UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The tsimshian crest system : a study based on museum specimens and the Marius Barbeau and William Beynon… Halpin, Marjorie Myers

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1973_A1 H34_5.pdf [ 37.35MB ]
Metadata
JSON: 831-1.0101054.json
JSON-LD: 831-1.0101054-ld.json
RDF/XML (Pretty): 831-1.0101054-rdf.xml
RDF/JSON: 831-1.0101054-rdf.json
Turtle: 831-1.0101054-turtle.txt
N-Triples: 831-1.0101054-rdf-ntriples.txt
Original Record: 831-1.0101054-source.json
Full Text
831-1.0101054-fulltext.txt
Citation
831-1.0101054.ris

Full Text

I5I2X THE TSIMSHIAN CREST SYSTEM: A STUDY BASED ON MUSEUM SPECIMENS AND THE MARIUS BARBEAU AND WILLIAM BEYNON FIELD NOTES by MARJORIE MYERS HALPIN B.A. , The George Washington U n i v e r s i t y , U.S.A., 1962 M.A., The George Washington U n i v e r s i t y , U.S.A., 1965 1 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF I DOCTOR OF PHILOSOPHY i n the Department of ANTHROPOLOGY AND SOCIOLOGY We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH .COLUMBIA A p r i l , 1973 In presenting t h i s thesis i n p a r t i a l fulfilment of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t freely available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the Head of my Department or by his representatives. I t i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of dA^XJO^/^dJljD (P^ The University of B r i t i s h Columbia Vancouver 8, Canada i ABSTRACT This thesis i s about the r e l a t i o n s h i p of a r t and so c i e t y . Spe-c i f i c a l l y , i t inv e s t i g a t e s Tsimshian crest a r t and i t s r e l a t i o n s h i p to s o c i a l organization. The a n a l y t i c a l framework i s s t r u c t u r a l , with explana-tory formulations derived i n part from the wr i t i n g s of Claude Levi-Strauss and V i c t o r Turner. The study i s unusual i n that i t i s based upon museum specimens and records, data not often perceived as amenable to treatment wit h i n the context of contemporary s o c i a l anthropology. I t i s also the most systematic examination of Northwest Coast iconography yet undertaken. The data include the f i e l d notes of Marius Barbeau and William Beynon, c o l l e c t e d from the Tsimshian between 1914 and 1957, and preserved i n the National Museum of Man. These data were used to construct an i c o -nographic framework or g r i d w i t h i n which Tsimshian objects i n museums can be i d e n t i f i e d as c r e s t s . The crest system was analyzed as a s e r i e s of statements about Tsimshian s o c i a l s t r u c t u r e . There are seve r a l hundred d i s t i n c t named cr e s t s i n the Tsimshian system (these are l i s t e d i n an appendix), which i s consid-erably more elaborate than the crest systems of t h e i r neighbours, the Haida and T l i n g i t . This elaboration was p r i n c i p a l l y produced by the a p p l i c a t i o n of a s e r i e s of "operators" ( a t t r i b u t e s ) to crest animals i n order to produce new forms. Thus, the Haida had a s i n g l e raven c r e s t , while the Tsimshian had over a dozen (White Raven, S p l i t Raven, A l l Copper Raven, e t c . ) . S t i l l other forms were produced by merging features of d i f f e r e n t animals i n t o com-po s i t e "monsters." i i This complexity of forms is related in the thesis to a parallel elaboration and complexity in social structure, notably the greater elabor-ation of ranking and chieftainship in Tsimshian society. An analytical distinction was developed between "crests of differentiation" and "crests of integration." Crests of differentiation are totemic; that i s , they employ distinctions between natural species in order to express differences between human descent groups. Crests of integration are iconographically monsters, which blur the natural (species) distinctions upon which totemic systems are based, in order to express integrative tendencies in social organization at both clan and "tribal" levels. A sub-category of complex monster crests was defined and shown to be related to a cannibal theme in Tsimshian mythology. The cannibal was interpreted as a metaphor expressing the redistributive function of the chiefly role. Representations of complex monsters were found on totem poles, house front paintings, frontlets, and raven rattles (the face on its "stomach"). A number of these representations are illustrated. While the focus of the study is crest art, a non-crest iconogra-phic system based on spirit (naxn?7x) names was also defined and illustrated. This iconographic system is presented as the first ethnographically substan-tiated interpretation of Tsimshian masks. TABLE OF CONTENTS Abstract L i s t of Tables L i s t of Plates Acknowledgment Phonetic Key INTRODUCTION Chapter One: The Tsimshian Chapter Two: "Ethnology Under Glass" Chapter Three: S o c i a l Organization Chapter Four: The Crest Context Chapter Fi v e : The Crest System I: General Structure Chapter Six: The Crest System I I : Iconography of Museum Specimens EPILOGUE Plates 70-85 Bibliography APPENDIX I: Ranked Houses of the Tsimshian APPENDIX I I : Tsimshian Crests i i i Page i i v i v i x x i 1 7 19 51 113 138 171 249 253 270 279 327 i v LIST OF TABLES Page Table I. Tsimshian Clans and Their P r i n c i p a l Crest Animals. 61 Table I I . Tsimshian, Haida, and T l i n g i t Clans and Crests. 61 Table I I I . Schematic Representation of the Tsimshian Crest System; 150 LIST OF PLATES Page 1. Cedarbark headdress 82 2. Cedarbark headdress 83 3. Halibut hook 172 4. Halibut hook 173 5. a and b - Halibut Hook, two views 173 6. Halibut hook 174 7. Halibut hook 175 8. Halibut hook 175 9. Round r a t t l e 177 10. Human face mask 188 11. Human face mask 189 12. Human face mask 189 13. Human face mask 190 14. Human face mask 190-15. Human face mask 191 V Page 16. Human face mask with b i r d 191 17. Human face mask with b i r d 192 18. Animal mask . 192 19. Animal mask 193 20. Animal maskl 193 21. Animal mask 194 22. B i r d mask-headdress 195 23. B i r d mask-headdress 196 24. B i r d mask 197 25. B i r d mask 198 26. Bird-man mask 199 27. Eagle headdress 200 28. Eagle headdress 200 29. B i r d headdress 201 30. B i r d headdress 202 31. Animal headdress 203 32. Animal headdress 203 33. Frog headdress 204 34. Frog headdress 204 35. Painted hide armour 207 36. Ermine and abalone cap 212 37. C h i l k a t blanket 215 38. Chilkat blanket 215 39. Chilkat blanket 215 40. Chilk a t blanket 217 v i Page 41. F r o n t l e t 221 42. F r o n t l e t 221 43. F r o n t l e t headdress 222 44. F r o n t l e t 223 45. F r o n t l e t 224 46. F r o n t l e t 224 47. F r o n t l e t 225 48. F r o n t l e t headdress 226 49. F r o n t l e t 226 50. F r o n t l e t 227 51. F r o n t l e t headdress 227 52. F r o n t l e t 228 53. F r o n t l e t headdress 229 54. a and b - F r o n t l e t headdress, two views 230 55. F r o n t l e t headdress 231 56. F r o n t l e t 231 57. F r o n t l e t 232 58. Fr o n t l e t 232 59. F r o n t l e t 232 60. Fr o n t l e t 233 61. Raven Rattle ' 237 62. Raven Rattle 237 63. Raven Rattle 238 64. a and b - Raven R a t t l e , two views 238 65. a and b - B i r d r a t t l e , two views 239 v i i Page 66. Chilk a t dance apron 240 67. Painted s k i n legging 243 68. P a i r c l o t h leggings with appliqued designs 243 69. Painted screen 245 70. Representations of the gi^b9lk 254 a) F r o n t l e t b) mgnE 'sk's f r o n t l e t headdress c) Peter n i . s y j q wearing dance apron and f r o n t l e t headdress 71. Representations of the spmgi'k 255 a) Headdress b) Mask c) D e t a i l of lux j.n's totem pole, Kitwancool 72. Representations of the xske/ms9m 256 a) Mask b) Mask c) D e t a i l of wixE 's totem pole, Kitwancool 73. Representations of the imdz-aks 257 a) D e t a i l of wog.al?'s totem pole, Kitsegukla b) l E l t wearing f r o n t l e t headdress 74. Representations of the Thunderbird 258 a) F r o n t l e t headdress b) Carved figure 75. Representation of Hooked Nose; d e t a i l of wixE's pole, Kitwancool 259 76. Representations of the Decayed Corpse 260 a) F r o n t l e t v i i i Page b) D e t a i l of xsg.Jg.omlaxE /'s J. totem pole, Kitsegukla c) D e t a i l of haxp3gvo/tu's totem pole, Kitsegukla 77. Representation of the hagwglax; d e t a i l of ksg.? gomlaxE 7 -' s totem pole, Kitsegukla 261 78. House front paintings of the g* i { b alk 262 a) mjnE^sk's house at Gitlaxdamiks b) sg.agwe't's house at Fort Simpson 79. Painted housefront c o l l e c t e d by J.G. Swan i n 1875 263 80. Chief sgmadi'-k i n three costumes 264 a) Chilkat and mask-headdress b) Chilkat and eagle headdress c) and d) Crest robe and hat, two views 81. a) Chief Laganitz i n chi e f ' s costume 265 b) Albert Williams wearing F l y i n g Frog headdress c) Chief sqat'i.'n i n chief's costume 82. Kitwanga chief i n chi e f ' s costume 266 83. a and b - m?nE /sk's crest robe showing S p l i t Eagle and Person of G l a c i e r ; front and back views 267 84. Chief's chest 268 85. sqat'i'n's legging with appliqued g r i z z l y design 269 i x Acknowledgment This study i s based on materials c o l l e c t e d by Marius Barbeau and William Beynon, h i s Tsimshian associate. I have a great indebtedness to these two ethnographers, and to the q u a l i t y and quantity of the data they c o l l e c t e d . I hope that I have i n part discharged t h i s debt with the d i s s e r t a t i o n i t s e l f . Barbeau did not do these same data j u s t i c e i n hi s many pu b l i c a t i o n s , with the r e s u l t that h i s and Beynon's great service to the d i s c i p l i n e i n c o l l e c t i n g them has not been adequately recognized. Since t h i s i s the f i r s t major study to have been done by someone else using the Barbeau/Beynon f i e l d notes, i t may d i r e c t new attention to t h e i r singular and s i g n i f i c a n t c o n tribution. The Barbeau/Beynon notes were generously made a v a i l a b l e to me by Professor Wilson Duff, who i s also Chairman of my Advisory Committee. His continued generosity and patience i n helping me to understand the data, and the underlying order of which they are r e f l e c t i o n s , was the very best kind of teaching a g r a t e f u l student can receive. The other members of my committee, Professor Michael Ames, K. 0. L. Burridge, Harry Hawthorn, and Barrie Morrison, were c r i t i c a l and h e l p f u l readers of e a r l i e r versions of the d i s s e r t a t i o n . Professor Ames also c o n t r i -buted to i t s s t r u c t u r a l o r i e n t a t i o n through h i s rigorous course i n i n s t r u c t u r a l theory. Another of my teachers whose influence i s obvious i n these pages i s Professor David Aberle, who kind l y read and c r i t i c i z e d Chapter Three. X I photographed Tsimshian c o l l e c t i o n s and gathered a d d i t i o n a l information f o r t h i s study i n a number of museums, and wish to. acknowledge here the p r i v i l e g e s extended to me by the B.C. P r o v i n c i a l Museum, the Museum of Anthropology of the U n i v e r s i t y of B r i t i s h Columbia, the National Museum of Man, the Museum of the American Indian, the Royal Ontario Museum, the Thomas Burke Memorial Museum of the Uni v e r s i t y of Washington, and the F i e l d Museum. I also received h e l p f u l information by mail from the U.S. National Museum and the Museum and Laboratories of Ethnic Arts and Technology, UCLA. I am also g r a t e f u l to Lonnie Hindle, formerly of Kispiox, who held informal seminars on the Gitksan d i a l e c t i n his home i n Vancouver during 1971-72. F i n a n c i a l support was provided me by a Un i v e r s i t y Fellowship (1969-70) and Doctoral Fellowships from the Canada Council (1960-71, 1971-72). I was also permitted to take time from my duties as an ass i s t a n t curator i n the Museum of Anthropology during 1972-73 i n order to f i n i s h w r i t i n g the d i s s e r t a t i o n . X I Phonetic Key The orthography used i n t h i s study i s Marius Barbeau's f i e l d orthography, which he developed at the National Museum with the help of h i s colleague, Edward Sapir., I t has been s l i g h t l y s i m p l i f i e d . Vowels 1 as i n e as i n E as i n a as i n 9 as i n o as i n ? as i n u as i n seek" l a t e " pet" f a t h e r " again" 'mole" law" boot" Consonants p b t d k g as i n English q.g. f a r t h e r back than k and g k* g' (k^ g^) f a r t h e r forward than k and g, adding a y sound, as i n "thank you" s z as i n English 1 surd 1, l i k e the t h l i n " a t h l e t e , " whispered and s l u r r e d i n t o a s i n g l e sound with tongue i n 1 p o s i t i o n x as i n German " i c h " c t plus s, as i n cats the g l o t t a l stop, as i n "Hawai*!" 