s for Jenness, Barnett, Collins and Codere), Coast Salish communities were not then actively involved in making protests over the land question. They had been very active in the 1870s (Tennant 1990), and again during the McKenna-McBride commission hearings in the 1910s. However, with the ban hiring attorneys for making legal arguments about land title brought these arguments to a temporary halt. This is in contrast to the Tsimshian communities who were actively engaged in protests continually through the early parts of this century (and indeed well into the end of it with the Calder and Delgamuukw cases and the Nisga'a Treaty) (Sterritt et al 1998). The anti-potlatch laws were also strictly enforced in Coast Salish communities, which were nearer the urban centres and the eyes of the law than their more northerly neighbours. In fact, it was in a Stó:lô (Coast Salish) community in Chilliwack that the first arrest and conviction under the anti-potlatch law was made (LaViolette 1973:70). Thus it was seywen (spirit dancing), cosmology and spirit power and not systems of property rights which caught the attention of this new generation of anthropologists interested in 'traditional culture'. In spite of these particular interests of the post-Boasian anthropologists working in Coast Salish communities, given the contemporary political context, it is important to see how Coast Salish communities may have framed oral traditions in terms of property, and how this is similar or different from the practices of other Northwest Coast groups.

Diamond Jenness produced two of the first post-Boasian ethnographies dealing with several Coast Salish communities, though one of these texts has languished as an unpublished typescript at the National Museum in Ottawa (now the CMC) since 1934 (Jenness n.d.; 1955). Jenness' unpublished manuscript reported on the Coast Salish communities of southeast Vancouver Island, where he had done several interviews with older men. His published account is the celebrated Faith of a Coast Salish Indian which is a verbatim reporting of a very long myth cycle told by the charismatic shaman Old Pierre (Jenness 1955). Jenness was always interested in collecting oral traditions, following the Boasian paradigm, as a means of understanding native cosmologies. Like Boas, Jenness was struck by the highly variable responses to his questions to informants about their beliefs and cosmology (Jenness n.d.: 105). However, by systematically describing the cosmological issues reported in the oral traditions he collected, Jenness was able to make several generalizations. Humans consist of a soul, thought, power and a shadow. Salmon have all of these, though the rest of the animals lack a soul. Other animate objects such as rocks, water, wind and stars have power and 'vitality' but no soul, thought or shadow. Also clear from these oral traditions was that humans and animals had corporeal frames (Jenness n.d.:108; 1955:35-37). Animals (and exceptional shamans) had the power to change their frame, as the salmon did every year when they returned to their homes out at sea and then back to salmon form for human food. First food rituals and the myths told around them are highly descriptive of these cosmological principles.

Though not the focus of his analysis, Jenness does make some observations about the connection between oral traditions and property. Some rituals and associated stories were held privately by the 'community' (it is unclear here if he means household, family or village as the property-holding group) (Jenness 1955:71). These private rituals and stories - collectively called 纜wtÑ'n - include mask dances, powerful dolls and stuffed animal rituals, supernatural fish, and funeral rituals. Jenness does not explicitly report these traditions as being linked to territory. However, the stories of the origins of many of specific ritual objects that may be used only by the family who owns the 纜wtÑ'n take place at specific, named locations to which those families have special association, including exclusive use of resources at these places. Additionally, in the long transformer-cycle myths recorded, the origins of particular 'communities' are recounted, with named individuals establishing communities at named locations (Jenness 1955:10-34). Like the Tsimshian adaawk, both the personal and place names are recognizable in modern, post-mythological times.

Homer Barnett, a student of Kroeber, was the next person to do substantial ethnographic work in Coast Salish communities during the 1930s. His primary interests were in collecting lists of cultural traits and collecting data for very traditional descriptive ethnography. Working entirely in English, he made little effort to collect oral traditions. However, stories were an important medium for his informants. Tommy Paul, a Saanich from Vancouver Island explained to Barnett about a violation taboo on telling the wrong stories (Barnett 1955:141). Stories are 'exclusive knowledge', particularly those about the original ancestors in the transformer-cycle myths. This exclusive knowledge (staalängan in Saanich, otakan in Comox) is "family owned and had to do with family history, myths and privileged acts".

One would not think of publicly claiming another's staalängan. He would be ridiculed unmercifully unless he could show a legitimate blood connection with its owners and could back up his claim with a property distribution (Barnett 1955:141).

This sounds very much like the 纜wtÑ'n described by Old Pierre to Jenness. Also similar to Jenness' observation on connections between myth, concept of the soul and spirit power is Barnett's discussion of a rite directed by a qualified ritualist to mark the proper date to start drying or smoking salmon. The ritualist had private knowledge of a myth of his ancestor's daughter who had been taken by a dog salmon as a wife. The daughter eventually returned to the village with the power to know when to begin smoking and drying salmon properly. This power has stayed with the family in the privately held myth (Barnett 1955:90). Though Barnett does not make much of this story, it is important to see how for the people who tell and hear these myths in an active social context, the myths have power of their own which have very real physical consequences. How much social and cultural context would the outside observer need to have the myth imbue its power? Barnett never recorded these stories, so from his work, this question cannot be answered.

In 1945, Helen Codere participated as a graduate student in an ethnographic field school out of Colombia University in Stó:lô territory. Out of this brief project Codere produced a number of important articles, one of which was a discussion of the Coast Salish swai'xwe mask and associated origin myths (Codere 1948). The swai'xwe is a mask and regalia which is danced by a man who has been granted the privilege from female relatives (affinal or consanguineal).⁽⁶⁾ Women whose families can claim the right to dance this mask trace the privilege from specific stories which tell about the origin of the mask at a particular place, and how it is that the family came to possess it. After providing a full recollection of the myth from three different tellers and ample ethnographic details surrounding its importance in the community, Codere goes on to posit a very unsatisfying analysis (see Suttles 1957), claiming the myth and mask resulted from a mixing of more sophisticated northern Northwest Coast ideas of 'crests' with the indigenous southern/central Northwest Coast ideas of spirit power. This is clearly a revival (or continuation) of Boas' theories of story-structure being a useful indicator of history, and the same criticism mentioned for Boas would apply here.

June Collins' research with the Upper Skagit (one of the Coast Salish peoples of Puget Sound) was ongoing from 1942 through to the 1970s. In an early article, Collins described the relationship between Skagit oral traditions and their attitudes towards animals (Collins 1952). She outlines how respect for animals is expressed as a relationship of reciprocity. Taboos are outlined in oral traditions set in the time of the myth-world, and examples are provided for how to be courteous to animals. The origins of the first salmon ceremony explained in these myth, and the philosophies behind the ritual practices of returning the bones of the first salmon to the river are described (Collins 1952:354). Like the ontological findings of Jenness, Collins discusses the relationship between the power of animals and humans to change form in myth and the actual experiences of contemporary Skagit individuals in their personal guardian sprit quest (ibid. 35). Another important ontological distinction is made visible in Collins' analysis of myth: that between food and non-food animals. Animals which can be eaten as food are also competing with humans in access to food: whales eat seals, as do humans; bears eat berries, as do humans (ibid. 353). These are the animals which marry humans (ibid. 356), thus the production-reproduction metaphor in these myths, and the necessity for respect. Those which are not eaten as food (lizards, salamanders, etc) are considered dangerous and are associated with shamanic powers in both myth and practice (ibid. 358).

After several decades of research, Collins published an ethnography of the Upper Skagit, where many of these ontological principles are restated (1974). Here, she considers the function of myth as history, with history being divided into two periods - mythological and folklore (ibid. 211). These historical periods are generally separated by the actions of the Transformer, who during the end of the myth-time set 'order', or natural laws in the world (ibid.). Beyond using these stories to mark historical time, Collins recognizes the importance of myth for the relationships people have with their guardian spirits and with each other. Myths have moral content which is often quite explicit. Violations of the moral values told of in myths have real world consequences in that the victims of the violation will be protected by their guardian spirits. This concrete relationship between the telling of moral stories and the actual physical (spiritual) danger of breaking these moral values create a cautious and circumspect social system of interactions between individuals (ibid. 214). This caution is extended from the unknown personal guardian spirits of other persons to the well-known, powerful spirits which live at particular places, which are also the subject of Upper Skagit myth and provide guidance for proper behaviour at places away from the community (ibid. 241).

More recently, Collins has asked how do people resolve for themselves the paradox of having a class-divided society that is essentially organized around kin units - where members of your own kin could occupy very different social statuses (Collins 1994). For Collins, this problem was resolved for the Coast Salish by borrowing traditions from the Kwakiutl, who it was argued at the time, had worked out a more complex system of social hierarchy than their more simple southern neighbours. Here we have Boas' pervasive argument again! However, the problem still remains, and may be better answered by the meanings bound up in the oral narratives themselves.

Sally Snyder developed this perspective of oral traditions as a means of expressing unresolved conflict three decades earlier in her highly interesting but almost always overlooked PhD dissertation (1964).⁽⁷⁾ In this comprehensive analysis of over 100 narratives she collected in Skagit communities, Snyder moves back and forth between a picture of social organization constructed from ethnography to the content of myths and stories which bears on social organization. In the latter, she finds social tensions abound (Snyder 1964:12). In-laws are caught in the tension between privilege and obligation. Women and older men are both positive and negative in that they have power. Myths chart courses through these social tensions, giving the listeners (whom she calls 'sophisticated') food for thought about the social tensions in their own lives (ibid. 59).

Like Jenness and Barnett, Snyder recognized myth as being a kind of private knowledge - 'advice' x^wdikw - owned by Coast Salish upper class families (1964:167-8, 210). Upper class genealogies are connected to oral traditions in the 'true stories' told which outline the rights and privileges belonging to that family. The advice given concerning myths is not only the narrative of myth itself (which is important and also held privately) but advice concerning the interpretation of morals and proper (upper-class) behaviour embedded in the myths. This concern with the private, proprietary nature of myths was in part responsible for her Skagit informants' concern to get the stories 'right'. They said they would rather stay silent than to guess at meanings, or "pad, improvise, paraphrase or omit" the telling a story (ibid. 21).

It is clear from this discussion of the Coast Salish material that these oral traditions serve an important function in expressing moral values and examples of how to (or how not to) have relationships with fellow community members (within and between genders and social classes) and with the spirit world. In the latter regard, these are not unlike myths told across North America. Of particular interest, though, is the role of Coast Salish myth as a discourse on property. Unlike the Tsimshian adaawk, the ethnographies of Coast Salish communities do not have a harmonious interpretation about the role of myth in the native discourse of ownership of property. The four terms presented (纜wtÑ'n (Katzie), staalängan (Saanich), otakan (Comox), x^wdikw (Skagit)) which could be analogous to the Tsimshian term adaawk, do not between them show clear similar cognates, even though they are all from related Salishan languages. However, what does seem clear is that the stories behind certain ritual practices are owned by families and that these stories also discuss original ancestors' resource locations. This seems to form a an analogous system to the Tsimshian adaawk (Thom 1998b), but this needs further clarification through fieldwork.

Structuralism and the Search for 'Meaning' in Northwest Coast Oral Traditions

The massive corpus of texts made by the Boas-Tate (Boas 1912, 1916) and Barbeau-Beynon (Barbeau 1961; Barbeau & Beynon 1987) collaborations has provided the basis for several major forays into structural analysis. When Claude Lévi-Strauss turned his eye to the Northwest Coast, it was the mythology collected by Boas and his collaborators to which Lévi-Strauss would turn. John Cove's dense hermeneutic exploration of Tsimshian mythology (Cove 1987) was based largely on these collections. These kinds of analysis use mythology as the core of their investigations, but unlike Boas' goals, are less interested in native ideas from the native point of view.

Lévi-Strauss' widely read analysis of one of these myths, the Tsimshian 'Story of Asdiwal" (Lévi-Strauss 1958, 1967) has been celebrated and debated (see Adams 1981:368-370 for a review). After the dust had settled, which took the better part of three decades, Lévi-Strauss' particular arguments about the structure of Northwest Coast mythology have been challenged from both an empirical perspective (ie: Adams 1974) and from the perspective of the internal logic of his argument (Codère 1974; Thomas, Kronenfeld & Kronenfeld 1976). Possibly the best summary of why Lévi-Strauss' structural approach to myth should be abandoned comes from Berman, who rejects

its inexplicable blindness to the importance of cultural provenience; its divorce of narrative form from the social, cultural and personal contexts in which that form is manifested; its preoccupation with discovering a single, simple universal type of narrative form; and finally its inability to demonstrate how analytical 'structure' or 'form' relates to the visual and aural 'texture' of actual oral-literary performances, to what is actually said and done in them. (Berman 1991:6)

Although the structural approach to the study of myth on the Northwest Coast has not been successful overall, there are some useful observations that can be made from examining the literature and some of the critiques. The debate provided an opportunity to bring into shaper focus some issues which had not been particularly well articulated, specifically the skill with which some myths explore the consequences of broken social rules.

For Lévi-Strauss, this Tsimshian myth provides resolution and balance where "contradictions" of cross-cousin marriages that Lévi-Strauss identifies in the story are all resolved (Lévi-Strauss 1967:25). Unconcerned with original narrative form, or the quality of translation, Lévi-Strauss defines on his own terms, four levels of symbolic opposition which, like music, operate on different levels of abstraction (ibid. 17-21). These opposites are not based on an ethnographically informed reading of the myth, nor on native views about the meaning of their own stories. Lévi-Strauss goes so far to claim that these levels cannot be separated by the native mind. These structural opposites serve the function of subconsciously resolving problems posed by actual contradictions in society (ibid. 28).

In critiquing the idea that the Tsimshian myth provides a symbolic balance, Adams says that Lévi-Strauss set up "a non-existent problem, found a solution to it (in a form of marriage) and then declared the solution inherently unstable, thus 'proving' his assertion that normative rules only appear to resolve inherent contradictions"(Adams 1974:171). Adams, whose reading of Tsimshian myth is informed by his own extensive ethnographic research in the area, has an alternate reading: no two people who share rights to the same resources (ie: who share kin ties) can marry each other. The only time that kin marriages can occur is when a person has been adopted into a different resource owning group than the one into which they were born. For Adams, this myth is not about resolving subconscious contradictions, but rather plays the very familiar role of title or charter of resource-owning groups. Ultimately, Lévi-Strauss sees that the form of the myth and its meaning are linked, but his methodology for describing the form of myth draws on his own imagination and cultural constructions, rather than being attentive to native literary style.

John Cove (1987) has presented a more recent investigation of symbolic meaning in Tsimshian mythology. Cove looks at the Tsimshian myths recorded by Tate, Boas, Beynon and Barbeau to build a model of the Tsimshian model of reality. Cove sorts out myths, which are common property (ibid. 46) from house narratives (adaawk), shamanic narratives, secret society narratives, legends and folktales. He suggest that these commonly known myths contain a central, integrated message for the Tsimshian who tell them. The message is a kind of answer to the 'basic' philosophical question of what does it mean to be human.

Cove discusses in detail over a hundred myths for what they reveal about being human. Tsimshian myth, he suggests, creates a model of humanity which makes the basic ontological distinctions that (1) all beings have soul, body and powers; (2) to be human is to be other-than-0n-human; and (3) humans have a continuing preoccupation with death. Cove looks at the symbolic significance of particular elements of myths -- a stone, the salmon, skin -- in order to get at these categories of human-ness. The stone represents immortality and fixedness in this world. The salmon are immoral and can powerfully move back and forth between their world and the world of humans. Skin represents transformation between human and non-human form. Cove reads meaning into these symbols by close exegesis of the content of the myths he has assembled, rather than in discussion with informants (who when asked, was not able to give him the kinds of philosophical answers he was looking for) as Jenness and Barnett did, or by seeing how myths were used by people to make meaning in actual living social situations.

The reading of Tsimshian mythology as being foundationally about the 'basic' question that has also pre-occupied Western philosophy seems remarkable in light of the wide array of characters, themes, and incidents in the stories. Cove himself recognizes that only a single story, the very short and enigmatic Elderberry and Stone (see Kroeber 1986 for an enlightened discussion of the performances of this narrative), addresses directly this 'basic' human dilemma. For the other stories, the symbolic meaning is not on the surface, but must be investigated by a hermeneutic process of going back and forth between readings to see how the strands of symbolic significance connect form one story to the next, creating a deeper web of meaning than the one on the surface.

In drawing his hermeneutic circles, however, Cove underplays other interpretations of the very stories he is trying to come to terms with. When his conclusion describe myth as being about one basic philosophical question, it places in the shadows the meanings that individual narrators and listeners might ascribe. Such a symbolic study freezes meaning, or risks claiming as illegitimate alternate readings of these stories. Cove's use of the authoritative third-person voice throughout the book reinforces this authoritative stance.

If one takes a different theoretical perspective and proposes that the ultimate meaning of oral traditions are completely contingent on the particular moment of the social context in which they are told, then such a hermeneutic analysis as given by Cove is unsatisfying. However, if like Cove, one considers that meaning is held in the semiotic relations of the literary elements within the story, then these oral traditions have meanings which stand on their own outside any particular use of them in social practice. Of course, every time an oral tradition is heard (or read) is in a social context of one kind or another, so taking such an extreme view will lead to the problems discussed for Lévi-Strauss. But if we take the highly post-structuralist (deconstructionalist) view that meaning only exists at the moment of practice, we neglect the important observation that language and discourses are structured in ways so that meanings can be shared and understood by groups of people. Otherwise, communication would be impossible. Thus, we are led to the middle ground of trying to see the relationship between structure and agency, between semiotic relationships and the momentary performance of an communicative act. Cove, I would suggest, does a better job than Lévi-Strauss in trying to understand agency, but could have spent more time sorting out the social contexts of recalling oral traditions to see more the more nuanced ways in which meaning unfolds.