1. The English equivalents of Tsimshian sounds and the d e s c r i p t i o n of Barbeau's popular system are from Duff. (1964: 109-110). x i i Phonetic Key cont. Accent or stress (') a f t e r the vowel of the stressed s y l l a b l e . Length vowels marked with a dot are sounded f o r double the normal length of time. Barbeau's Popular Orthography In h i s publications Barbeau used a system based on the English alphabet. Correspondence between t h i s popular system and the preceding f i e l d system are as follows: Popular System F i e l d System ae E aw 3 r g-rh x h i 1 gy g' Examples: Gitrhahla g ' i t x a ' l a (Kitkatla) Legyarh legE'x (Legaic) Gitrhawn g"itxon (Gitkun) Rhaida xay'd? (Haida) There i s no point i n proposing a museum study of f o l k c l a s s i f i c a t i o n s , or of ethnoesthetics, or of attitudes about objects, or of the d e t a i l e d r e l a t i o n s between material culture and s o c i a l organization, f o r the data do not ex i s t i n museums. William C. Sturtevant, 1973 1 INTRODUCTION This i s a study i n museum anthropology, to which I have had a personal commitment since 1963, when I was h i r e d as the Smithsonian I n s t i t u t i o n ' s f i r s t docent (interpreter) i n anthropology. Since 1968, I have been p a r t i c i p a t i n g i n a modest renaissance i n object-oriented 1 research at the U n i v e r s i t y of B r i t i s h Columbia, stimulated by the example and teachings of Professor Wilson Duff. My choice of a problem i n museum anthropology f o r t h i s d i s s e r t a t i o n was a r e s u l t of these two influences on my personal and s c h o l a r l y development. In order to make any kind of concrete f a c t meaningful — whether i t be event or thing — i t i s necessary to i n t e r p r e t i t i n terms of a t h e o r e t i c a l construct or a n a l y t i c a l framework. This basic s t r i c t u r e of science has been overlooked by most anthropologists who write about objects i n museum c o l l e c t i o n s , with the r e s u l t that we have a great many catalogues which increase our f a m i l i a r i t y with p a r t i c u l a r pieces, but do l i t t l e to increase our understanding of them. In t h i s sense, we can say that the a r t of the Northwest Coast Indians i s one of the best-known and l e a s t understood aesthetic expressions of the non-Western world. The notable exception i n studies of t h i s art i s B i l l Holm's Northwest Coast A r t : An Analysis of Form (1965), which masterfully explicated some of the formal p r i n c i p l e s of the a r t s t y l e . The present study i s an attempt to explicate some of the p r i n c i p l e s underlying i t s iconography. 1. The term "object-oriented research" comes from Scott and Segmen (1970: 1005). 2 According to Panofsky (1962: 3), "iconography i s that branch of the h i s t o r y of art which concerns i t s e l f with the subject matter or meaning of works of a r t , as opposed to t h e i r form." I f we follow h i s well-known d i s t i n c t i o n s , a d e s c r i p t i o n of the "primary or natural mean-ings" of a work of art i s a "pre-iconographical" d e s c r i p t i o n ( i b i d . : 5). This would be the d e s c r i p t i o n , f o r example, of the subject matter of a painting as a woman holding a c h i l d . I t i s when we connect such repre-sentations with themes or concepts which have meaning i n terms of the culture of t h e i r o r i g i n , which he c a l l s turning motifs into images, that we reach the domain of "iconography i n the narrow sense" ( i b i d . : 6). This would be the i d e n t i f i c a t i o n of the woman and c h i l d as the V i r g i n Mary and Jesus. Iconography i n the narrow sense i s the p r i n c i p a l domain of the present study, although i t w i l l also attempt to fathom what Panofsky c a l l s "iconography i n a deeper sense": "those underlying p r i n c i p l e s which reveal the basic a t t i t u d e of a nation" ( i b i d . : 8). I am s p e c i f i c a l l y concerned with those images which had meaning to the Tsimshian as cr e s t s , although i n order to adequately understand crests I have had to look at c e r t a i n systems of non-crest representations. Fortunately, rather than working from the motif to i t s meaning, as have previous inv e s t i g a t o r s of Northwest Coast icono-graphy, I have had data at my disposal which permitted me to work from a system of meanings to motifs. This was poss i b l e because of the sys-tematic and c l a s s i f i c a t o r y nature of the Tsimshian crest system, but should also be poss i b l e f o r other bodies of totemic a r t . Tsimshian crests formed a c l a s s i f i c a t i o n system of the type defined by Levi-Strauss (1966) as totemic, that i s , an ethno-logical 3 construct used to c l a s s i f y s o c i a l groups. The terms i n a totemic c l a s -s i f i c a t i o n , as i n other taxonomic systems of n o n - l i t e r a t e peoples, are p r i m a r i l y drawn from nature, the d i s t i n c t i o n s between plant and animal species being used to represent differences between human groups. The d i s s e r t a t i o n i s based on l i s t s of several hundreds of crests owned by s p e c i f i e d houses or lineages of the Tsimshian, and represented iconographically on c e r t a i n items of material culture. These l i s t s were compiled from the unpublished f i e l d notes of Marius Barbeau and William Beynon and are contained i n Appendix I I . They constitute one of the most comprehensive l i s t s of "totemic" crests ever made a v a i l a b l e . A previous l i s t of Gitksan crests was published by Barbeau (1929: 158-169), although t h e i r systemic character appears to have gone unnoticed, even by Barbeau himself. The present l i s t s include also the crests of the Niska and Coast Tsimshian, and the Gitksan crests are reported more accurately and i n more d e t a i l than i n Barbeau's 1929 l i s t . The crest l i s t s were analyzed f o r r e g u l a r i t i e s or patterns suggestive of the rules underlying both t h e i r generation and t h e i r d i s t r i b u t i o n throughout the s o c i a l system. One of the most s i g n i f i c a n t and s u r p r i s i n g discoveries based on the crest l i s t s was a class of "monster" c r e s t s , defined as animal-like forms without natural prototypes, but composed of a t t r i b u t e s such as wings, f i n s , and beaks borrowed from nature. They thus b l u r the species d i s t i n c t i o n s upon which a totemic system i s based, and represent a new and evolving order of c r e s t s . Otherwise c a r e f u l scho-4 l a r s have co n s i s t e n t l y c l a s s i f i e d these monster forms as hawks, k i l l e r -whales, bears, or other animals on the basis of c e r t a i n perceived sim-i l a r i t i e s i n Tsimshian art to natural species on the pre-iconographic l e v e l , and thus they have with equal consistency misinterpreted the meaning of these c r e s t s . A sub-class of complex monster crests was defined and found to be r e l a t e d to a pervasive Cannibal theme i n Tsimshian thought. This, too, was unexpected and had not been anticipated from the l i t -erature, which deals only with a form of r i t u a l c a n n i b a l i s t i c dramati-zation believed to be recently borrowed from the Northern Kwakiutl. I i n t e r p r e t monster crests as a movement or transformation ^ i n the Tsimshian crest system from d i f f e r e n t i a t i o n to i n t e g r a t i o n , from d i v e r s i t y to unity. A s i m i l a r and p a r a l l e l process was found i n Tsimshian s o c i a l organization. The development toward the two u n i t i e s , the one metaphorical, the other s o c i e t a l , i s r e f l e c t e d most c l e a r l y i n the r o l e s and symbols of the c h i e f , e s p e c i a l l y among the Coast Tsimshian. This suggested, using the developmental stages of contemporary c u l t u r a l e v o l u t i o n i s t s , that the evolving s o c i a l system was i n t r a n s i t i o n from a t r i b a l l e v e l of i n t e g r a t i o n to a chiefdom l e v e l . Potlatches were the r i t u a l context i n which crests were v a l i -dated and displayed, with attendant d i s t r i b u t i o n of wealth to guest-witnesses. Potlatches permitted, i n e f f e c t , communal p a r t i c i p a t i o n i n and authorization of the construction of the metaphorical structure and ensured i t s f i t or congruity with the s o c i a l structure. The pot-l a t c h can thus be int e r p r e t e d , following Leach (1965), as a r i t u a l c e l e b r a t i o n of Tsimshian s o c i a l organization. 5 A f t e r i n v e s t i g a t i n g crests as a system of meanings, I used the system i n i d e n t i f y i n g and i n t e r p r e t i n g crest images i n Tsimshian a r t . These were found to be f a r l e s s ubiquitous than previously thought, and not to be synonomous with animal representations ( i . e . , an animal image i s not always a c r e s t ) . In the process of e s t a b l i s h i n g a r e s i d u a l category of non-crest images, I defined a previously undescribed sys-tem of mask iconography — naxnp 'x or s p i r i t names and t h e i r represen-t a t i o n i n masked dramas. I t i s the f i r s t ethnographically substantiated i n t e r p r e t a t i o n of Tsimshian masks i n the l i t e r a t u r e . Tsimshian crest art was a s p e c i a l use of a r t i s t i c images to make Tsimshian s o c i a l structure v i s i b l e . This suggests that a r t , at le a s t totemic a r t , i s more than aesthetic design; i t i s an i n t e l l e c t u a l or cognitive process by means of which man explores h i s world and renders i t meaningful and i n t e l l i g i b l e . As Levi-Strauss has argued, the so-c a l l e d p r i m i t i v e a r t i s t , and those who commission and use h i s a r t , are more cerebral than many have thought. There are seven chapters, with two appendices that contain supporting data. Chapter One i s an introduction to the culture and contains a taxonomy of l o c a l groups, intended to help the reader through a plethora of l o c a l and v i l l a g e names. Chapter Two describes the museum data and c o l l e c t i o n s used i n t h i s study. Chapter Three, " S o c i a l Organ-i z a t i o n , " i s a c r i t i c a l background chapter to the rest of the argument, which depends upon an a n a l y t i c a l d i s t i n c t i o n between two s t r u c t u r a l orders i n Tsimshian society. Chapters Four through Six describe the 6 crest system and present the major substantive findings of the d i s s e r -t a t i o n . The Epilogue i s a concluding statement. Appendix I contains a synoptic presentation of Tsimshian lineages by rank, t r i b e , and clan. Appendix II contains a c l a s s i f i c a t i o n of a ser i e s of crest l i s t s , the basic data upon which the study i s based. I have t r i e d to spare the reader from the necessity of lea r n -ing to recognize Tsimshian words. However, there are s i x which cannot be avoided. These include the names of the four clans: l a x s k ' i ' k , laxk' i b u / , g.anha^d^, and g'ispgwudwa'd-a. The l a s t two cannot be trans-l a t e d and i t would be misleading to use them while t r a n s l a t i n g the other two into E n g l i s h . The other two words are naxna /x, which, can be glossed as " s p i r i t , " and h a l a / i t , which i s a general word that can mean, according to i t s context, "shaman," or "dancer," as w e l l as r i t -uals such as secret society i n i t i a t i o n s i n which supernatural powers are involved. 7 CHAPTER ONE THE TSIMSHIAN The Tsimshian l i v e i n northwestern B r i t i s h Columbia, along the Nass and Skeena Rivers and on the coast and islands between t h e i r two estuaries and extending as f a r south as Milbanke Sound. With t h e i r neighbours, the Haida and T l i n g i t , they comprise a d i s t i n c t and d i s t i n c -t i v e c u l t u r a l grouping, known i n Northwest Coast studies as the Northern 1 Province. P r i o r to the modern period, which can be sa i d to have begun about 1880, the t r a d i t i o n a l cultures of the Northern Province shared a number of s t r u c t u r a l features which combined to give them the e s s e n t i a l u n i t y recognized i n t h i s anthropological c l a s s i f i c a t i o n . These included totemism, the p o t l a t c h , m a t r i l i n e a l descent, p r e f e r e n t i a l marriage, pre-ferred avunculocal residence, and ranking (see Rosman and Rubel, 1971: 2 7, 34). They also shared a great many c u l t u r a l t r a i t s , ranging from tobacco chewing to art and f o l k l o r i s t i c motifs, which r e i n f o r c e our 3 perceptions of t h e i r relatedness. Within the unity of the Northern Province, however, the d i s -t i n c t i v e n e s s of Tsimshian culture i s beginning to receive increasing a t t e n t i o n . In 1951, V i o l a G a r f i e l d summarized the then-current view of 1. The Northern Province was defined by Drucker (1955a: 187), who also included the northern Kwakiutl H a i s l a as marginal members. 2. Not a l l of these features, of course, were unique to the Northern Province; the combination was. 3. The Northern Province was, i n f a c t , defined by Drucker on the basis of t r a i t l i s t s (see 1955a: Chapter Nine). 