An example from analysis of Coast Salish mythology can more clearly explicate the difference in these two theoretical positions concerning meaning. Jay Miller and Lushootseed story teller Vi Hilbert have published a short article which describes the symbolic meaning of Lushootseed myths (Miller & Hilbert 1996). These myths describe the origin of contemporary villages, each of which has an ancestor who dropped from the sky. Aside from providing satisfying 'origin stories', Hilbert says myths teach you to live with your mistakes. Such a brief, almost obvious interpretation of Lushootseed stories by Hilbert in an academic publication defies the complex relationship she has with their power and meaning. Hilbert has worked with linguist Thom Hess and anthropologist Crisca Bierwert for many years, both translating texts and providing personal insight into their meanings and contexts (Bierwert et al. 1996; Bierwert 1999). The thin explanation given in the academic publication is further illuminated by Hilbert's comments in her published collaboration with Bierwert et al.: "If I were to explain fully some incidents, I would disgrace myself. On these points, therefore, I ... trust the intelligence of our audience to perceive the full significance of the event" (Hilbert in Bierwert et al. 1996:4). Trusting the intelligence of the audience is a recurring theme in the presentation of Northwest Coast texts.

For Bierwert, the depth of meaning of Hilbert's Lushootseed stories becomes richer for the anthropologist who has made themselves familiar with the culture from which the stories emerge:

The depth and resonances of the stories are difficult to grasp at first because the narrations are thin in descriptive detail; they acquire depth on study and relistening and they are best at cuing a personally constructed image for someone who knows the culture (Bierwert 1999:147).

The problem becomes building cultural knowledge through personal experience. Does this mean that the texts in and of themselves have no meaning? Of course not. However the meanings that are ascribed to a text taken out of its social context are going to necessarily be 'thinner' or less socially potent than those texts which are analysed in active discourse. As a methodology, Bierwert suggests that if we can see something of the form and style of their original telling, at least some of the intended meaning will be clear to the new listeners/readers (Bierwert 1999:154). The greater the listeners knowledge of the rules form and style, the greater their own take on meaning. It is from a reading like this that we can more to more recent 'literary' approaches to Northwest Coast oral traditions.

Literary Approaches to Northwest Coast Oral Traditions

Out of this century of academic discourse on Northwest Coast discourses emerges a fairly concise set of theoretical and methodological concerns. Paramount among these is a concern with cultural meaning. As Morantz has asked, how can one presume to know the meaning of oral traditions when there is great social, temporal, cultural and linguistic distance between teller and interpreter (Morantz, in press). In many of the cases presented above, at least one of these interpretive barriers poses a central problem for an analysis of Northwest Coast myth. Since the important work of Dell Hymes in the early 1970s, several anthropologists and linguists working on the Northwest Coast have taken steps to address these problems. Some of these scholars have posited that oral narratives have been and continue to be an important part of native cultural and political discourses. Also important is the renewed recognition that there is an important connection between language and culture which must be attended to in the translation of oral traditions. However, their most important theoretical innovation has been to recognize that the performance of oral traditions, where culturally and linguistically encoded messages are transmitted, are critical to understanding their meanings.

Translation

When our theoretical concerns turn to meaning, the problem of translation becomes a critical issue. The translator is faced with the problem of not only providing correct morphological equivalents for the words uttered, but must give them shape in a syntax which reads well in the new language. This can be a problem where systems of syntax are very different. In Haida, (like other Northwest Coast languages) phrases and sentences are centred on verb, while in English they are centred on nouns. In making a smooth gloss translation into English, Bringhurst (1999) struggled with trying to evoke the motion and action being expressed in a verb-centred language. The English version given is clearly different in both form and meaning than another translation which was not so attentive to the shape of English in making this translation (ie: Swanton 1905). The distinction between nouns and verbs is similarly ambiguous in Kwakw'ala, making possible fine punning which is lost in translation without extra context being given by the translator (Berman 1992:143). The translator has to consider the kinds of meanings the audience will read into the term when translating it. Some terms may be best left untranslated, and explained in a footnote. Others may be translated using awkward or heavy phrasing in English to give it the same kind of sense. In other times the original sense may be simply lost or modified. Much, as we have seen in the discussion of poetic form, depends on the audience.

In the fall of 1999 a flurry of controversy blew up in prominent newspapers and magazines in Canada over the publication of a book of 'new translations' of Haida myths.⁽⁸⁾ The problem is rooted in the relationship between a non-native scholar and poet, Robert Bringhurst (whose book is a 'new' translation of Haida mythology collected by one of Boas' proteges, John Swanton) and members of the Haida community, who argue that their myths belong to particular houses who have exclusive rights to tell the stories. Bringhurst's high-profile, critically acclaimed publication and the controversy surrounding it raises a number of important issues. The book itself is centrally concerned with the problem of translating oral traditions across boundaries of time (100 years), medium (oral to written), languages (Haida to English), and, of course, culture. However, the main intended audience is the non-native, interested, sympathetic reader. The book is an excellent presentation of Haida history and oral narratives, and does take into account many of the problems of presenting native texts discussed here. However in excluding Haida people from the process, Bringhurst has alienated one of his most important audiences.

Bierwert et al (1996) have considered multiple audiences in their excellent presentation, translation and contextualiztion of Lushootseed Texts. Here, the audience is taken to be in three primary groups: (1) academic readers interested in Native oral literature in general and Salishan languages in particular; (2) Lushootseed community members who are interested in the stories or studying the Lushootseed language in a language program (3) a general audience of readers. To achieve a presentation that would appeal to all three audiences, the text was presented in a carefully transcribed, poetically organized Lushootseed form with a smooth English gloss (made at the level of the sentence) on the facing page. Explanatory notes for culturally and linguistically particular features are left out of the translation of the text (not adding anything significant at the level of the sentence) but rather provided in endnotes and in short contextualizing essays at the beginning of each story. Tapes are made available on request so that the stories can be heard as well as read. Significantly (and possibly uniquely) the copyright notice inside the cover of the book states: "The texts of the Lushootseed stories are understood to be part of native cultural tradition and therefore no claim of copyright is here made upon them." This is a work that clearly acknowledges the community as both owner and potential audience of the stories. Unlike Bringhurst's Story as Sharp as a Knife, community members (particularly renowned story teller Vi Hilbert) were intimately involved in producing the translations and explanations. In the recently published Lushootseed Dictionary, the editors (one of whom is linguist Thom Hess) has made cross references to the occurrence of words in these texts so that those who are looking to see words in action may cross-reference back and forth between the Dictionary and Texts. This is a rare kind of publication that is highly embedded in the community from which the stories emerged.

Several different schemes for presenting translations have been tried, including (1) English-only publication as translated by editor; (2) Native texts with technical morpheme-by-morpheme translation followed by smooth gloss; (3) native text and smooth English gloss on facing pages; (4) English-only publication as given by original teller. Several problems in meaning should be considered. In method (1), the text given is furthest from the original, with the most opportunities for mis-communications to occur. The second approach is the most widely practised; linguists are frequently the ones recording the texts. The morpheme analysis is an enormous task and often leaves little opportunity for further consideration of the context social meanings in the stories. However, they are the stuff of grammars and dictionaries and are vital academic productions. Tedlock (1983:31) has argued that the kinds of highly literal, morpheme-by-morpheme, phrase-by-phrase translations of Boas and his colleagues and students is for the most part "what professional translators would call a 'crib' or a 'trot' - not a true translation into literate English, but rather a running guide to the original text, written in an English that was decidedly awkward and foreign" (Tedlock 1983:31). The smooth English glosses produced in methods (2) and (3) raise their own interesting problem. They are intended to give a close translation of the native text but become an Native English not spoken by anyone, not the diction of native speakers using English, but the invented native English of the scholar working on the text. This might best be done in collaboration with a poet and native speaker to give the fullest and richest translation. The critically acclaimed Canadian poet Robert Bringhurst (1999) has risen admirably to this task in his translation of Haida oral traditions. The problem raised in method (4) is a matter not of 'authenticity', but rather that of the possibility of losing devices, and forms which are inherent in native language that gives stories a unique meaning in their own context. It may be that a native speaker is bilingual enough to reproduce these aspects in an English versions, but this has not been explored on the Northwest Coast like it has been by Cruikshank (1999) in the Yukon.

Richard and Nora Dauenhauer have long been active in recording, translating and publishing Tlingit oral narratives from the genres of life histories (Dauenhauer & Dauenhauer 1994), oratory (Dauenhauer & Dauenhauer 1990) and mythological oral narrative (Dauenhauer & Dauenhauer 1987). This work has formed an important body of accessible texts for students in Alaska wanting to learn and practice Tlingit. They typically present the Tlingit version, organized into verse form at the level of the phrase, and a smooth English gloss given on the facing page. Technical translations are left out of the texts, privileging the native original, not the scholarly goals of linguistic analysis. Extensive notes are included for each text to provide cultural and historical context, including extensive biographical information about the tellers and the event where the narratives were given and recorded. As a collection, these three large texts form a useful standard for language learning in a community which is already close to many of the cultural issues which the texts raise.

Richard Dauenhauer has published a number of scholarly analyses of Tlingit texts, some of which focus on the particular problems of translation. In an interesting, brief consideration of the problem of translating native texts, Dauenhauer turns to two of the texts of mythological songs recorded in Tlingit by John Swanton during his fieldwork under the direction of Franz Boas (Dauenhauer 1981). Dauenhauer demonstrates the process of making careful translation at the level of both phrase and morpheme. He pays particular attention to the complex set of Tlingit kinship terms, which often have no English equivalents. He shows how in Swanton and Boas' translations into English and German there is a fundamental loss in

...everything functional and contextual - the sense of what moved the singer to compose it, and a sense of the song's impact on the Tlingit audience. Such deletions of kinship terms are the chief cause of the anger and hostility knowledgeable Tlingits universally experience and generally express when they see such translations in print... From the Tlingit point-of-view, such versions are not only un-Tlingit, but anti-Tlingit. (Dauenhauer 1981:362)

These comments reflect the importance of oral traditions to contemporary native groups as part of their identity. Getting the words right means not only being respectful during field work process, but in on-going collaborative efforts with the community. Old-style anthropology where translations of texts are made thousands of miles away with little subsequent involvement from native speakers who are engaged in their communities are no longer politically welcome, nor do they 'cut the mustard' as useful academic translations. In addition to just being able to translate the language, it is clear that a knowledge of linguistics and a poetic sense of English are both required to do justice to a translation of native oral traditions.

Julie Cruikshank's important work on matters of translation suggests another alternative methodology. Cruikshank has worked for years with elder storytellers. She has recently concluded that the best method for her work is to document stories in English by storytellers who are bilingual. One reason for this conclusion is that by working in English, storytellers want to make sure that their intended listeners hear what they have to say. The English translations that are generally made from the more academic projects discussed above are more 'flat' than their original performance, while English renditions told by storytellers compensate for the shift in the language of performance (Cruikshank 1999:103-4). This methodology places the social life of the text as paramount to its meaning. Such is also the perspective of the Dell Hymes school of ethnopoetics.

Ethnopoetics & the Performance of Myth

The song/narrative, poetry/prose distinctions led Boas away from making observations about narrative form, in spite of his being interested in describing languages in their own processes and categories (Hymes 1999:85). Scholars working on the Northwest Coast in the generation after Boas had a keener interest in trying to describe the structure of the languages in their own terms. Jacobs, for instance, suggested that oral traditions might best be presented as theatre (Jacobs 1959b:7), rather than the straight prose presented in Boas' texts. Theatrical figuring of oral traditions would allow for more literary kinds of analysis which could understand the relationships between character, plot, and theme. Jacobs trained a generation of scholars who worked on the Northwest Coast, making new recordings, transcriptions and translations of these discourses.

The most well-known of these students is Dell Hymes. Hymes has built on Jacobs' literary approach to native oral traditions to lay the foundations for the school of the ethnography of speaking. Hymes is keenly interested in the importance of conventions of narrative form which generate meaning in oral traditions beyond the semantics of the words in the texts themselves. Hymes' analysis of Northwest Coast oral narratives has been worked mainly from Boas' texts, 're-presenting' them in verses, stanzas and acts to reflect something of their narrative structure (Hymes 1981). In Hymes' own analysis of these old texts, the main statements he has been able to make about inferring meaning have been in recognizing patterned use of culturally important numbers (like 3, 4, or 5) in narrative rules (Hymes 1999:92). Other scholars (like Dennis Tedlock) have tried to capture more of the performance aspects of oral narrative - like the particular evocative sounds and stresses in speech - in their transcriptions and translations, but these kinds of analysis have rarely been used on the Northwest Coast.

Respected linguist John Dunn, who has spent a career working on the Tsimshian language regards oral narrative as a kind of literature which can be usefully studied using the techniques of poetic analysis . This is an analysis of a 'first contact' narrative that is still owned by the house descending from one of the sisters of the main character in the narrative (Dunn 1989:396). Where this fairly typical Tsimshian oral narrative becomes literature is in the stichometric (lines and stanzas) and strophometric (aural performance features) structures. Lines can be identified in the story by the use of the Tsimshian marker "ada", (which is something like the English 'then') which occurs with great frequency during the telling of the story. The stories' stanzas form quatrains, which is significant as the Tsimshian culture number has been well established as being 'four'. The strophometric features include regular pauses after the use of 'then', dramatic length added to the syllables of some words and a regular pattern of cadences in the telling of the story. These performative features reveal an elaborate but regular set of parallelisms in the telling of this story and of stories of this class in general. However, beyond a detailed description of this structure and the recognition of the culture number pattern, Dunn draws little else from his analysis of this text as literature.

In contrast to Dunn's claim that Tsimshian narratives are a form of poetry, Halkomelem linguist Brent Galloway has made a careful examination of the narrative form of a Salishan Mink story to conclude that "there is no feeling of poetry here, but... artistry and tradition" (Galloway 1996:159). Rhymes and tight metres are what Galloway uses to distinguish a poem, but Halkomelem (and other Northwest Coast) narrative forms lack these features. Galloway describes several dominant narrative patterns (of discourse conjunctions, timing, repetition) from the story, but concludes that "there is not necessarily one, and only one, correct structure that can be proposed" (Galloway 1996:167). This long, technical analysis by Galloway shows that even if very careful attention is payed to form, (which in translation can help bring the story more authentically alive), a rigorous understanding of meaning may not ensue.

Poetic form has been attended to by other writers who do not concern themselves with trying to represent performance, as much as they want to represent something of the literary feeling of oral narrative. Brian Swann (1985), for example, has 're-presented' narratives collected by Swanton as free-verse poetry. Swann's argument is that though his work does not capture the literal or performance quality of the original oral traditions, he offers an accessible representation for a Western audience. Such representations mask real differences between cultural concepts embedded in the poems (Krupat 1992:15).

Presentations like these shy away from the often difficult task of either trying really to understand native languages, or work collaboratively (with all the power-sharing that this requires) with someone who does, as the collaboration between Bierwert, Hilbert, Hess, and Langen has masterfully done (Bierwert et al. 1996). The form of this text includes both poetic & prose arrangements of the native texts (depending on the style of the oratory), drawing on performance features, an English translation which reveals Lushootseed poetic structures on the facing page (ie: pauses, voice, changes in scene and shifts in focus/narrative repetitions are marked, false starts are ignored, and ordinary and inverted word orders are reflected in the translation), and extensive footnotes to provide linguistic and cultural annotations for the text given (ibid. 24). The editors suggest that this kind of presentation can "carry their reader farther into the organization of thought and the art of the Lushootseed language and storytellers" (ibid. 37) than more sparse translation projects might do.

In another effort to mediate this debate, Salishan linguist Anthony Mattina has also insisted that the "understanding which readers gain from the script is in direct proportion to what they know about the tradition and the context of the texts" (Mattina 1987:143). Thus the results of the labourious efforts to make poetic transcriptions which capture something of the form of the performance, can be equally well achieved using prose transcription. The prose transcriptions Mattina provides are highly accessible, yet maintains an authentic relationship to the speakers of the oral narratives. The key to these accessible translations are his attention to the use of "Red English" (a close-to-literal translation, with the use of the kinds of English words that would commonly be used by the native speakers). This technique is useful for traditions which are living, which continue to be told and recorded, as in the Haida stories presented by Eastman & Edwards (1991), who also give a closer line-by-line translation in the appendix to their volume of stories.

Form, Memory and Meaning

In an interesting analysis of form in Lushootseed stories, Langen recognizes a certain seeming paradox in the form of storytelling: the narratives often have very few words, but are highly repetitious. Langen asks us to think about how these two patterns of speech might interact with each other in the telling of Lushootseed narratives as a vehicle for information not declared (Langen 1989:80). Langen's first conclusion is that the very regular form of repetition in the oral narratives told by Lushootseed tellers likely serves as a mnemonic device for the complex set of incidences that are recounted in the long stories, that the storyteller "held the story in her mind as much by its shape as by its content" (Langen 1989:81). This mnemonic function is likely important, but still does not solve the problem of the paradox raised.

To understand how the laconic (but repetitive) telling of these stories can make sense to the listener, Langen compares them to the tellings done in English by story tellers who choose not to follow these narrative traditions so closely. Langen finds that in the stories which keep these narrative forms, a parallelism of actions and events is used by the narrator to highlight the deeds of the main character. In the same story told by the other tellers without holding strictly to these narrative conventions, a short 'sermon' was given at the end to capture the meaning of morals (ibid. 90). It would appear then, that such morals should become evident to those listeners who are attuned to the conventions of repetitive parallelisms. In a similar discussion, Langen critiques curriculum projects which have developed new tellings of Lushootseed stories and emphasizing plot (linear-thinking) rather than form (lateral-thinking) (Langen 1992). Langen suggests that Lushootseed story tellers working in their own language and in English use conventions of form to create richly embedded stories, but this is missed in following Western conventions of emphasizing linear developments in plot (Langen 1992:196).