8 the differences between the Tsimshian and t h e i r neighbours as follows: Three elements of Tsimshian culture set them o f f most d i s t i n c t l y from t h e i r neighbors. The language i s d i s t i n c t and, to date, no r e l a t i o n s h i p between i t and any others i n the area has been demonstrated. The Tsimshian have four exogamous kinship d i v i s i o n s i n contrast to the dual d i v i s i o n s of the T l i n g i t and Haida, though a l l four phratries are not represented i n every Tsimshian town. The Coast Tsimshian and Nisqa elevated c e r t a i n lineage heads to t r i b a l c h i e f s whose presti g e was greatly enhanced by t r i b a l economic sup-port and properties, and by t r i b u t e from a l l members of the l o c a l group regardless of clan a f f i l i a t i o n ( G a r f i e l d , 1966 [1951]: 4). In contrast, the present generation of anthropologists studying the Northwest Coast are concerned with discovering b a s i c s t r u c t u r a l and cognitive features of these cultures. The most s i g n i f i c a n t published r e s u l t of the a p p l i c a t i o n of s t r u c t u r a l anthropology to the Northwest Coast i s Abraham Rosman and Paula Rubel's Feasting with Mine Enemy (1971), i n which they investigate the r e l a t i o n s h i p s between the s t r u c -ture of the p o t l a t c h and aspects of s o c i a l structure f or s i x Northwest Coast s o c i e t i e s . They conclude that s i g n i f i c a n t s o c i a l s t r u c t u r a l d i f -ferences i n the Northern Province are that "the Tsimshian have matri-l a t e r a l cross-cousin marriage while the T l i n g i t and Haida have p a t r i -l a t e r a l cross-cousin marriage" ( i b i d . : 7) and "the Tsimshian exhibit a complex ranking of groups i n r e l a t i o n to one another while the Haida and T l i n g i t do not" ( i b i d . : 194). They then examine and explain p o t l a t c h v a r i a t i o n s i n the s o c i e t i e s i n terms of a model developed from these v a r i a b l e s . 9 This study i s another attempt to penetrate the patent oneness of the Northern Province and to discover s t r u c t u r a l v a r i a b l e s underlying differences i n r i t u a l and a r t . I t does not look at the T l i n g i t and Haida, except i n passing, but focusses rather on differences between the three d i v i s i o n s of the Tsimshian — the Coast Tsimshian, the Niska, and the Gitksan. While the reader can be r e f e r r e d to the published l i t e r a t u r e f o r general summaries of Tsimshian culture ( G a r f i e l d , 1939, 1966; Boas, 1916) , some of i t s basic patterns are reviewed here. The Tsimshian include both coastal and inland peoples, and oppositions of land and sea, coast and i n t e r i o r , and the animals and plants c h a r a c t e r i s t i c of each, provided a major symbolic theme running throughout t h e i r expressive c u l t u r e . This co a s t a l : i n t e r i o r opposition was mediated by the two great r i v e r s — the Nass and Skeena — which united them and provided t h e i r main food resource: salmon. Every family owned salmon f i s h i n g t e r r i t o r i e s along the r i v e r banks, where they spent the summer months catching the running salmon and smoking and drying them f o r winter. Another river-spawning f i s h whose p e r i o d i c i t y had great influence on Tsimshian l i f e was the eulachon, which ascended the Nass River i n prodigious numbers i n the early spring. This was s a i d to be a time of famine or near-famine f o r the Tsimshian, and they a l l moved eagerly to the mouth of the Nass to intercept the eulachon, whose a r r i v a l heralded the plenty of summer. Eulachon grease was a much p r i z e d commodity and condiment on the Northwest Coast, and the Tsimshian gained great wealth from trading i t to the T l i n g i t and 10 Haida, who owned no eulachon f i s h i n g grounds and who came to the Nass i n the spring to trade f o r i t . Nass River eulachon grease was famous a l l over the Northwest Coast f o r i t s s p e c i a l f l a v o u r . Winter brought the "time of taboos" and the people moved back into t h e i r great cedar plank houses i n permanent "totem pole" or winter v i l l a g e s to wait again f o r spring. This was the r i t u a l season, the time when the s p i r i t s came down from the mountains to l u r k i n the f o r -ests surrounding the v i l l a g e s . Certain houses were s a n c t i f i e d f o r the season by p l a c i n g sacred rings of red cedar bark on the doors; in s i d e the members of dancing s o c i e t i e s i n i t i a t e d new members and people pos-sessed by the Cannibal and Dog Eater s p i r i t s roamed the woods. I t was also a time f o r f e a s t i n g . People traveled from a l l over the Tsimshian country to attend the potlatches of the great c h i e f s , which l a s t e d f o r days, were staged with great pomp and ceremony, and were remembered f o r generations. Guests at potlatches were i n v i t e d to witness the r i t u a l t r a n s f e r of names and p r i v i l e g e s from t h e i r custo-dians i n one generation to the successors i n the next. Succession was m a t r i l i n e a l , and the most elaborate potlatches were those held to mark the assumption of a ch i e f ' s name — perhaps "name-title" i s more des-c r i p t i v e — by h i s successor, i d e a l l y h i s s i s t e r ' s son. These were the occasions at which the new chief erected a totem pole as a memorial to his predecessor. Before the p o t l a t c h , the Tsimshian often held s p e c i a l dramatic 4 performances in v o l v i n g another kind of name: s p i r i t (naxno'x) names. 4. Among the Gitksan e s p e c i a l l y , the c h i e f ' s name I t s e l f was often a naxio x name. 11 These were events of pure theatre, involving humour, suspense, and de-nouement; song and dance; masks, costumes, and props. The audience i t s e l f was often brought i n t o the a c t i o n and, as w e l l as we can deter-mine from the fragmentary record, everyone present seems to have thor-oughly enjoyed these occasions. The p l o t , as i t were, of the dramatic presentation was the ac t i n g out of the meaning of a name. I t was usu-a l l y done i n the manner of a charade, the audience being presented with various clues as to the meaning of the name being dramatized. A l l of these r i t u a l events — dancing society i n i t i a t i o n s , potlatches, and s p i r i t name dramatizations — required wealth to stage, and some of them required i n h e r i t e d p r i v i l e g e or r i g h t , so that they were normally a f f a i r s of the c h i e f l y f a m i l i e s . Except f o r the dancing society i n i t i a t i o n s , however, the e n t i r e community was often involved, at l e a s t i n the capacity of audience i f not minor p a r t i c i p a n t s . To be a high-ranking Tsimshian man was to be Tsimshian i n the f u l l e s t sense the culture afforded. I t was to be highborn and pure i n descent, to have assumed important names at potlatches of proper pomp, to speak w e l l and observe, most c a r e f u l l y , the rules of r i g h t behaviour and e t i q u e t t e , to have been w e l l - r a i s e d by one's parents and c a r e f u l l y taught by one's uncle. I t was to have r i t u a l l y encountered and c o n t r o l l e d dangerous supernatural powers, and to demonstrate t h i s control several times more at considerable expense. I t was to assume the noblesse  oblige of one's rank and, above a l l , i t was to accept the great respon-s i b i l i t i e s of one's name and the moral imperative to expend one's l i f e and one's wealth i n e f f o r t s to elevate i t . 12 As were t h e i r neighbours i n the Northern Province, the Tsim-shian were masters of the northern s t y l e of Northwest Coast a r t , the s t y l e characterized by Haida a r t i s t B i l l Reid as one of " c l a s s i c a l c o n t r o l " (Duff, et a l , 1967: n.p.). I t i s the meaning or iconography of t h i s a r t , i t s r i t u a l context, and i t s s o c i a l motivations that concern t h i s study. We w i l l also look at some outstanding examples of i t s form. * * * * * There i s confusion and inconsistency i n the l i t e r a t u r e regard-ing l i n g u i s t i c , c u l t u r a l , and geographic d i v i s i o n s of Tsimshian-speaking peoples. The following c l a s s i f i c a t i o n i s proposed and used herein to avoid ambiguity and c l e a r l y d i s t i n g u i s h l i n g u i s t i c (e.g., Tsimshian) and major (e.g., Coast Tsimshian) and minor (e.g., Lower Skeena Tsimshian) cultural-geographic units or d i v i s i o n s : 5 Tsimshian i s a l i n g u i s t i c designation f o r people speaking two r e l a t e d languages (RIgsby, 1969): 6 1. Coast Tsimshian 2. Nass-Gitksan. These two language groups are divided into three broad c u l -t u r a l and geographic d i v i s i o n s : 6 I. Coast Tsimshian, l i v i n g along the lower Skeena River up to and inc l u d i n g i t s canyon, and the coasts and islands from the mouth of the Nass south to Milbanke Sound; 5. Native speakers use "Tsimshian" to r e f e r to the Coast Tsimshian language only, and normally d i s t i n g u i s h between Niska and Gitksan d i a l e c t s . 6. Although the c l a s s i f i c a t i o n uses "Coast Tsimshian" to r e f e r to both a language and a c u l t u r a l group, the two are coterminous. 13 I I . Niska, l i v i n g along the Nass River and Portland Canal. I I I . Gitksan, l i v i n g along the upper Skeena River or i t s t r i b u -t a r i e s . The Coast Tsimshian and Niska are further subdivided: I. Coast Tsimshian A. Southern Tsimshian B. Lower Skeena Tsimshian C. Canyon Tsimshian I I . Niska A. Lower Nass B. Upper Nass. The Gitksan are often r e f e r r e d to i n the l i t e r a t u r e as the people of the Upper Skeena. T r i b e s : The Tsimshian l i v e d i n some twenty^six l o c a l groups usually r e f e r r e d to as t r i b e s . Each t r i b e customarily occupied a s i n g l e winter v i l l a g e , often of the same name. The t r i b e s are l i s t e d below according to the preceding geo-cultural d i v i s i o n s . The numbers of the t r i b e s (1-26) are sometimes used i n t h i s study and i n Appendix I I as a short-hand designation f o r the t r i b a l name. The s p e l l i n g of t r i b a l names i s based upon 'Marius Barbeau's f i e l d orthography and English t r a n s l a t i o n s are approximations derived from h i s f i e l d notes, except where another source i s given. The accepted names and s p e l l i n g s of present Tsimshian bands are given i n brackets (from Duff, 1964: 18-20). Minimal discus-sion of population movements i s included, p r i m a r i l y to a i d i n provenience designations of museum specimens. 14 I. Coast Tsimshian A. Southern Tsimshian The three southernmost t r i b e s had t e r r i t o r i e s on the outer coasts and i s l a n d s , although the t r a d i t i o n a l narratives of the g'id^stsu 7 and g'itg.a 7 i >atg t e l l of migrations from the Skeena River. The g'idgstsu' now l i v e i n the modern v i l l a g e of Klemtu with the HaiHais Kwakiutl. 1. g ' i d a s t s u / (meaning?) [Kitasoo] 2. g ' i t g . a ^ a t a ("people of the [ceremonial] cane") [Hartley Bay] 3. g ' i t x a ' i a ("people of the channel" [ G a r f i e l d , 1939: 176]) [ K i t k a t l a ] . B. Lower Skeena Tsimshian These ten t r i b e s had winter v i l l a g e s on the lower Skeena River u n t i l l a t e p r e h i s t o r i c times when they extended t h e i r t e r r i t o r i e s coastward, at the expense of the T l i n g i t , and b u i l t winter v i l l a g e s on the i s l a n d s of Metiakatla Pass, where the weather was milder. They continued to return to t h e i r t e r r i t o r i e s on the Skeena i n the summers. Af t e r the Hudson's Bay Company b u i l t Fort Simpson i n 1834, nine of the t r i b e s (the g'itwilksabE' became ext i n c t as a t r i b e ) moved t h e i r winter houses and r e b u i l t them on separate v i l l a g e sections near the Fort. These t r i b e s now comprise the Port Simpson and Metiakatla bands i n B r i t i s h Columbia, and there i s an offshoot population at New 7 Metiakatla on Annette Island In Alaska. 7. New Metiakatla was founded i n 1887 by the Anglican missionary William Duncan and some 800 C h r i s t i a n Tsimshian from the e a r l i e r C h r i s t i a n v i l l a g e of Metiakatla, which was b u i l t on the precontact s i t e of the winter v i l l a g e s of the ten Lower Skeena Tsimshian t r i b e s (see Duff, 1964: 92-94; Arctander, 1909). 15 4. g ' i t w i l g V t s ("people of the kelp" [ G a r f i e l d , 1939: 176]) 5 ' g ' l t z a x l E ^ ("people of the ," an e l d e r b e r r y - l i k e shrub, sp. unknown) g ' i t s K s ("people of the salmon trap") 7. g ,iriax''ang ,i'k ("people of the mosquitos") 8. g ,ina ,dJ. /iks ("people of the swift water") 9. g'It*and?.' ("people of the weirs") 10. g'ispaxl:/* ts ("people of the elderberries") 11. g'itwjlksabE' (meaning?) 