Haida oral tradition also uses particular oratory devices to embed culturally-specific meanings (Boelsher 1991). Boelsher observed the operation of some of these practices in contemporary pubic discourse, which when practised with cultural outsiders, poses a potential problem of cross-cultural communication, where serious misunderstandings can occur from not recognizing narrative forms. She takes the example of political oratory where allusion, inversion, repetition, pausing and silence are important rhetorical devices. These are typical of small-scale, face-to-face societies, where it becomes possible to say things without actually articulating them. These ideas which are only hinted at would be lost on the outsider (ibid. 118-121). Thus when public oratory centres around land claims or fishing disputes, these matters of sensitive literary translation hive highly political consequences.

Boas and Ethno-Poetics?

These literary approaches to oral narrative have been discussed as providing a useful perspective from which to approach oral traditions made in close collaboration between the storyteller and the researcher. The texts produced are seen as being engaged in a highly 'active' social context, where meanings are negotiated between the tellers and the audience, sometimes through the interpreting medium of the translator, who would ideally operate with an almost-invisible, highly poetic touch. So what can be made of older recordings of stories? Can it be that there is no meaning that can be brought out from the shelves of texts recorded by Boas and his colleagues? Or would an investigation of 'meaning' in such texts be left to structural or symbolic analysts? There is one analysis which stands out as a highly successful attempt to draw on this literary perspective to investigate the problems of meaning in Boas' Kwakiutl texts, attending to the problems of the social context of the production of the texts, their social context as an ethno-literary category, the difficulties of negotiating both linguistically accurate and literate translation, and the connection between form and meaning in the texts. This is the excellent work of Judith Berman, who has drawn on these important insights into the analysis of oral narrative in a major re-evaluation of Boas' Kwakiutl texts (Berman 1991; 1992; 1994; 1996).

Berman proposes that we must examine how the texts came to exist, how they were translated, and build an ethnographic understanding of the role of texts as social objects. Boas, she asserts, knew that getting texts in the native language was important, but often made mistakes in his translations and more importantly said very little in the way of interpreting the patterns of performance (rhetorical) and genre form, which we now understand to be critical for interpreting meaning from stories (Berman 1992:126). In fact, in the case of the story "Oolachan-Woman's Robe", Boas could not have been more wrong in both his translation and interpretation of what Berman has shown to be a pun-filled, highly metaphorical, ribald story (Berman 1992).

In her PhD dissertation, Berman sets out to investigate systematic relationships between the form and rules of oral literature, and the cosmological concepts and cultural meanings expressed in the stories published by Boas. The analysis is substantial. She begins by giving a biographical review of Boas' collaborator George Hunt and the historical and social context for how the volumes of "Kwakiutl Texts" were generated (Berman 1991: 58-116; see also Berman 1994; 1996 which rework this section of her dissertation), and follows this with a robust description of 19^th century Kwagul social organization, based on a linguistically informed reading of Hunt's work with Boas. Berman looks to the myths to say something about the relationship between tribe-village-household-hearth (chapter 3), providing some new insight into the dynamic nature of these groups over time, commentary on social rank within and between groups, the chief as metaphor of the social body.

From this ethnographic position, she works her way back to conducting an analysis of myth by discussing Kwagul ethno-literary categories, placing the nuyºm into context with other genres of Kwagul oral narrative (see Table 2). These 'house-story' myths have few plot structures, but a great diversity of individual formation, being that they are "charters of social, political and personal identity..." like a "...birth certificate or passport" (Berman 1991:128-9). This observation is important, as myths have often been taken to be "Durkheimian collective representations that transcend the opinions and understandings of any one individual" which can be interpreted as forming "a neat symbolic or conceptual system" (Berman 1991:132). For Berman, however, the patterned variability of the text provides a better picture of the cosmological opinions held by Kwagul society (ibid.). Thus having deconstructed static or largely structural-symbolic interpretations of Boas' texts, Berman moves on to present a picture of what the patterns in narrative actually are, and how these may help in making more sound interpretations.

Berman examens one story closely, "Night Hunter and Day Hunter", starting with looking at the context of how it was produced by Hunt and (mis-)translated by Boas, then providing a re-translation, based on a highly sophisticated understanding of Kwakw'ala grammar (Berman is a linguist specializing in Wakashan languages), organizing the new translation into ethnopoetic notation to reveal its form of verses, stanzas, scenes, and parts. She describes patterning in (1) auxiliaries used, (2) performance form, (3) plot-types, zones and dramatis personae in nuyºm generally, (4) names & imagery used in performance, (5) thematic aspects of characters in nuyºm generally, (6) motifs in nuyºm generally, and finally (7) the symbolic oppositions posed in structure of genre (and corresponding forms in real life). This part of her analysis takes up some 476 pages of double-spaced text, and is a careful, well-illustrated linguistic analysis.

At the end of the dissertation, Berman comes back to the first question - what can be understood from Boas' texts, or this text in particular. In response to Boas' unspoken theoretical concern with the problem of relativity, Berman suggests that Hunt's texts do, in fact, present what is important and meaningful to the Kwagul - Hunt himself participated in Kwagul winter ceremonial life both as a dancer and in initiating his son. What is important and meaningful is a basic concern with reversal and reciprocity, "in which relations between human and sprit, power and powerlessness, giving and taking" are expressed in a variety of imaginative ways by different presenters of the myths told (Berman 1991:712).

This is a very satisfying analysis and seems to get to the heart of the many of the methodological and theoretical issues raised. However, weighty re-analyses like these speak mostly to an academic audience. They lack a level of accessibility that would make them useful for people not schooled in this kind of linguistic analysis. However, if the goal of this kind of work was to be accessible, the author would have to rely on the audience to bring the requisite knowledge to make an informed interpretation. Such an audience, presumably already fluent in the native language, would likely not need to have these very basic principles about language and narrative form stated. The anthropological project is about producing stories about stories in a context other than in which the original stories were presented. By attending to the problems of the social and political contexts of the stories, the manner in which they are translated produced as texts, and the relationship between form and meaning, the anthropological project can be very good indeed at bridging the cultural and linguistic gap that might otherwise make these 'other' discourses so difficult. However, as most of these efforts have produced anthropology which is accessible to very few others than other anthropologists, we need to ask if the job of bridging the cultural divide is being adequately done. Clearly there is room for more accessible but rigorous discussion of native discourses.

Conclusions: The Past & Future of Oral Tradition Research on the Northwest Coast

The anthropology of Northwest Coast oral traditions has been a long tradition, which has seen many changes in its theoretical goals and methods of analysis. Painted coarsely, the anthropology of Northwest Coast oral traditions has completed something of its own hermeneutic circle, starting with Boas' project of documenting native texts in their own languages and his inductive analyses, which was countered by the British-influenced project of looking for the function of oral traditions in Northwest Coast societies. Lévi-Strauss moved the discourse to a discussion of meaning and symbolic structure, which was countered by a return to a Boasian attention to individual 'performances' of oral traditions, but this time with an discussion of meaning through the interplay of narrative form and content. Though there are many problems which each approach, each brings important insight which is worth summing up here, together with suggestions for future research.

Distinguishing the genre or 'ethno-literary' categories of the oral traditions being studied has been a general practice in the studied considered. The usual two-fold distinction between 'myth' and 'legend' is clearly not sufficient in the Northwest Coast context, though it may easily be elicited from native vocabularies. So that appropriate questions are asked of the traditions being studied, a hierarchical classification scheme seems more appropriate, taking into account the social context in which the stories are told.

Boas' efforts transcribing texts in native languages has provided an essential methodology for learning principles of native grammars and to a certain degree, native world views. Boas' monumental efforts of publishing native language texts is obviously an important legacy now that many of these languages are endangered, moribund or extinct. Following this tradition is of utmost importance if these languages are to survive a transition into a textual world.

The prodigiousness of material left by Boas and his collaborators have been so influential in all of these studies that few are produced without at least reference to them, if not using them as the central set of 'data' for analysis. One wonders what the anthropological discourse on the Northwest Coast would have looked like had Boas spent his inexhaustible energies in the Southwest? However, given problems of translation and the frequent difficulty of seeing the social context for the performance of these collections of narratives, there is much room for scholars using this material to misinterpret the intent of the narrators and a more general pattern of cultural meaning. Continuing work like Berman's and Hymes' re-translations of Boas' efforts into a poetically sophisticated English is a worthwhile task. Translations of oral traditions work best when done collaboratively with researchers familiar with linguistic principles of native languages (and a poetic sense of their own language) in collaboration with fluently bilingual native speakers. This kind of work must not eclipse the more important task of continuing to work with holders of these traditions, listening to how they are used in native communities and providing new opportunities and venues -- from the court room to the class room -- for narrators to use oral traditions as a way of expressing themselves.

The other of Boas' major contributions was his attempts at grand comparisons of Northwest Coast oral traditions. Boas' Tsimshian Mythology stands unparalleled in seeing the patterns of motifs, episodes, and themes across Northwest Coast oral traditions (and again in larger contexts of North America and Siberia). Miller (1989) has drawn on Boas' analysis to make some general statements about the style and content of Northwest Coast mythology that can be held as some important defining elements of a vastly complex range of discourses. A few of these are worth repeating here. Concern with social ranking is shown through a mirroring of mythological worlds and human social statuses. Supernatural power is essential in acquiring food and prestige, with one of life's complexity being how that power is obtained, controlled and maintained. To this, we can add Adams' summary of the contribution of structural studies of Northwest Coast myths, which hold that "in-law relations are important and that the relations between husband and wife are considered to be particularly, even mystically close" (1981:380).

Symbolic and structural accounts have been heavily criticized for moving too far from the intellectual lives of the people whose meanings they purport to uncover. This critique leaves open the question of how is it that we can try to make sense of the oral traditions that have been recorded over the last century. One way to move between 'flat' historic narrations and translations of oral traditions and contemporary social contexts is by comparing the kinds of meanings they have held over time. Harkin (1998) provides a recent exploration of the persistence of symbolic metaphorical meaning in Nuu-chal-nulth oral traditions. Harkin's analysis shows how overcoming the whale in traditional narratives is a highly potent and pervasive metaphor for how leaders must be 'clean' in their own lives - overcoming their personal whales - to be respected by their communities. Harkin sees these oral narratives being important in Nuu-chal-nulth communities today, as leaders face new kinds of 'whales' in their efforts in co-management and self-determination. This approach to narratives is refreshing in appreciating the complexity of historic cultural patterns, and contemporary ways in which oral traditions continue to hold power and meaning for community members. More work should be done to see the contemporary social role of oral traditions and to see how these play off of other (ie: popular culture, contemporary politics, school curriculum) discourses. Studies like Boelsher's work with contemporary Haida political discursive practice (1988) are highly relevant in the current political milieu where 'native culture' is drawn on to make arguments for social justice and equality.

Finally, we must ask what have been the contributions of the literary turn in anthropology to the study of Northwest Coast oral traditions. Clearly, they bring a careful and creative methodology to the processes of translation and in the discovery of the relationships between narrative form and the range of meanings expressed in a story. These literary perspectives remind us of Boas' cautions in ascribing certain singular meanings to narratives, and move us towards an appreciation of the complexity and interplay of language and culture, tradition and artistry. They also encourage anthropology to engage in conversation with the people we study, to see how our work is seen by others and to appreciate that these stories and narratives have social lives outside the scope of our academic discussion. If we read the end-pages of these books (ie: Enrico 1995, Bierwert et al 1996), they provide some moral guidance on issues of cultural property rights and copyright, sometimes taking the indigenous social contexts of the stories being reported to heart, bringing them along with their messages and artistry into the textual world.

In all this discussion of Northwest Coast oral traditions, I have not brought any of the stories themselves forward. This has been a critical study of an anthropological tradition and less so about a set of Northwest Coast narratives themselves. I hope now that we have reflected on these treatments of Northwest Coast oral traditions that we can return again to the stories -- the Raven and the Transformer, the Boy who received his mask and spirit power from the lake, the Mouse who would not marry Beaver and Snake -- and read, hear or tell them with a new hope and appreciation for how they can be meaningful and powerful in all their social lives to come.

Bibliography

Adams, John. (1974). Dialectics and Contingency in 'The Story of Asdiwal.' An Ethnographic Note. In Ino Rossi (ed.), The Unconscious in Culture: The Structuralism of Claude Lévi-Strauss in Perspective. (pp. 170-178). New York: E.P. Dutton.

Adams, John. (1981). Recent Ethnology of the Northwest Coast. Annual Review of Anthropology, 10, 361-392.

Ames, Kenneth, & Herbert Maschner. (1999). Peoples of the Northwest Coast: Their Archaeology and Prehistory. London: Thames & Hudson.

Andrade, Manuel. (1931). Quileute Texts. New York: Columbia University Press.

Barbeau, Marius. (1917). Review of Boas' 'Tsimshian Mythology'. American Anthropologist, 19, 548-563.

Barbeau, Marius. (1961). Tsimsyan Myths. Ottawa: Anthropological Series 54, National Museum of Canada Bulletin, 174.

Barbeau, Marius, & William Beynon. (1987). Tsimshian Narratives. Ottawa: Canadian Museum of Civilization.

Barnett, Homer. (1955). The Coast Salish of British Columbia. Eugene: University of Oregon Press.

Berman, Judith. (1991). Dissertation. The Seals' Sleeping Cave: The Interpretation of Boas' Kwakw'ala Texts. Department of Anthropology, University of Pennsylvania;

Berman, Judith. (1992). Oolachan-Woman's Robe: Fish, Blankets, Masks, and Meaning in Boas's Kwakw'ala Texts. In Brian Swann (ed.), On the Translation of Native American LIteratures. (pp. 125-162). Washington: Smithsonian Institution Press.

Berman, Judith. (1994). George Hunt and the Kwak'wala Texts. Anthropological Linguistics, 36, 483-514.

Berman, Judith. (1996). 'The Culture as it Appears to the Indian Himself': Boas, George Hunt, and the Methods of Ethnography. In George Stocking (ed.), Volksgeist as Method and Ethic: Essays on Boasian Ethnography and the German Anthropological Tradition. (pp. 215-256). Madison: University of Wisconsin Press.

Bierwert, Crisca. (1996). Lushootseed Texts: An Introduction to Puget Salish Narrative Aesthetics. Lincoln: University of Nebraska Press.

Bierwert, Crisca. (1999). Brushed by Cedar, Living by the River: Coast Salish Figures of Power. Tucson: University of Arizona Press.

Boas, Franz. (1888). Myths and Legends of the Çatloltq of Vancouver Island. The American Antiquarian, 10, 201-211-366-373.

Boas, Franz. (1889). Notes on the Snanaimuq. American Anthropologist, 2, 321-328.

Boas, Franz. (1891). Second General Report on the Indians of British Columbia: The Lku'ñgEn. Sixth Report on the North-Western Tribes of Canada. Report of the British Association for the Advancement of Science, 1890, 582

Boas, Franz. (1894). The Indian Tribes of the Lower Fraser River, Ninth Reort on the North-Western Tribes of Canada. Report of the British Association for the Advancement of Science, 1894, 463

Boas, Franz. (1895a). Indian Myths and Legends from the North Pacific Coast of America. Bouchard, Randy and Kennedy, Dorothy (eds). Translated by Dietrich Bertz, from Verhandlungen der Berliner Gesellschaft für Anthropologie, Ethnologie und Urgeschichte. Victoria, BC Indian Language Project.

Boas, Franz. (1895b). Salishan Texts. Proceedings of the American Philosophical Society, 34, 31-48.

Boas, Franz. (1898a). The Mythology of the Bella Coola Indians. New York: American Museum of Natural History.

Boas, Franz. (1898b). Traditions of the Tillamook Indians. Journal of American Folklore, 11, 28-38-133-150.

Boas, Franz. (1898c). Introduction to James Teit 'Traditions of the Thompson Indians of British Columbia'. In Franz Boas (ed.), Memoirs of the American Folklore Society 6. (pp. 1-18). New York:

Boas, Franz. (1910). Kwakiutl Tales. Columbia University Contributions to Anthropology. Volume 2. New York, New York.

Boas, Franz. (1912). Tsimshian Texts (New Series). Publications of the American Ethnological Society, 3, 65-285.

Boas, Franz. (1916). Tsimshian Mythology. Based on Texts Recorded by Henry W. Tate. 31st Annual Report of the bureau of American Ethnology for the Years 1909-1910, 29-1037.

Boas, Franz. (ed) (1917). Folk-tales of Salishan and Sahaptin Tribes. Memoirs of the American Folk-Lore Society, Number 11. New York: New York.

Boas, Franz. (1935). Kwakiutl Culture as Reflected in Mythology. New York: American Folk-Lore Society, Memoir 28.

Boas, Franz. (1940). Review of G.W. Locher, 'The Serpent in Kwakiutl Religion: A Study of Primitive Culture'. In Franz Boas (ed.), Race, Language and Culture. (pp. 446-450). Chicago: University of Chicago Press.

Boas, Franz. (1981). Indian Folktales from British Columbia, Translated by Dietrich Bertz from Indianische Sagen von der Nord-Pacifischen Küste Amerikas. The Malahat Review, 45-77.

Boas, Franz, & George Hunt. (1905). Kwakiutl Texts. Memoirs of the American Museum of Natural History. Volume 5.. New York, New York.

Boas, Franz, & George Hunt. (1906). Kwakiutl Texts. Publications of the Jesup North Pacific Expedition. Volume 10. New York, New York.

Boelscher, Marianne. (1988). The Curtain Within: Haida Social and Mythical Discourse. Vancouver: University of British Columbia Press.

Boelscher-Ignace, Marianne. (1991). Haida Public Discourse and its Social Context. The Canadian Journal of Native Studies, 11, 113-135.

Bringhurst, Robert. (1999). A Story as Sharp as a Knife: The Classical Haida Myth-tellers and Their World. Vancouver: Douglas & McIntyre.

Codere, Helen. (1948). The Swai'xwe Myth of the Middle Fraser River: The Integration of Two Northwest Coast Cultural Ideas. Journal of American Folklore, 61, 1-18.

Codère, Hélène. (1974). La Geste du Chien d'Asdiwal: The Story of Mac. American Anthropologist, 76, 42-47.