12. g'ilodza'ua ("people of the way i n s i d e , " r e f e r r i n g to a canoe route through a slough i n the Skeena) 13. g'itlE.'n ("people of two passing canoes," r e f e r r i n g to the shape of two mountains). C. Canyon Tsimshian The g ' i t s g l a 'st l i v e d i n two winter v i l l a g e s on e i t h e r side of K i t s e l a s Canyon on the Skeena. These v i l l a g e s were abandoned between 1870 and 1890, the people moving to New K i t s e l a s and Port Essington. The g'itsamg.E/lam l i v e d below them near the mouth of the Kitsumkalum River. 14. g'itsamg.E^lam ("people of the plateau") [Kitsumkalum] 15. g ' i t s a l a ' s a ("people of the canyon") [Ki t s e l a s ] a. g'itlaxdzg'ks ("people at the foot of the s l i d e " or "peo-p l e at the edge of a precipice") b. g'itxtsE'x ("people at the edge of the la k e " ) . 16 II. Niska Generally speaking, the four Niska t r i b e s have not retained c l e a r and separate i d e n t i t i e s to the same degree as have t h e i r Coast Tsimshian and Gitksan neighbours. There was considerable population movement i n t h i s area, and the people who comprised the t r i b e s and l i v e d i n the d i f f e r e n t v i l l a g e s are not always c l e a r l y d i s t i n g u i s h a b l e on the basis of present data. A. Lower Nass The people of the lower Nass c a l l e d themselves the g' i t x a t ' i a i and were divided into two t r i b e s : the g'itxat^iTn (proper) and the g*itg*ig'E.'nix, who were a small offshoot population that moved a short distance upriver to the v i l l a g e of antegwalE y. The g'itxat* i fn did not conform to the usual pattern of l i v -ing i n a s i n g l e winter v i l l a g e , and they and the g'itg'ig'E.'nix exper-ienced a number of population s h i f t s so that t h e i r h i s t o r y of s e t t l e -ment i s complicated. Since the l a t e decades of the 19th century they have been l i v i n g at the two modern v i l l a g e s of K i n c o l i t h ("place of scalps") and G r e e n v i l l e or laxg.aldzap ("on deserted v i l l a g e s i t e " ) . The four v i l l a g e s l i s t e d below f o r the g' i t x a t 1 i.'n were 19th century "totem pole" v i l l a g e s . 16. g * i t x a t 7 i . n ("people using f i s h traps") [ K i n c o l i t h and Green-v i l l e ] a. g* i t ' i k s ("people of i k s ! " an exclamation) b. kwunwj'q ("where people sleep" when t r a v e l l i n g ) c. ang'edE' ("where they catch eulachon with rakes") d. g'itlax^a'us ("people on the sandbar") 17 17. g'itg'ig'E/nix ("people of up the r i v e r " ) [ K l n c o l i t h and Green-v i l l e ] a. antegwalE' ("place of happiness"). B. Upper Nass The people of the upper Nass were the g* i t g n w i l i k s ("people staying temporarily," r e f e r r i n g to t h e i r movement down the r i v e r at eulachon f i s h i n g time). They were divided into two t r i b e s : the g* itwgnksi' Ik, who l i v e d at the canyon, and the dominant g* itlaxda. xmks, who l i v e d a few miles above them. Around the turn of the century the o l d v i l l a g e of g'itwanksi ^ k burned down and the people s e t t l e d at Gwinaha. They have since moved to Canyon C i t y . The C h r i s t i a n s of g'itlaxdaCmks moved to the missionary v i l l a g e a i ^ y a ^ c ("early leaves") and, when i t was flooded ca. 1918, they returned to g*itlaxda f m k s , which was then given the name a i , y a / n c . Recently, the g* i t l a x d a IImks have moved to the modern v i l l a g e of New Aiyansh. 18. g'itwanksi^k ("people of the place of l i z a r d s " ) [Canyon City] 19. g'itlaxda^rnks ("people on the place of springs") [Gitlakadamix] III. Gitksan The seven Gitksan t r i b e s each occupied a s i n g l e winter v i l l a g e , s i x of them on or near the Skeena and one, k'itw^nlku^l, to the north on the "grease t r a i l " to the Nass. About 1880 another small t r i b e , the anlag.asgmdE'x, joined the k'isg.ag.a's. 18 Three C h r i s t i a n communities were founded between 1890 and 1910. Glen Vowell, the only one s t i l l i n existence, drew i t s converts from k'ispayaks, k'isg.ag.a's, and qaid a ?; Andimaul ("where they f i s h with hand l i n e s " ) mostly from k'itsgigu^kla; and Meanskinisht ("at the base of the big mountain") from k'itwang.E / and k'itw^ntku^l. g'it''anma /ks, at the s i t e of the white settlement of Skeena Forks or Hazelton (founded i n the l a t e 1860s), attracted people from neighbouring v i l l a g e s and i t s o r i g i n a l l y small population now exceeds a l l of the others; the k*isg.ag.a^s have now completely amalgamated with them. The people of qald •> have amalgamated with k* ispayaks, and many of the k' i tWj}niku / ? 1 moved to the Nass i n the l a t e 19th cen-tury to l i v e at Aiyansh and K i n c o l i t h . The t r i b e s i n t h e i r order upriver are: 20. k'itwpng.E' ("people of the place of rabbits") [Kitwanga] 21. k'itwaniku ' J l ("people of the l i t t l e place" or "people of the narrow v a l l e y " ) [Kitwancool] 22. k'itsagu'kla ("people of sggu'kla," a mountain) [Kltsegukla] 23. g'it*anma'ks ("people where they f i s h by torch l i g h t " ) [Hazelton] 24. k'ispayaks ("people of the hiding place") [Kispiox] 25. k'isg.ag.a's ("people of the sea-gulls"?) [Kisgegas] 26. qald:> ("wilderness") [Kuldo]. 19 CHAPTER TWO 1 "ETHNOLOGY UNDER GLASS" Mary Douglas, an outsider to Northwest Coast studies, reviewed 2 the published ethnographic l i t e r a t u r e on the Tsimshian and concluded that, " a l a s , very l i t t l e i s known about t h i s t r i b e , " and that one wish-ing to study them "has to make do with very poor ethnographic materials" (Douglas, 1967: 66). In the pages to follow, I s h a l l have occasion to evaluate the adequacy of such published ethnography as there i s ; here I am more concerned to e s t a b l i s h i t s s c a r c i t y . For Mary Douglas was quite r i g h t ; the student who wishes to understand the t r a d i t i o n a l Tsim-shian way of l i f e must work with spotty and incomplete ethnographic reportage. There are only three published monographic sources on the Tsimshian: one on Coast Tsimshian s o c i a l organization ( G a r f i e l d , 1939) , one on Coast Tsimshian myths (Boas, 1916), and one on Gitksan totem poles (Barbeau, 1929). There are also two c o l l e c t i o n s of Nass River texts (Boas, 1902, 1912), important early a r t i c l e s on s o c i a l organiza-t i o n (Sapir, 1915; Barbeau, 1917a) and kinship terms (Sapir, 1920) , a missionary's biography (Arctander, 1909), a lengthy book review (Barbeau, 1917b, on Boas, 1916), a short d e s c r i p t i o n of language (Boas, 1910), c o l l e c t i o n s of myths (Barbeau, 1953, 1961), some o r i g i n a l data on totem 1. The chapter t i t l e was taken from Harrison (1937). 2. In order to assess Levi-Strauss' (1967a) s t r u c t u r a l analysis of the Tsimshian myth of Asdiwal. 20 poles, dancing s o c i e t i e s , and shamanism i n larger works (Barbeau, 1950; Boas, 1897; Drucker, 1940; Barbeau, 1958), and a b r i e f h i s t o r y of Kitwancool (Duff, 1959). There are two popular summaries ( G a r f i e l d , 1966; Drucker, 1965, Chapter 7), and four recent doctoral theses: one on contemporary Coast Tsimshian reserve l i f e ( I n g l i s , 1970), one on Coast Tsimshian phonology (Dunn, 1970), and two on the contemporary 3 Gitksan ( J . and A. Adams, Harvard). There are a few more published a r t i c l e s , plus some unpublished papers i n c i r c u l a t i o n (notably, Rigsby, 1967 and 1969) , but the sources c i t e d here constitute the basic l i t e r a -ture on which ethnological understanding of Tsimshian culture and society must be based. Hence, the existence of Marius Barbeau's s o - c a l l e d "Tsim-shian F i l e " i n the D i v i s i o n of F o l k l o r e at the National Museum of Man assumes considerable importance. I t consists of the f i e l d records of Barbeau and William Beynon, h i s Tsimshian i n t e r p r e t e r and co l l a b o r a t o r , and an ethnographer i n h i s own r i g h t , and spans a period of 43 years, from Barbeau's a r r i v a l i n Port Simpson i n December, 1914, u n t i l Beynon's death i n 1957 (Barbeau died i n 1969). The range of data i n the F i l e has been described by Wilson Duff (1964b), who spent the pre-xerox year 1958-59 i n Ottawa, copying i t s contents and working on i t with Barbeau. B r i e f l y , the F i l e contains extensive l i s t s of names, c r e s t s , houses (lineages and branch l i n e a g e s ) , and t e r r i t o r i e s , plus several hundred house-owned narratives and variants of the t r a d i t i o n a l h i s t o r i e s of 3. I was denied access to J . Adams' thesis by the author, and have therefore w ritten t h i s study without the benefit of eit h e r t h e s i s . 21 the Tsimshian people. Scattered throughout are b r i e f ethnographic ob-servations, but there i s only one systematic observer's account i n the F i l e — a valuable d e s c r i p t i o n by Beynon of f i v e days of potlatches, dances, and totem pole r a i s i n g s at Kitsegukla i n 1945. Some 407 of the narratives and variants were compiled by Barbeau into four typed manu-s c r i p t s , one f o r each c l a n , which he hoped to have published. He t i t l e d them The Gwenhoot of Alaska i n Search of a Bounteous Land ( l a x s k ' i ^ k ) , Temlarh'am; The Land of Plenty on the North P a c i f i c Coast (g'ispgwudwa /dg), The Larhkibu Migrate South ( l a x k * i b u / ) , and The Kanhada Outlaws (g.anha /dg) . The r e s t of the data now i n Duff's copy of the F i l e are i n the form of h i s handwritten abstracts of Barbeau's and Beynon's notes. The time period the f i e l d notes r e f e r to cannot be f i r m l y es-tablished. Barbeau worked with Indian people between 1914 and 1929 who were i n t h e i r 60s and 70s, and who were reaching back i n memory to per-iods when t r a d i t i o n a l systems of Tsimshian culture were s t i l l f unction-ing. This would suggest the period ca. 1860 to 1880, although dated h i s t o r i c a l events which occurred both before and a f t e r that time are mentioned i n t r a d i t i o n a l contexts. As i s probably true i n most cases of memory ethnography, the people were attempting to describe what they perceived as the e s s e n t i a l l y timeless c u l t u r a l patterns of, or p e r s i s t i n g from, an immediately preceding t r a d i t i o n a l age. In the case of two of the primary classes of the f i e l d data used i n the present study — crests and names — i t does not matter whether the Indian people were r e f e r r i n g to the pre-1860 period or l a t e r . These were i n h e r i t e d names and e n t i t i e s which were, to the Tsimshian, t h e o r e t i c a l l y constant. 22 The data i n the F i l e form the basis of a long seri e s of pub-l i c a t i o n s by Barbeau, one matched and exceeded i n Northwest Coast eth-nography only by Franz Boas (Duff, 1964b; see the bibliography f o r a 4 p a r t i a l l i s t c f Barbeau's p u b l i c a t i o n s ) . The r e l a t i o n s h i p between the f i e l d data and the publications based on them make an indisputable case f o r the a r c h i v a l preservation of anthropological f i e l d notes. For the data were c o l l e c t e d and recorded c o n s i s t e n t l y and d i l i g e n t l y by men who knew and understood the people they were working with (and Bey-non was f l u e n t i n the two Tsimshian languages), whereas Barbeau's pub-l i c a t i o n s are contaminated by outmoded and ecce n t r i c t h e o r e t i c a l pre-occupations and poor, even sloppy, scholarship. Duff (1964b) has ex-amined and refuted or modified a number of Barbeau's most misleading conclusions, which need not be re-examined here, and anyone who has attempted to use h i s exhaustive survey of Northwest Coast totem poles 5 (Barbeau, 1950) has endured h i s awkward orthography and suffered from his inconsistency, inadequate documentation, and the generally poor organization of the book. Yet, the f i e l d records remain, preserved unaltered and un-excelled as primary documents on Tsimshian t r a d i t i o n a l h i s t o r y . In 1969, when I was attempting to formulate a research project to inves-t i g a t e Northwest Coast a r t , Wilson Duff made his copy of the Tsimshian 4. To date, no anthropological obituary and l i s t of publications has appeared for Barbeau and, to the best of my knowledge, none i s i n preparation. There i s a short popular biography of Barbeau i n Swayze, 1960. 5. I am r e f e r r i n g here to h i s published or popular orthography only; Barbeau's f i e l d orthography seems quite adequate. 23 F i l e a v a i l a b l e to me. The data i t contains provide information on Tsimshian iconography which was not, and could not now be, matched or exceeded f o r any other Northwest Coast society. While I was f i r s t f a m i l i a r i z i n g myself with the content of the F i l e , I re-read L e v i -Strauss' The Savage Mind (1966) and r e a l i z e d that the data on crests could be used to tes t h i s model of totemism, and, combined with the study of museum specimens, could extend i t by adding another dimension: the v i s u a l symbols (art) by which totemic messages (crests) were com-municated. I therefore formulated a research design to explore the r e -la t i o n s h i p s between Tsimshian cognitive systems, as these are r e f l e c t e d i n the Barbeau/Beynon notes, and t h e i r v i s u a l expressions, as these are preserved i n museums. These are r e l a t i o n s h i p s which can no longer be tested i n the f i e l d , and I hope that t h e i r e x p l i c a t i o n w i l l demonstrate the usefulness of museum documentation and c o l l e c t i o n s i n anthropological research. I do not, i n t h i s d i s s e r t a t i o n , investigate the en t i r e range of Tsimshian material c u l t u r e , but focus on crest a r t , or totemic a r t , which i s the v i s u a l expression of a complicated c l a s s i f i c a t i o n system and one, moreover, f o r which the f i e l d data are e s p e c i a l l y com-p l e t e . In the process of de f i n i n g crests and d i s t i n g u i s h i n g t h e i r representations from other a r t i s t i c representations, however, I deal to a considerable extent with c e r t a i n categories of non^crest a r t , notably masks. The f i r s t stage i n the research was to master the contents of the F i l e and transf e r what emerged as data s i g n i f i c a n t to the problem 24 at hand to 5 x 8" cards, so that i t would be r e t r i e v a b l e as needed. As I worked through the data, another category of information which had v i s u a l or a r t i s t i c expression became incr e a s i n g l y s i g n i f i c a n t : the masked dramatisations of s p i r i t (naxno 'x) names. These comprised a separate and uniquely Tsimshian r i t u a l expression previously unre-ported i n the l i t e r a t u r e (although noted by Duff, 1964b: 68-69). Not only d i d t h i s system warrant d e s c r i p t i o n , i t also provides an a n a l y t i -c a l l y u s e f u l contrast to the totemic system already being investigated. The contrast between these two systems — totemic and s p i r i t name — i n turn suggested a new way of conceptualizing Tsimshian s o c i a l organization. I began to see i t i n terms of two d i s t i n c t s t r u c t u r a l orders — which I c a l l the p o t l a t c h order and the h a l a * i t order — that employ d i s t i n c t kinds of material culture and.distinct forms of r i t u a l expression. This l e d , again, to a broadening of the scope of the study to include a long chapter (Chapter Three) on Tsimshian s o c i a l organization. Three categories of data, then, i n a d d i t i o n to ethnographic observations, were extracted from the Barbeau Tsimshian F i l e to form the basis of the present study. They are 1) l i s t s and descriptions of c r e s t s , 2) ranked l i s t s of houses or lineages, and 3) l i s t s of s p i r i t names, plus some data as to how these were dramatized. Each of these categories of data w i l l be discussed separately. The master l i s t of crests (Appendix II) was compiled from hundreds of p a r t i a l l i s t s and references i n crest myths (ada/ox) r e -corded by Barbeau and Beynon from dozens of Tsimshian people between 25 1914 and Beynon's death i n 1957. While the l i s t s recorded from d i f -ferent people were seldom i n complete agreement, they overlapped and complemented each other i n numerous ways so as to make possible a master l i s t which i s more complete and more balanced than could have been known or remembered by any s i n g l e person. In order to appreciate the necessity of working from a composite l i s t , and to better appreciate the completeness of the Barbeau/Beynon data, i t i s h e l p f u l to r e f e r to the much shorter Tsimshian crest l i s t recorded by Henry Tate and published by Boas (1916: 503-506). Tate said at the time that "there were only a few old women who remember the crests" ( i b i d . : 503), which was demon-strated to be untrue by Barbeau's l a t e r work i n the same area, and Boas pointed out discrepancies i n the published l i s t with "a l i s t w ritten about seven years ago," apparently also by Tate. Not only does the second l i s t omit some of the important animal crests of the Coast Tsimshian (e.g., mountain goat, shark, mosquito), the crests l i s t e d "are, i n many cases, inaccurate, and never i n d i c a t e t h e i r owner" (Barbeau, 1917b: 561). Sapir's (1915) published l i s t of Niska crests i s also i n -accurate and incomplete (Wilson Duff was given a copy of t h i s a r t i c l e with extensive c o r r e c t i o n a l marginal notes by Barbeau). I t i s , how-ever, a f a r more useful l i s t than Tate's. Discrepancies i n the various recorded l i s t s of crests are due to more than f a u l t y memory and f a u l t y recording. Crest ownership . was enormously important to the Tsimshian. Crests were a measure of prestige to a people who competed f i e r c e l y f or p r e s t i g e , and i t i s to 26 be expected that someone would l i s t the crests owned by h i s or her own house i n considerably more d e t a i l than he would l i s t the crests of others. Each person, then, reported on t h i s highly complex system from the perspective of h i s own p a r t i c i p a t i o n i n i t , and some used the occasion of reporting to enhance t h e i r own po s i t i o n s . Another source of discrepancy i n the l i s t s was the tendency of people to over-look the more general or widely owned crests i n favour of the more highly s p e c i a l i z e d and prestigious crests associated with high rank. Those which were commonly owned tended to be taken f o r granted. Undoubtedly, some crests were simply forgotten, e s p e c i a l l y by the Coast Tsimshian l i v i n g around Port Simpson. The population of 6 these t r i b e s dropped from an estimated 3,000 i n 1835 to 632 i n 1924 (G a r f i e l d , 1939: 332-333). The crest system as i t survived i n the memory of th i s remnant population must have been mutilated. Fortu-nately, disease, trader, and missionary a c t i v i t y were les s a c t i v e i n the other, more remote Tsimshian settlements. The Coast Tsimshian also retained fewer of the tangible expressions of t h e i r crests into the twentieth century than did the Niska and Gitksan, both of whom used totem poles as memory aids when working on crest l i s t s with Barbeau. In 1862, James Deans (1899: 7) v i s i t e d Fort Simpson and "was astonished at the amount of carvings and paintings on the houses and t a l l columns, to be seen everywhere." But by 1878, George Dawson (1880: 115B) reported that "among the 6. This drop would include the 823 people who moved to New Metlakatla with Duncan ( G a r f i e l d , 1939: 333). 27 Tshimsians at Port Simpson, most of the o r i g i n a l carved posts have been cut down as missionary influence spread among the people." In con-t r a s t , the totem poles at Gitlakdamiks on the upper Nass were not des-troyed out of C h r i s t i a n zeal u n t i l 1917 or 1918 (Barbeau, 1929: 1, footnote) and those of the Gitksan never were. S i m i l a r l y , i f the records of such industrious c o l l e c t o r s as the Newcombes, Emmons, and Barbeau himself are accurate r e f l e c t i o n s of what there was s t i l l to be c o l l e c t e d , the people of Port Simpson had very l i t t l e movable material culture l e f t to s e l l (or, by inference, to use as memory aids) by the turn of the century. Crest data extracted from the Barbeau f i l e have been c l a s s i -f i e d and are included herein as Appendix I I . These data include the names of the crests i n Coast Tsimshian and Nass-Gitksan, English trans-l a t i o n s , such descriptions of the material representations of the crests as were reported, rules of use, and the houses (lineages) claim-ing each crest. Certain a d d i t i o n a l data, such as important crest con-f l i c t s and notes on the transf e r of crests from one house to another by capture or i n compensation for murder, are included as footnotes. The crests are l i s t e d by clans — laxk* i b u / , g* isp^wudwa 7dg, g.anha'da, laxak*i'k — and by d i v i s i o n s — Coast Tsimshian, Niska, Gitksan — within each clan. They are c l a s s i f i e d into the following categories, which are discussed i n Chapter Fiv e : primary animal, secondary animal, human, monster, plant, natural phenomenon, and a r t i f a c t . Ranked l i s t s of Tsimshian houses arranged by clan and t r i b e are included i n Appendix I. These l i s t s were synthesized by myself and 28 Wilson Duff from a number of l i s t s Barbeau recorded from Tsimshian people (Duff prepared the Coast Tsimshian l i s t s , which have been mini-mally expanded by Halpin, and Halpin prepared the others from data which had been p a r t i a l l y ordered by Duff). These l i s t s must not be taken as i n f a l l i b l e or "true" i n any absolute sense. They do represent a consensus of the data provided by Barbeau's Tsimshian teachers and, as such, probably correspond c l o s e l y to ranking arrangements as they existed i n the minds of the Tsimshian generally — at the time Barbeau recorded them (between 1914 and 1929). There i s no reason to question the clan assignments, since these did not change (except under extreme conditions, such as incestuous mar-riages) , and the Coast Tsimshian d i v i s i o n of houses into two status 7 l e v e l s was probably r e l a t i v e l y s t a t i c and unchanging. But the rankings, expressed i n the l i s t s by Roman numerals, were i n constant f l u x , and could have, t h e o r e t i c a l l y , changed each time a potlatch occurred (see Rosman and Rubel, 1971, Chs. II and V I I I ) , although such changes, i f they had been recorded, would probably have shown a very slow and v a r i -able r i s e and f a l l of houses i n rank. I t was a system not meant to be written down and frozen i n time. The implications of rank order among the Tsimshian, and the differences between the three d i v i s i o n s , are further discussed i n Chapter Three. 7. I am using the term "status l e v e l s " i n t h i s study i n place of the more t r a d i t i o n a l term " c l a s s " ; the difference i s discussed i n Chap-ter Three. 29 The p a r t i c u l a r value of the l i s t s here i s that they permit the arrangement of crests according to the houses which owned them i n order to give a v i s u a l p i c ture of Tsimshian s o c i a l organization. The arrangements reveal s u f f i c i e n t patterning to suggest some rules f o r the generation of crests as material expressions of s o c i a l p o s i t i o n . The l i s t s also provide a guide to clan and v i l l a g e a f f i l i a t i o n s of names and houses, which can be very confusing when encountered i n the l i t e r a t u r e . The Coast Tsimshian and Gitksan l i s t s of houses are probably very close to complete, and the i n t e r n a l evidence of the l i s t s provided by d i f f e r e n t people suggests that the s o c i a l organizations of these two d i v i s i o n s were f a i r l y s table. The s i t u a t i o n on the Nass was another story, and there was s i g n i f i c a n t disagreement among those who gave l i s t s to Barbeau. Part of t h i s was undoubtedly due to the movements of population from v i l l a g e to v i l l a g e during the h i s t o r i c period (see Chapter One), but i t i s also l i k e l y that the s o c i a l organization of the Niska had never s t a b i l i z e d to the same extent as i n the other two d i v i -sions due to the i r constant warring, marrying, and trading with Haida, T l i n g i t , and other Tsimshian who came to the Nass River f i s h i n g grounds during eulachon season. I t should also be pointed out that Barbeau spent less time on the Nass than he did among the Gitksan and Coast Tsimshian, and that there i s very l i t t l e published or unpublished data from the Nass which can be used to amplify h i s notes. I have, there-fore, not attempted to reco n c i l e the d i f f e r e n t versions of rank order f o r the Niska, but have included i n Appendix II each s i g n i f i c a n t version recorded. 