Collins, June. (1952). The Mythological Basis for Attitudes Towards Animals Among Salish-Speaking Indians. Journal of American Folklore, 65, 353-359.

Collins, June. (1974). Valley of the Spirits: The Upper Skagit Indians of Western Washington. Seattle: University of Washington Press.

Collins, June. (1994). Kinship, Social Class, and Religion of Northwest Coast Peoples. In Raymond DeMallie & Alfonso Ortiz (eds.), North American Indian Anthropology: Essays on Society and Culture. (pp. 82-107). Norman: University of Oklahoma Press.

Cove, John. (1987). Shattered Images: Dialogues and Meditations on Tsimshian Narratives. Ottawa: Carleton University Press.

Cranmer-Webster, Gloria. (1997). Consumers, Then and Now. Paper Presented at the Centenary Conference: Constructing Cultures Then and Now,The Legacy of Franz Boas and the Jesup North Pacific Expedition (1897-1902), 1-10.

Cruikshank, Julie. (1999). The Social Life of Texts: Editing on the Page and in Performance. In Laura Murray & Keren Rice (eds.), Talking on the Page: Editing Aboriginal Oral Texts. (pp. 97-119). Toronto: University of Toronto Press.

Dauenhauer, Richard. (1981). Notes on Swanton Numbers 80 and 81. Journal of American Folklore, 94, 358-364.

Dauenhauer, Nora Marks, & Richard Dauenhauer. (1987). Haa Shuká, Our Ancestors: Tlingit Oral Narratives. Seattle: University of Washington Press.

Dauenhauer, Nora Marks, & Richard Dauenhauer. (1990). Haa Tuwunáagu Yís: for Healing Our Spirit, Tlingit Oratory. Seattle: University of Washington Press.

Dauenhauer, Nora Marks, & Richard Dauenhauer. (1994). Haa Kusteeyí: Our Culture, Tlingit Life Stories. Seattle: University of Washington Press.

Duff, Wilson. (1959). Histories, Territories and Laws of the Kitwancool. Anthropology in British Columbia, Memoir, 4, 1-45.

Dunn, John. (1989). Tsimshian Poetics. In Mary Keyand & Henry Hoenigswald (eds.), General and Amerindian Ethnolinguistics: In Remembrance of Stanley Newman. (pp. 397-406). New York: Mouton de Gruyter.

Eastman, Carol, & Elizabeth Edwards. (1991). Gyaehingaay: Traditions, Tales and Images of the Kaigani Haida. Seattle: University of Washington Press.

Farrand, Livingston. (1902). Traditions of the Quinalut Indians. Memoirs of the American Museum of Natural History. (Pp. 77-132). Volume 4, Number 3. New York, New York.

Farrand, Livingston, & Theresa Mayer. (1919). Quileute Tales. Journal of American Folklore, 32, 251-279.

Finnegan, Ruth. (1992). Oral Traditions and the Verbal Arts: A Guide to Research Practices. London: Routledge.

Galloway, Brent. (1993). A Grammar of Upriver Halkomelem. Berkeley: University of California Press.

Galloway, Brent. (1996). An Upriver Halkomelem Mink Story: Ethnopoetics and Discourse Analysis. In Papers for the 31st International Conference on Salish and Neighboring Languages. Vancouver: ICSNL.

Garfield, Voila. (1966). The Tsimshian and Their Neighbours. In Viola Garfield & Paul Wingert (eds.), The Tsimshian Indians and Their Arts. (pp. 3-70). Seattle: University of Washington Press.

Gunther, Erna. (1925). Klallam Folk Tales. University of Washington Publications in Anthropology, 1, 113-170.

Haeberlin, Herman. (1924). Mythology of Puget Sound. Journal of American Folklore, 37, 371-438.

Harkin, Michael. (1998). Whales, Chiefs, and Giants: An Exploration into Nuu-Chah-Nulth Political Thought. Ethnology, 37, 317-332.

Hill, Jane. (1999). The Meaning of Writing and Text in a Changing Americanist Tradition. In Lisa Philips Valentine & Regna Darnell (eds.), Theorizing the Americanist Tradition. (pp. 181-194). Toronto: University of Toronto Press.

Hill-Tout, Charles. (1978). The Salish People: The Local Contribution of Charles Hill-Tout (in Four Volumes). Vancouver: Talonbooks.

Hymes, Dell. (1981). 'In Vein I Tried to Tell You' Essays in Native American Ethnopoetics. Philadelphia: University of Pennsylvania Press.

Hymes, Dell. (1990). Mythology. In Wayne Suttles (ed.), Handbook of North American Indians, Volume 7, Northwest Coast. (pp. 593-601). Washington: Smithsonian Institution Press.

Hymes, Dell. (1996). Coyote, the Thinking (Wo)man's Trickster. In A. James Arnold (ed.), Monsters, Tricksters and Sacred Cows: Animal Tales and American Identities. (pp. 108-137). Charlottesville: University Press of Virginia.

Hymes, Dell. (1999). Boas on the Threshold of Ethnopoetics. In Lisa Philips Valentine & Regna Darnell (eds.), Theorizing the Americanist Tradition. (pp. 84-107). Toronto: University of Toronto Press.

Jacobs, Melville. (1959a). Folklore. In Walter Goldschmidt (ed.), The Anthropology of Franz Boas. (pp. 119-138). Menasha, Wis.: American Anthropological Association.

Jacobs, Melville. (1959b). The Content and Style of an Oral Literature. New York: Viking Fund Publications in Anthropology.

Jenness, Diamond. (n.d.) Coast Salish Mythology. VII-G-9M, Box 39, F.1. Hull, Canadian Museum of Civilization.

Jenness, Diamond. (1955). Faith of a Coast Salish Indian. Victoria: British Columbia Provincial Museum.

Jorgensen, Joseph. (1980). Western Indians: Comparative Environments, Languages, and Culture os 172 Western American Indian Tribes. San Francisco: W. H. Freeman.

Kinkade, M. Dale. (1992). Translating Pentlatch. In Brian Swann (ed.), On the Translation of Native American Literatures. (pp. 163-175). Washington: Smithsonian Institution Press.

Kinkade, M. Dale, & Anthony Mattina. (1996). Discourse. In Ives Goddard (ed.), Handbook of North American Indians, Volume 17, Languages. (pp. 244-274). Washington: Smithsonian Institution Press.

Kroeber, Karl. (1986). Elderberry and Stone. Studies in American Indian Literature, 10, 38-42.

Krupat, Arnold. (1992). On the Translation of Native American Song and Story: A Theorized History. In Brian Swann (ed.), On the Translation of Native American LIteratures. (pp. 3-32). Washington: Smithsonian Institution Press.

Langen, Toby C. S. (1989). The Organization of Thought in Lushootseed (Puget Salish) Literature: Martha Lamont's 'Mink and Changer'. Melus, 16, 77-93.

Langen, Toby C. S. (1992). Translating Form in Classical American Indian Literature. In Brian Swann (ed.), On the Translation of Native American Literatures. (pp. 191-207). Washington: Smithsonian Institution Press.

LaViolette, Forest. (1973). The Struggle for Survival: Indian Cultures and the Protestant Ethic in B.C. Toronto: University of Toronto Press.

Lévi-Strauss, Claude. (1958). La Geste d'Asdiwal. In Anonymous, Extrait de l'Annuaire, 1958-9, Ecole Prâtique des Hautes Etudes. (pp. 2-43). Paris: Section des Sciences Religeuses, Ecole Prâtique des Hautes Etudes.

Lévi-Strauss, Claude. (1967). The Story of Asdiwal. In Edmund Leach (ed.), The Structural Study of Myth and Totemism. (pp. 1-47). London: Tavistock.

Malinowski, Bronislaw. (1948). Myth in Primitive Psychology. In Bronislaw Malinowski (ed.), Magic, Science & Religion and Other Essays. (pp. 72-124). Boston: Beacon Press.

Matson, R. G., & Gary Coupland. (1995). The Prehistory of the Northwest Coast. New York: Academic Press.

Mattina, Anthony. (1987). North American Indian Mythography: Editing Texts for the Printed Page. In Brian Swann & Arnold Krupat (eds.), Recovering the Word: Essays on Native American Literature. (pp. 129-148). Berkeley: University of California Press.

Maud, Ralph. (1982). A Guide to B.C. Indian Myth and Legend. Vancouver: Talonbooks.

Maud, Ralph. (1989). The Henry Tate-Franz Boas Collaboration on Tsimshian Mythology. American Ethnologist, 16, 158-162.

McIlwraith, T. F. (1948). The Bella Coola Indians (2 Volumes). Toronto: University of Toronto Press.

Miller, Jay. (1989). Northwest Coast Mythology. Northwest Anthropological Research Notes. 20, 263-270.

Miller, Jay, & Vi Hilbert. (1996). Lushootseed Animal People: Mediation and Transformation from Myth to History. In A. James Arnold (ed.), Monsters, Tricksters and Sacred Cows: Animal Tales and American Indian Identities. (pp. 138-156). Charlottesville: University of Virginia Press.