30 Barbeau and Beynon recorded 662 s p i r i t (naxn? 7x) names from s p e c i f i e d houses of the three d i v i s i o n s of the Tsimshian (when the same name was claimed by more than one house, I counted i t separately each time i t was recorded). Each name was recorded i n Coast Tsimshian or Nass-Gitksan, with English t r a n s l a t i o n s , and f o r w e l l over h a l f of the names he also recorded b r i e f , usually one-line, descriptions of the manner i n which the name was dramatized. While these descriptions are s i g n i f i c a n t , they are too truncated to permit understanding of s p i r i t name enactments as a dramatic form. Fortunately, however, the Baynon account of the potlatch se r i e s at Kitsegukla i n 1945 contains h i s eye-witness record of some two dozen naxnj'x performances. These permit construction of a model f or the dramatic form, by means of which the other 662 naxn^ /x names recorded by Barbeau can be interpreted. While the naxnp'x system must s t i l l await a f u l l d e s c r i p t i o n and analy-s i s , I have described i t s major features i n Chapter Four and i l l u s t r a -ted some of the masks, with t h e i r attendant documentation, i n Chapter Six. In the early spring of 1971, having worked through the Barbeau/Beynon notes, I turned to museums to search out Tsimshian a r t i -f a c t s which would add a palpable three-dimensional r e a l i t y to the data on Tsimshian s o c i a l organization, cres t s , and s p i r i t names. From clues i n the l i t e r a t u r e , notably John E. Hunter's Inventory of Ethnological  C o l l e c t i o n s i n Museums of the United States and Canada (1967), I chose to v i s i t the following museums which were l i k e l y to have s i g n i f i c a n t 31 8 Tsimshian c o l l e c t i o n s (hereinafter abbreviated as i n d i c a t e d ) : B r i t i s h Columbia P r o v i n c i a l Museum, V i c t o r i a (PM) Unive r s i t y of B r i t i s h Columbia Museum of Anthropology, Vancouver (UBC) National Museum of Man, National Museums of Canada, Ottawa (NMC) Royal Ontario Museum, Toronto (ROM) Museum of the American Indian, New York (MAI) F i e l d Museum, Chicago (FM). Other p o t e n t i a l l y s i g n i f i c a n t Tsimshian c o l l e c t i o n s are i n the American Museum of Natural History (Emmons, Boas), the Univ e r s i t y Museum of Ph i l a d e l p h i a (Louis Shotridge), the Museum of Ethnic Arts and Technology, UCLA (items c o l l e c t e d by Beynon f o r S i r Henry Wellcome), and the National Museum of Natural H i s t o r y , Smithsonian I n s t i t u t i o n (Swan). , Museum research has i t s own kinds of t r i b u l a t i o n s (see Sturtevant, 1973, for an expert's i n t r o d u c t i o n ) . In the f i r s t place, the objects preserved i n museum c o l l e c t i o n s are <->but a sampling, c o l -l e c t e d by others and normally according to aesthetic rather than s c i e n -t i f i c c r i t e r i a , of the t o t a l universe of objects they represent. Again, 8. I also v i s i t e d the Thomas Burke Memorial Museum, University of Wash-ington, S e a t t l e , where there are some dozen, poorly documented Tsim-shian pieces c o l l e c t e d by Emmons and Walter C. Waters. They are not included i n th i s study. I had also intended to v i s i t the American Museum of Natural H i s t o r y , New York, but i n s p i t e of two l e t t e r s from me and one from my advi-sor (none of which were answered), they had made no provisions f o r me to view objects i n t h e i r c o l l e c t i o n . Rather than press my case there, I used the time I had l e f t i n New York f o r further work at the MAI where Vincent Wilcox, Curator of Research, was e s p e c i a l l y h e l p f u l . 32 of the a r t i f a c t s one believes to be i n a c o l l e c t i o n , because they are l i s t e d i n the museum's catalogue, there w i l l be a s i g n i f i c a n t proportion unavailable to the v i s i t i n g researcher because 1) they are i n s p e c i a l storage, 2) they are on exhibit and cannot or w i l l not be removed, and 9 3) no one can f i n d them. Then, as i n most kinds of anthropological i n v e s t i g a t i o n , the researcher i s an outsider to a closed system, one i n which people per-form customary duties within a seemingly (at f i r s t ) i n e x p l i c a b l e con-text of values and s o c i a l h i e r a r c h i e s , and one i n which there are no 10 i n s t i t u t i o n a l i z e d roles or f a c i l i t i e s f o r the stranger or outsider. Inexcusably, to me, one's an t i c i p a t e d colleague and counterpart, the resident museum ethnologist, a l l too often does not share the v i s i t i n g researcher's enthusiasm for and i n t e r e s t i n the c o l l e c t i o n s under h i s or her care, and the v i s i t o r must seek out a photographer, r e g i s t r a r , or technician f o r sympathetic and knowledgeable assistance. 9. Fortunately, I did not have any experiences quite as f r u s t r a t i n g as the one reported by Sturtevant (1969: 635) when 83 percent of the specimens he selected from the catalogue of an unnamed European museum could not be found. 10. In s p i t e of claims to the contrary; Sturtevant (1973: 14) believes that t h i s i s because research v i s i t o r s are so infrequent and notes that museum " i d e a l s support research." I would q u a l i f y t h i s and say that museum i d e a l s , as revealed by museum s t a f f behaviour, sup-port research by in-house researchers. I am basing t h i s observation not only on my experiences as a graduate student doing museum research f o r t h i s study ( i n which case my low status can be expected to have accorded me d i f f e r e n t receptions from those accorded Dr. Sturtevant), but also upon some seven years' experience on the s t a f f s of a na-t i o n a l museum and a u n i v e r s i t y museum. However, I am sympathetic, as i s Sturtevant ( i b i d . : 14), with the chronic museum problems of "low s t a f f i n g , lack of space, l o s t specimens, m i s l a i d records, and the demands of ex h i b i t programs" which are normally offered i n apology. 33 The rewards of museum research are simple enough. F i r s t , f i n d i n g a r t i f a c t s of the kinds one has predicted or hoped would be there, and, second, f i n d i n g , a l l too r a r e l y , that the f i e l d c o l l e c t o r had c o l l e c t e d supporting documentation pertinent to one's problem. Of the c o l l e c t o r s responsible f o r the Tsimshian pieces upon which t h i s study i s based, only one — and i t was Marius Barbeau himself — c o l -l e c t e d supporting documentation adequate to my purposes. The others, and they include some of the best-known and respected c o l l e c t o r s who have worked on the Northwest Coast, were p r i m a r i l y c o l l e c t o r s of ob-je c t s only, and to judge from the records they made, had l i t t l e i n -te r e s t i n the people and cultures these objects represented. This i s , i n i t s e l f , an h i s t o r i c a l problem worth i n v e s t i g a t i o n : "the h i s t o r y of c o l l e c t i n g and of museums i s part of Euroamerican c u l t u r a l and i n t e l l e c -t u a l h i s t o r y that ought to be examined as evidence on the changing i n t e r e s t s i i n exotic peoples and ideas about them" (Sturtevant, 1973: 13). The only published examination of c o l l e c t o r s ' a t t i t u d e s toward 11 donor populations on the Northwest Coast i s Macnair (1971), although current opinions about c o l l e c t i n g p ractices of the past are shared f r e e l y i n pri v a t e conversations. Someefifteen c o l l e c t o r s were responsible for most of the Tsim-12 shian pieces i n the museums I v i s i t e d . They f a l l i nto two groups, 11. Macnair's references to "Mr. C o l l e c t o r " i n t h i s a r t i c l e are t h i n l y v e i l e d references to C F . Newcombe. 12. This includes p r i n c i p a l c o l l e c t o r s only; pieces acquired by auc-t i o n , from dealers, or s i n g l e and i n small l o t s (e.g., there i s one item at the NMC [VII-C-19] c o l l e c t e d by Franz Boas) are not included. 34 separated on the basis of t h e i r c o l l e c t i n g purposes, which have s i g -n i f i c a n t implications f o r the q u a l i t y of supporting documentation associated with t h e i r c o l l e c t i o n s . The f i r s t group, who might be c a l l e d f i e l d c o l l e c t o r s , are those persons, resident outside of Tsimshian t e r r i t o r y , who made t r i p s to that area f o r the express purpose of c o l l e c t i n g a r t i f a c t s from the Tsimshian, or, i f resident i n the area, were commissioned by persons from outside to make such c o l l e c t i o n s . With two exceptions, which are i n themselves i n s t r u c t i v e , the q u a l i t y of supporting documentation f o r the c o l l e c t i o n s made by t h i s group i s superior to the documentation of the other group. The second group of c o l l e c t o r s , who might be c a l l e d resident  c o l l e c t o r s , are those non-Tsimshian who were resident i n the Tsimshian area f o r purposes other than c o l l e c t i n g , and whose c o l l e c t i o n s were therefore made i n c i d e n t a l to these other and p r i o r purposes. These are e s s e n t i a l l y " c u r i o " c o l l e c t i o n s : the objects being acquired more as mementos of the period of residence than i n and f o r themselves. I t i s therefore understandable that the c o l l e c t o r usually had l i t t l e or no in t e r e s t i n documentation. Regrettably, these tend to be e a r l i e r c o l -l e c t i o n s than those more purposefully made. Paradoxically, the resident c o l l e c t o r s were i n a better p o s i -t i o n to acquire the more complete documentation, due to prolonged con-tact with the donor population, had he but seen t h i s as a desirable end. The two exceptions to t h i s c l a s s i f i c a t i o n are the I.W. Powell and C.V. Smith c o l l e c t i o n s . Powell's primary purpose f o r v i s i t i n g 35 Tsimshian t e r r i t o r y was h i s r o l e as Superintendent of Indian A f f a i r s , although he c o l l e c t e d a r t i f a c t s from them i n 1879 on commission from the Geological Survey i n Ottawa (the forerunner of the National Museums). Hence, i n terms of t h i s c l a s s i f i c a t i o n , h i s dual purpose might be sa i d to have d i l u t e d or worked at cross-purposes with the f i e l d c o l l e c t o r ' s purpose and, therefore, to account at le a s t i n part for h i s poor documen-ta t i o n . G.V. Smith was a Hazelton resident who amassed a c o l l e c t i o n which he sold to the National Museum i n 1925. As a resident c o l l e c t o r , he predictably provided no documentation. However, Barbeau, a f i e l d c o l l e c t o r , was able to secure good documentation f o r the c o l l e c t i o n from other informed persons. Hence, a resident c o l l e c t o r ' s c o l l e c t i o n has the l e v e l of documentation normally provided by a f i e l d c o l l e c t o r . These p r i n c i p a l c o l l e c t o r s and t h e i r c o l l e c t i o n s are l i s t e d below, together with dates, pertinent ranges of museum catalogue numbers (not n e c e s s a r i l y continuously Tsimshian; other pieces are often i n t e r -spersed) , proveniences, some i n d i c a t i o n of the range of items i n the c o l l e c t i o n , and notes as to the q u a l i t y of the supporting documentation. F i e l d C o l l e c t o r s : CM. Barbeau, anthropologist and f o l k l o r i s t , National Museum of Canada 1915 (Jan. - Mar.). Coast Tsimshian. NMC. VII-C-491 - VII-C-728. Well-balanced ethnographic c o l -l e c t i o n . Documentation includes t r i b e , owner, native name f o r specimens, notes on use and iconography. 36 1920-21 (Aug. - Jan.). Gitksan. NMC. VTI-C-739 - VII-C-1047. Well-balanced ethnographic c o l -l e c t i o n . Documentation more de t a i l e d than f o r 1915 c o l l e c t i o n . Includes t r i b e , owner, maker, native name, notes on use and iconography. 13 1923 (summer). Gitksan. NMC. VII-C-1048 - VII-C-1099. Large c o l l e c t i o n of masks, other items. Documentation s i m i l a r to 1920-21 c o l l e c t i o n . Includes t r i b e , owner, maker, native name, notes on use and iconography. 1924. Gitksan and Niska. NMC. VII-C-1103 - VII-C-1182. Ethnographic c o l l e c t i o n . Documentation includes t r i b e , owner, maker, native name, notes on use and iconography. ROMti HN-613 - HN-654. Coast Tsimshian, K i t l o p e , Nass River (some pieces c o l l e c t e d by Beynon; some purchased from Pat P h i l l i p s o n , Prince Rupert dealer). Includes well-documented serie s of chief's paraphernalia from sqat^i'n (laxk*ibu) and mynEsk (laxsk* i'k) of g* itlaxda?mks. Documentation poor f o r Beynon and P h i l l i p s o n pieces. TT. Duff (1964b: 65) does not l i s t a f i e l d t r i p f o r Barbeau i n 1923; however, Barbeau's typed f i e l d notes f o r t h i s c o l l e c t i o n are under the heading " L i s t of specimens c o l l e c t e d i n summer of 1923 by CM. Barbeau i n the Upper Skeena Country, B.C. (mostly Gitksan specimens)." 37 1927 (summer). Kispayaks, Kisgegas, and Gitlaxdamiks. ROM. HN-680 - HN-826. Ethnographic c o l l e c t i o n . Documenta-ti o n includes t r i b e , native name, owner, and excellent notes on use and iconography. (July-Sept.). Gitksan and Niska. NMC. VTI-C-1364 - VII-C-1426. Ethnographic c o l l e c t i o n , i n -cludes some items purchased from Pat P h i l l i p s o n , Prince Ru-pert dealer. Documentation includes t r i b e , owner, native name, and notes on use and iconography. 1929 (summer). Gitksan and Niska. NMC. VII-C-1441 - VII-C-1476. Ethnographic c o l l e c t i o n . Doc-umentation s i m i l a r to above. (June). Gitksan. ROM. HN-1214 - HN-1263. C o l l e c t i o n of shamans' paraphernalia from Kispiox made for Dr. Harold M. T o v e l l , who presented i t to the Academy of Medicine, Toronto, i n 1930. Now i n the ROM. . Coast Tsimshian, Nass River. ROM. HN-867-HN-926. P r i m a r i l y headdresses and masks. Docu-mentation not as complete as usual. George A. Dorsey, anthropologist. 1897. Gitksan. FM. Accession No. 537 (accessioned October 30, 1897). Cata-logue Numbers 53051-53142. This c o l l e c t i o n resulted from a 38 f i e l d t r i p to Hazelton i n August, 1897, and includes some items purchased from Cunningham & Son, Port Essington. Docu-mentation i s disappointing; provenience given as "Port Essing-ton" or "Hazelton" only, although the c o l l e c t o r did t r y to get the names of the masks (e.g., "Peace Maker" or "represent-ing Sicanee Indians"). G.T. Emmons, r e t i r e d U.S. naval o f f i c e r , p r o f e s s i o n a l c o l l e c t o r . MAI. Specimens from a l l three d i v i s i o n s of the Tsimshian, although p r i m a r i l y Niska and Gitksan. Between 1905 and 1936, the MAI accessioned between 200 and 300 Tsimshian pieces (that I was able to locate) received from Emmons. This museum does not maintain an accession record or other catalogue r e t r i e v a l system to f a c i l i t a t e searching i t s holdings. I t i s necessary to go one by one through some one m i l l i o n catalogue cards or to work d i r e c t l y from the specimens i n storage. Hence, i t i s l i k e l y that some Tsimshian c o l l e c t i o n s were missed. Catalogue numbers f o r Emmons' c o l l e c t i o n s which include Tsimshian pieces are l i s t e d below, together with the approximate dates they were received i n the museum (accession dates are not entered on the catalogue cards; these dates were taken from a cata-logue number/date key maintained i n the Research Annex): 39 1905 4316 1919 9/6730-9/6742 1907 1/4165-1/4368 1920 9/7843-9/8152 1909 2/4318-2/4335 9/8010 1910-12 2/6934-2/6998 1921 10/4576-10/4585 1914 3/5008-3/5037 1922 11/1741-11/1825 1915 4/508-4/534 11/3843-11/3929 1916 5/5017-5/5071 11/5400-11/5464 5/5424-5/5425 1924 12/6628 5/6896^5/6906 13/3955-13/4049 5/8805-5/8823 1926 15/1321-15/1367 1917 6/524-6/549 1928 15/8947-15/8998 6/6279-6/6326 1936 19/765-19/812 1918 7/4411-7/4437 8/2606-8/2609 8/7996-8/8016 8/8466-8/8470 The above c o l l e c t i o n s include a wide range of a r t i f a c t types, inc l u d i n g an e s p e c i a l l y good s e l e c t i o n of masks and four com-ple t e Gitksan shamans' bundles. The documentation i s v a r i a b l e , ranging from provenience given as "Tsimshian," to v i l l a g e a t -t r i b u t i o n s , to occasional pieces with the name of the owner and notes on use and iconography. I was able to locate Emmons' o r i g i n a l catalogue notes only f o r the 1914 (3/5008-3/5037), 1915 (4/508-4/534), 1918 (8/7996-8/8016), and 1920 (9/7843-9/8152) c o l l e c t i o n s . There i s also a catalogue dated 1932, 40 "Lt. G.T. Emmons Northwest Coast C o l l e c t i o n (1-193)" but MAI catalogue numbers are not included, so that c o r r e l a t i o n of the notes with s p e c i f i c pieces i s d i f f i c u l t . S i r A l f r e d Bossom, 1881-1965, English a r c h i t e c t . NMC. VII-C-1710 - VII-C-1797. Coast Tsimshian, Niska, Kispiox. Accessioned 1955, 1960. This c o l l e c t i o n was quite c e r t a i n l y made for Bossom by Emmons, probably between 1903 and 1916. Bossom wrote as follows i n "Some personal notes on the cre-ating of t h i s c o l l e c t i o n of the works of the North West Coast Indians" (n.d., ms. i n Ethnology D i v i s i o n , NMC): "to help me gathering t h i s large assortment of the Arts and Crafts of these people f o r about f i f t y years I was most fortunate i n e n l i s t i n g the a i d of an old and highly respected American f r i e n d , Captain Emmons, who knew these Indians w e l l . " Bar-beau wrote i n a memorandum regarding the c o l l e c t i o n (n.d., Ethnology D i v i s i o n , NMC): "the c o l l e c t i o n was made, I be-l i e v e , from 1900 to 1910. How d i d S i r A l f r e d Bossom acquire t h i s c o l l e c t i o n ? Solely through L i e u t . G.T. Emmons of the U.S. Navy i n Alaska .... I met L t . Emmons i n V i c t o r i a i n the spring of 1916; he was then an old man." The c o l l e c t i o n includes some 95 pieces i d e n t i f i e d as Tsim-shian. Some of them are among the f i n e s t known examples of Tsimshian a r t . A r t i f a c t types include masks, r a t t l e s , spoons, gambling s t i c k s , and a few other pieces. Many of them were included i n an exhibit at the Imperial I n s t i t u t e , London, November, 1954 - January, 1955. 41 The NMC received with the c o l l e c t i o n a typed catalogue pre-pared by someone with the i n i t i a l s E.H.S. I t makes several references to an " o r i g i n a l " catalogue which cannot now be located, but which must have been made by Emmons. The s t y l e i n which entries i n the second catalogue were written i s s t i l l unmistakably Emmons' s t y l e . Compare the following entries from the Emmons catalogue notes at the MAI with the Bossom c o l l e c t i o n catalogue at the NMC: 14 MAI 7/8044. Shaman's mask of maple, representing the face of a very old woman of high caste, the wrinkles and the grey h a i r representing the age, and while t h i s i s also represented by the s i z e of the l a b r e t , yet i t s s i z e and i n l a y i n g of h a l i o -t i s s h e l l also show the rank. The labret i s a very large and b e a u t i f u l specimen, and was o r i g i n a l l y worn by a woman, and l a t e r the mask was made i n order to use i t . From the Nishka v i l l a g e of Aiyansh on the upper Nass River. I t was worn by  the shaman i n h i s p r a c t i c e about the s i c k and represented a  p a r t i c u l a r s p i r i t ( i t a l i c s added). NMC VII-C-1759. Wood mask representing the face of a land o t t e r woman. The mask i s painted with a black face, red ( n o s t r i l s ) , ears, and l i p s . The mouth i s open and the wood teeth unpainted. The forehead i s ornamented with black bear f u r . The old b e l i e f throughout the Northwest Coast from the Skeena River to the Copper Delta was that the drowned turned into h a l f land o t t e r and h a l f human beings. The T l i n g i t c a l l e d them "Koushta" (land otter) "KA" (man) and believed when peo-pl e were drowning t h e i r beings assumed a human form and came o f f e r i n g to save them. The mask was worn by the Shaman i n  hi s p r a c t i c e about the s i c k and bewitched. Niska ( i t a l i c s added). MAI 9/7879. Headdress ornament of wood that can be attached to a head piece or wooden hat. I t i s carved to represent a double-headed sea animal of the snake v a r i e t y , a wholly imag-inary animal .... Used i n the winter dances which are t h e a t r i -Tsimshian shamans did not wear masks i n t h e i r r o l e as shamans. Emmons i s obviously extrapolating here from h i s knowledge of the T l i n g i t , whose shamans did wear masks while curing. 42 c a l i n character, each one t r y i n g to surpass i n grotesqueness and o r i g i n a l i t y everyone e l s e , i n masks and dress. From the Nishka ( i t a l i c s added). NMC VTI-C-1713, Dance implement used i n the more t h e a t r i c a l  winter dances. I t represents the crest of the dancer, a beaver. I t i s a wood head, carved as a beaver with copper teeth holding a s t i c k of wood that i s divided on each side i n three parts that can be opened and closed by the dancer. I t i s held by the c a r r i e r between the teeth by means of a s t i c k , p r o j e c t i n g out of the rear of the beaver head. Painted red and black. Niska ( i t a l i c s added). Mrs. 0. Morison, Tsimshian c o l l e c t o r . 1892. Nass and Skeena Rivers. FM. Accession 60, Catalogue Numbers 14609, 17826, 18001-18646. Mrs. Morison was a Tsimshian, described by James Deans (1899, quoted i n Barbeau, 1950: 460) as "an exceedingly i n t e l l i g e n t h a l f caste, her mother being a native Simshian." She made t h i s c o l l e c t i o n of some 50 pieces i n 1892, commissioned by Franz Boas for the 1893 Chicago World's Columbian Exposi-t i o n . The c o l l e c t i o n includes a wide range of a r t i f a c t types, including three house models which can no longer be found (Nos. 17826, 18001, 18003). FM catalogue notes i n d i c a t e that the f i r s t two were sent to the Brooklyn Museum, which has no record of them. This i s a r e a l l o s s , f o r her records describe housefront paintings and totem pole models which could be ex-15 ceedingly important. 15. I have recently heard that the housefronts (only) were seen at the F i e l d Museum within the l a s t few years, but have no d e t a i l s . 43 She provided r i c h documentation i n English and Tsimshian (although the orthography i s somewhat i d i o s y n c r a t i c ) . Her notes (13 typed pages) are u s e f u l primary documents i n them-selves, f o r they go beyond describing and naming the specimens and include some observations on c h i e f s , secret s o c i e t i e s , and the r e l a t i o n s h i p of iconography to myths. Regrettably, she did not provide provenience data other than, and i n only some cases, "Nass River," or "Skeena River." C.F. Newcombe, medical doctor and pr o f e s s i o n a l c o l l e c t o r , also associa- ted with the B.C. P r o v i n c i a l Museum. 1913. Niska, some Coast Tsimshian and Gitksan pieces. PM. 1499 - 1663. Niska ethnographic c o l l e c t i o n , good s e r i e s of masks, shamans' charms. 9534 - 9548. Niska c o l l e c t i o n , more charms. 9715 - 10067. Niska, Gitksan; seems to f i t into the 1499-1663 s e r i e s . The Newcombe c o l l e c t i o n was accessioned by himself f o r the P r o v i n c i a l Museum; he did not d i s t i n g u i s h between c o l l e c t i n g and accession dates. His most s i g n i f i c a n t Tsimshian c o l l e c t -ing t r i p seems to have been i n 1913, although he dates other pieces 1911, 1912, and 1914. The Newcombe papers have not yet been sorted and analyzed by the Ethnology D i v i s i o n (they are i n the B.C. P r o v i n c i a l Archives); when t h i s i s done the docu-mentation should be improved. E x i s t i n g documentation on these pieces i s p r i n c i p a l l y v i l l a g e provenience, although some 44 masks are more f u l l y documented (e.g., "potlatch mask, blue grouse, owner's c r e s t " ) . W.A. Newcombe, pro f e s s i o n a l c o l l e c t o r and associate of h i s father (above). 1905 (Nov. - D e c ) . Niska. NMC. VII-C-105 - VII-C-294. Ethnographic c o l l e c t i o n . Ac-cessioned with this c o l l e c t i o n are some few pieces from K i t -k a t l a which were c o l l e c t e d by C F . Newcombe between 1895-1901. V i l l a g e provenience only as documentation. Dr. I s r a e l Wood Powell (1837-1915), V i c t o r i a medical doctor and f i r s t Superintendent of Indian A f f a i r s f o r B r i t i s h Columbia, 1872. 1879. Coast Tsimshian and Niska. NMC. VII-C-2 - VII-C-104; VII-C-329 - VII-C-350; VII-C-434 -VII-C-435; VII-C-483. In 1873, Powell made a t r i p on the H.M.S. Boxer to the Queen Charlotte Islands, which seems to have awakened h i s i n t e r e s t i n Indian art (Robinson, 1942). In 1879, he was commissioned by the Director of the Geological Survey i n Ottawa to c o l l e c t Indian a r t i f a c t s ; some 350 pieces were sent to Ottawa i n De-cember of that year to become the nucleus of the c o l l e c t i o n i n the present National Museum of Man. The pieces were p r i -marily Haida, Nootka, Coast Tsimshian, and Niska. The Tsim-shian pieces include horn spoons, dishes, c h i l k a t s , ambelans, amhalaits, a house model, masks (including one of stone), boxes, charms, and r a t t l e s . The c o l l e c t i o n i s p r a c t i c a l l y 45 undocumented, excepting f o r loose provenience a t t r i b u t i o n s (e.g., "Tsimshian," "Fort Simpson," "Nass R i v e r " ) , some of which are questionable. For example, mask VII-C-2, i d e n t i -f i e d as "Skeena River," i s almost c e r t a i n l y Nootka. There i s a small notebook i n the B.C. P r o v i n c i a l Archives (No. AE, P87, 2) with some s c r i b b l e d p e n c i l notations by Powell which could, with considerable detective work, be used to amplify the records at the NMC. I t contains such c r y p t i c notes as " K i n c o l i t h , 20 Ju l y , a d d i t i o n a l curios purchased; war club, $20.00; 2 masks, $20.00." Some pieces c o l l e c t e d by Powell on thi s or another t r i p were sold by Emmons to the MAI. MAI 3/5010, a canoe figure head, MAI 3/5011, a s t a f f , and MAI 3/5013, a headdress, were s a i d by Emmons to have been " c o l -l e c t e d i n the 70s by the B r i t i s h Columbia Commissioner of Indian A f f a i r s . " Harlan I. Smith, archeologist, National Museum of Canada. 1925. Gitksan. NMC. VII-C-1276 - VII-C-1345. V i l l a g e provenience only. 1926. Gitksan. NMC. VII-C-1347 - VII-C-1354. Some dozen well-documented pieces purchased from Chief samedi^ks, Kitwanga. 1927. Gitksan. NMC. VII-C-1427 - VII-C-1436. Masks, other r i t u a l pieces, "bought of a Gitksan woman at Hazelton." No other documen-t a t i o n . 