Morantz, Toby. (in press). Plunder or Harmony? On Merging European and Native Views of Early Contact. In Germaine Warkentin (ed.), De-centring the Renaissance: New Essays on Early Modern Canada. Toronto: University of Toronto Press.

Ridington, Robin. (1999). Theorizing Coyote's Cannon: Sharing Stories with Thomas King. In Lisa Philips Valentine & Regna Darnell (eds.), Theorizing the Americanist Tradition. (pp. 19-37). Toronto: University of Toronto Press.

Sapir, Edward. (1912). Review Article. Kwakiutl Tales, by Franz Boas. Current Anthropological Literature, 1, 193-198.

Sapir, Edward. (1924). The Rival Whalers, a Nitinat Story (Nootka Texts with Translation and Grammatical Analysis). International Journal of American Linguistics, 3, 76-102.

Sapir, Edward, & Morris Swadesh. (1939). Nootka Texts: Tales and Ethnological Narratives, with Grammatical Notes and Lexical Material. Philadelphia: University of Pennsylvania, Linguistic Society of America.

Sarris, Greg. (1993). Keeping Slug Woman Alive: A Holistic Approach to American Indian Texts. Berkeley: University of California Press.

Seguin, Margaret. (1985). Interpretive Contexts for Traditional and Current Coast Tsimshian Feasts. Ottawa: National Museum of Man.

Sherzer, Joel. (1987). A Discourse-Centered Approach to Language and Culture. American Anthropologist, 89, 295-309.

Snyder, Sally. (1964). Dissertation. Skagit Society and its Existential Basis: An Ethnofolkloristic Reconstruction. Department of Anthropology, University of Washington;

Snyder, Sally. (1968). Stylistic Stratification in an Oral Tradition. Anthropologica, 10, 235-259.

Snyder, Sally. (1975). Quest for the Sacred in Northern Puget Sound: An Interpretation of Potlatch. Ethnology, 14, 149-161.

Sterritt, Neil, Susan Marsden, Robert Galois, Peter Grant, & Richard Overstall. (1998). Tribal Boundaries in the Nass Watershed. Vancouver: UBC Press.

Stocking, George. (1968). Race, Culture and Evolution. New York: The Free Press.

Suttles, Wayne. (1990). Introduction. In Wayne Suttles (ed.), Handbook of North American Indians, Volumve 7, Northwest Coast. (pp. 1-15). Washington: Smithsonian Institution Press.

Suttles, Wayne. (1957). The 'Middle Fraser' and 'Foothill' Cultures: A Criticism. Southwestern Journal of Anthropology, 13, 156-183.

Suttles, Wayne. (1991). Streams of Property, Armor of Wealth: The Traditional Kwakiutl Potlatch. In Aldona Jonaitis (ed.), Chiefly Feasts: The Enduring Kwakiutl Potlatch. (pp. 71-134). Vancouver: Douglas & McIntyre Press.

Swann, Brian. (1985). Song of the Sky: Versions of Native American Songs and Poems. Ashuelot: Four Zoas Night House Press.

Swanton, John. (1905). Haida Texts and Myths, Skidegate Dialect. Bureau of American Ethnology Bulletin. Volume 29. Washington.

Swanton, John. (1908). Haida Texts, Masset Dialect. Memoirs of the American Museum of Natural History. Volume 14. New York.

Tedlock, Dennis. (1983). The Spoken Word and the Work of Interpretation. Philadelphia: University of Pennsylvania Press.

Teit, James. (1898). Traditions of the Thompson River Indians of British Columbia. Memoirs of the American Folk-lore Soceity, 6, 1-137.

Teit, James. (1912). Mythology of the Thompson Indians. Memoirs of the American Museum of Natural History. Volume 12. New York.

Tennant, Paul. (1990). Aboriginal Peoples and Politics: The Indian Land Question in British Columbia, 1849-1989. Vancouver: UBC Press.

Thom, Brian. (1998a). Aboriginal Rights and Title in Canada after Delgamuukw. Bibliographic Essay, McGill University.

Thom, Brian. (1998b) Coast Salish Transformation Stories: Kinship, Place and Aboriginal Rights and Title in Canada. Paper Presented to the Canadian Anthropology Society, Toronto.

Thomas, L. L., J. Z. Kronenfeld, & D. B. Kronenfeld. (1976). Asdiwal Crumbles: A Critique of Lévi-Straussian Myth Analysis. American Ethnologist, 3, 147-174.

Thompson, Laurence. (1979). Salishan and the Northwest. In Lyle Campbell & Marianne Mithnun (eds.), The Languages of Native North America: Historical and Comparative Assessment. (pp. 692-765). Austin: University of Texas Press.

Valentine, Lisa Philips, & Regna Darnell. (1999). Introductions: Timely Conversations. In Lisa Philips Valentine & Regna Darnell (eds.), Theorizing the Americanist Tradition. (pp. 3-18). Toronto: University of Toronto Press.

1. In this review of the Northwest Coast, I have limited myself to an examination of the 'central' and 'northern' Northwest Coast, omitting the literature for the 'southern' region from the mouth of the Columbia River down into northern California. The extent of the "culture area"of the Northwest Coast is an old, probably unresolvable debate in the literature (see for instance Suttles 1990:1) which I do not intend to tackle here (though Jorgenson's Western Indians (1980:95) takes the idea of a culture area seriously and sees a 75% similarity over 292 cultural features in the northern region, 68% in the central and 74% in the southern, but only 55% for the region as a whole, with the greatest variation being between the southern area and the other two). From my own perspective, the large body of literature which has developed south of the Columbia River would add a level of complexity unnecessary for the questions I am asking here.

2. Oral traditions of Northwest Coast native people have also been at the core of the current political and legal debates in Canada over aboriginal title and rights. Since the Gitksan and Wet'suwet'en presented their oral traditions as evidence in the Delgamuukw case, scholars, legal experts and policy makers have had to grapple with the implications of these discourses. This particular important debate has been treated at length in an earlier paper (Thom 1998a) and though referred to, will not be reviewed again here.

3. A note on synonymy: In my own discussions, I use the contemporary names for social groups as defined by the communities themselves. However, in an effort to be true to the sources being reviewed, I have used the authors' terms for native communities they describe. Often, different terms are used for the same community. Authors use language names to describe socio-political groups. Also frequent is the use of smaller group (local descent group, village) names which may not be as familiar as their regional (ethnographic) name. A correspondence table is given in the appendix to clarify the often confusing naming practices of the discourses on the Northwest Coast.

4. Published texts of oral traditions from Coastal Salishan languages (my dissertation topic) include collections in Bella Coola (Boas 1895a, 1895b, 1898a, 1981), Island Comox (1888, 1895a, 1981), Klahuse (1895a), Sliammon (1895a), the now extinct Pentlatch (Boas 1895a; Kinkade 1992), Squamish (1895a), Island Halkomelem (1895a, 1889, 1981), Upriver & Downriver Halkomelem (1894, 1895a), Straits Salish (1891, 1895a, 1981), Tillamook (1898b), and a comparative analysis of Coast Salish Transformer stories (1916:586-610).

5. In some of his early English-only publications, the editor of the journal Boas published in made complete deletions from the translation of the text, indicated by a string of asterisks, because of "certain expressions which are not suitable for publication" (Boas 1888:206).

6. This has been of interest as one of the only matrilineal inheritance practices of the dominantly bilateral central and southern Northwest Coast peoples. This is in contrast to the more well known matrilineal property-holding groups of the northern Northwest Coast.

7. Unfortunately, only one short publication on her theoretical perspective and methodology for understanding oral tradition came out of the dissertation work (Snyder 1968) and another article on the potlatch (Snyder 1975), which has been reviewed very favourably (Adams 1981:376), but has received little subsequent attention. Snyder's work deserves closer attention in Northwest Coast studies, particularly her insights into the lives and social position of women.

8. Anne Moon, Whose culture is it, anyway?, Times Colonist (Victoria), June 20, 1999, p.10; Hugh Brody, Pilgrimage to the poem: Using raw materials collected nearly a century ago, an accomplished Canadian storyteller takes a journey to the cultural heart of the Haida nation, National Post, July 6, 1999, p.B11; Hans Werner, The soul of Haida Gwaii, The Toronto Star, July 11, 1999; Mark Abley, Haida tales brought back to life: B.C. poet translates record of oral verse made by anthropologist The Gazette (Montreal), April 10, 1999, p.J4; Candace Fertile, Haida mythology echoes inclusively, Calgary Herald, July 3, 1999, p.D7; Norbert Ruebsaat, Tricking the Trickster, Vancouver Sun, May 29, 1999, p.E11; The myths and the white man: experts on the stories of the Haida First Nation are infuriated by Vancouver poet Robert Bringhurst's new book, Globe & Mail, Nov 15, 1999, p.C3; John Bemrose, The timely wisdom of traditional tales: the penetrating beauty and great art of Haida poetry shines through in new translations by West Coast poet and linguist Robert Bringhurst, Maclean's, v.112(28) July 12, 1999 pg 56; Val Ross, Lost masterpieces regained Globe & Mail, July 17, 1999 pg. C8.