46 Resident C o l l e c t o r s : Thomas Crosby (1840-1914), Methodist missionary at Fort Simpson, 1870s and 1880s. 1870s, 1880s? Coast Tsimshian. MAI. 1/8900 - 1/8957, 1/8020 - 1/8199. Purchased by the MAI i n 1908. Pr i m a r i l y items from Port Simpson and Kitimat; i n -cludes r a t t l e s , masks, chests, dishes. Some pieces s a i d to have been c o l l e c t e d from the Tsimshian i n 1874. Documentation poor, p r i n c i p a l l y provenience data ssuch as "Kitamat," "Port Simpson," and "Tsimshian." NMC. VII-C-68, VII-C-84, VII-C-93. Crosby pieces purchased i n 1886. Charles C l i f t o n Perry, Indian Agent, Metiakatla. (1880s?) P r i m a r i l y Niska. NMC. VII-C-311 - VII-C-325, VII-C-331 - VII-C-336, VII-C-354 -VII-C-355, VTI-C-360 - VII-C-391, VII-C-407 - VII-C-408, VII-C-414 - VII-C-423, VII-C-451 - VII-C-453, VII-C-474 - VII-C-477. C o l l e c t i o n sold to the NMC i n 1911. Includes an e s p e c i a l l y f i n e c o l l e c t i o n of about a dozen Niska masks. Documentation i s u s e f u l , though spotty. Provenience ranges from "Aiyansh" to "Nass"; some notes on use (e.g., "Neishga Indian Dancing Mask, Halaid Dance, up to 1890"). There are no i n d i c a t i o n s as to when the c o l l e c t i o n was made. P r i e s t l y (no data on c o l l e c t o r ) . Nass River. PM. 9619 - 9696. Accessioned 1909. A t y p i c a l curio c o l l e c -t i o n : charms, horn spoons, soul catchers, etc. Four masks, 47 one f r o n t l e t , and a kerfed box. Documentation "Nass River" only. Dr. George H. Raley, Methodist minister to Kitamaat, 1893; Port Simpson,  1906; P r i n c i p a l of Coqualeetza R e s i d e n t i a l School, Sardis, 1914. (ca. 1893-1914). Coast Tsimshian, Niska, Gitksan. UBC. A-1482 - A-6568 (i n t e r m i t t e n t ) . The Raley c o l l e c t i o n covers the e n t i r e coast, and includes some 30 good Tsimshian pieces. I t i s e s s e n t i a l l y a connoisseur's c o l l e c t i o n , ac-quired piecemeal over the years. Documentation includes provenience data only, ranging from such a t t r i b u t i o n s as "Skeena River" and "Nass River" to s p e c i f i c l o c a l i t i e s , e.g., "Port Essington," "Metlakatla," and " K i t k a t l a . " Dr. Raley sold the c o l l e c t i o n to UBC i n 1948. C.V. Smith, Hazelton. Gitksan. NMC. VII-C-1183 - VTI-C-1275. A well-balanced Gitksan c o l -l e c t i o n . The pieces were purchased by the NMC i n 1925 and documented by Barbeau, apparently i n the f i e l d and probably before they were shipped to Ottawa (rather than from photo-graphs) . V i l l a g e s and oftentimes owners' names for many pieces, plus some a d d i t i o n a l information, though not of the c a l i b r e of Barbeau's documentation f o r pieces he himself c o l l e c t e d . Dr. Wv.F. Tolmie (1812-1886), medical doctor and Hudson's Bay Company fur trader. (ca. 1852). Coast Tsimshian. 48 PM. 4102 - 4123. A small c o l l e c t i o n of dishes, r a t t l e , f r o n t -l e t , masks, charms purchased i n 1927. No documentation. Mus-eum i d e n t i f i c a t i o n s "Tsimshian" or "Tsimshian or Haida," pos-s i b l y made by Newcombe. We know from Tolmie's diary (Tolmie, 1963) that he was at Fort McLoughlin ( B e l l a Bella) from 1833 to 1836, and that he v i s i -ted Fort Simpson i n 1834. He mentions i n a l e t t e r dated 1838 that he had made a l i s t of ethnological specimens f o r the In-verness Museum (which has no record of having received a c o l l e c t i o n from Tolmie), so we know he was c o l l e c t i n g by t h i s time, alghough no Tsimshian pieces are s p e c i f i c a l l y mentioned. Dr. H.C. Wrinch, Methodist medical missionary. Gitksan. NMC. VII-C-1483 - VII-C-1700. Wide range of a r t i f a c t types, inc l u d i n g s i l v e r jewelry, gambling s t i c k sets, masks. Wrinch a r r i v e d i n Kispiox i n 1901 and opened a h o s p i t a l i n Hazelton i n 1904. This c o l l e c t i o n was purchased by the NMC i n 1937, short l y before h i s death i n 1939. Documentation i s "Skeena River" only. * * * * Of the Tsimshian-made and Tsimshian-used a r t i f a c t s i n these c o l l e c t i o n s , I selected a working sample of some 1645 pieces, which I judged to have p o t e n t i a l ( i . e . , discoverable) iconographic and symbolic meanings (excluding basketry and native-made t e x t i l e s other than Chil k a t 49 16 blankets). Most of these are items decorated i n the Northwest Coast art s t y l e , but some, such as abalone earrings and l a b r e t s , are not. Of these pieces, I was able to photograph or purchase photographs of somewhat more than 1,000, some i n color, some i n black and white, some i n both. In addition to the museum specimens, I studied f i e l d photo-graphs taken by Barbeau, H.I. Smith, Emmons, C.F. Newcombe, and others, i n the PM and the NMC. I also studied the photographs of Tsimshian t o -tem poles published by Barbeau (1929, 1950), and a c o l l e c t i o n of photo-graphs taken by Wilson Duff i n 1952 of totem poles at Gitanmaks, Kispiox, 17 Kitsegukla, Kitwanga, and Kitwancool, now i n the PM (see Duff, 1952). To conclude t h i s chapter, i t i s important to consider the re l a t i o n s h i p s between the two classes of data used i n the study: the Barbeau/Beynon notes and the museum specimens. Sturtevant (1973: 9-10) distinguishes between two types of museum studies: contextual studies, which " t r e a t the r e l a t i o n between the objects and some n o n - a r t i f a c t u a l aspect such as s o c i a l structure, r i t u a l , psychology, or the character and l i f e h i s t o r y of i n d i v i d u a l a r t i s t s , " and formal studies, which " t r e a t the context as given and analyse the objects themselves." This study i s of the f i r s t or contex-t u a l type. By analyzing the Barbeau/Beynon f i e l d data, i t seeks to erect a contextual framework, or g r i d , within which items of Tsimshian material culture can be interpreted or made meaningful. While the f i e l d data, 16. The horn spoons i n t h i s sample include only those for which some s p e c i f i c data were c o l l e c t e d . There are hundreds more i n the c o l -l e c t i o n . 17. The Ethnology D i v i s i o n of the PM graciously supplied me with p r i n t s of most of these photographs free of charge. 50 notably the l i s t s of crests and s p i r i t names, c e r t a i n l y cannot be con-sidered complete, they are an adequate sample of the idea systems they represent f o r the purpose at hand. Further, because of acc u l t u r a t i v e loss of these idea systems by the contemporary Tsimshian, they are data which cannot be s i g n i f i c a n t l y increased. The sample of material representations or embodiments of these idea systems — the museum specimens used i n the research — i s not an adequate sample of the larger universe i t represents. Fortu-nately, i t can be increased as more of the Tsimshian a r t i f a c t s i n the world's museums are i d e n t i f i e d and interpreted i n the context developed here. The museum sample used i s , however, adequate to e s t a b l i s h the r e l i a b i l i t y of the i n t e r p r e t i v e framework, i n that crest representa-tions with the s p e c i f i c a t t r i b u t e s described i n the crest l i s t s were found i n museums. But a great many more museum specimens must be brought into t h i s framework before the goal of a reasonably complete material culture re-presentation of the Tsimshian crest and s p i r i t name systems w i l l be accomplished. Then, i t w i l l be possible i n Sturtevant's words, to " t r e a t the context as given and analyze the objects themselves." For the present, the sample of a r t i f a c t s i s only adequate to be used as i l l u s t r a t i o n s of a larger universe we now know was created, and can only hope has been preserved. 51 CHAPTER THREE SOCIAL ORGANIZATION The following o u t l i n e of Tsimshian s o c i a l organization i s included as background f o r the analysis which follows i n l a t e r chapters. I t i s not a complete statement, i n that i t does not develop a formal model and systematically follow i t s implications f o r Tsimshian s o c i a l behaviour. However, i t does amplify the discussions of Tsimshian s o c i a l organization i n the published l i t e r a t u r e , and therefore makes an o r i g i n a l contribution. The focus of the chapter i s the development of an a n a l y t i c a l d i s t i n c t i o n between two s t r u c t u r a l orders i n Tsimshian so c i e t y : the 1 one r i t u a l l y expressed i n the potlatch, the other i n h a l a f i t ' s . The potlatch r e f l e c t s and celebrates the order of descent groups and a f f i -n a l t i e s , of clanship and exogamy. Conceptually opposed and based upon a d i f f e r e n t s t r u c t u r a l p r i n c i p l e — that of c o n t r o l l i n g supernatural power — i s the hala.'it order, which r e s u l t s i n s o d a l i t i e s c u t t i n g across the clan structure. Tsimshian c h i e f t a i n s h i p was o r i g i n a l l y a p o s i t i o n a l locus i n the potlatch order; that i s , chiefs were clan chiefs and sub-clan c h i e f s . 2 Among the Coast Tsimshian, however, an o f f i c e of t r i b a l chief had dev-eloped which extended the chief's hegemony to include the l o c a l lineages 1. See page 74 for a d e f i n i t i o n of h a l a / i t . 2. " T r i b e " i s used herein to r e f e r to the 26 l o c a l groups of the Tsim-shian ( g ' i t x a ^ a , g ' i s p a x l ? " ts , g* i t x a t * i.'n, k* itwang.E', e t c . ) . Since each group usually occupied a s i n g l e winter v i l l a g e , " t r i b e " i s e s s e n t i a l l y synonomous with " v i l l a g e . " See p. 13 f f . above. 52 of other clans. Concomitantly, the chief was extending the s t r u c t u r a l basis of h i s power v i s - a - v i s the h a l a ' i t order. The same process, a l -though l e s s developed, can also be seen among the Niska and Gitksan. I i n t e r p r e t t h i s intermeshing of the two orders i n the p o s i t i o n of chief as evidence of a development i n Tsimshian s o c i a l organization from a t r i b a l l e v e l of i n t e g r a t i o n to a chiefdom l e v e l . Rather than attempting to account for the changing power of the chief by reference to economic or other f a c t o r s , I w i l l discuss the way i n which the tr a n -s i t i o n i s mirrored i n r i t u a l and i n the metaphors, both material and non-material, with which the Tsimshian expressed the r o l e and functions of the chief. Other aspects of s o c i a l organization are discussed as they bear upon the above d i s t i n c t i o n or, as i n the problems of post-marital residence and status l e v e l s , i n order to c l a r i f y what I see as issues l e f t unclear or unresolved i n the e x i s t i n g l i t e r a t u r e . The chapter i s organized as follows: descent and succession, marriage, descent groups, chiefs and s o d a l i t i e s , rank, status l e v e l s , and trading r e l a t i o n s h i p s . Appendix I i s a synoptic presentation of t r i b a l and descent group or-ganization, prepared i n support and a m p l i f i c a t i o n of the present chapter. In what follows, c u l t u r a l d e t a i l s , i n the sense that these constitute the idiom of s o c i a l r e l a t i o n s h i p s , have been kept to a min-imum. This follows a conception of the r e l a t i o n s h i p of society and culture which has been s u c c i n c t l y expressed by Geertz (1967: 233-234): "one of the most useful ways ... of d i s t i n g u i s h i n g between culture and s o c i a l system i s to see the former as an ordered system of meaning and 53 symbols, i n terms of which s o c i a l i n t e r a c t i o n takes place; and to see the l a t t e r as the pattern of s o c i a l i n t e r a c t i o n i t s e l f . " The purpose of the present chapter i s to present an out l i n e of those enduring or r e p e t i t i v e t i e s which bind the Tsimshian together and define t h e i r constituent parts. They are the r e l a t i o n s h i p s celebrated i n r i t u a l (see Leach, 1965: 15-16) and expressed i n a r t . The goal of the chapter, then, i s to express these r e l a t i o n s h i p s i n anthropological terms; the goal of the next three chapters w i l l be to show how the Tsimshian ex-press some of these same re l a t i o n s h i p s with t h e i r a r t . Descent and Succession Descent was reckoned m a t r i l i n e a l l y , with succession to names and positions going i n general to the eldest man most c l o s e l y r e l a t e d to an incumbent. G a r f i e l d (1939: 179) ranks succession i n the following order, one that i s obviously predicated on an avunculocal residence pat-tern i n which the successor would have been resident with and trained by h i s predecessor: 1. own next younger brother (same mother) 2. own eldest s i s t e r ' s eldest son 3. next younger p a r a l l e l cousin (man having same maternal grand-mother as holder) 4. eldest house nephew (son of a woman of the same house and gen-3 eration as holder) 3. This was probably the son of one of Ego's m a t r i l a t e r a l p a r a l l e l cousins. 54 5. eldest man of a r e l a t e d house, i n own or another t r i b e 6. adopted man. While t h i s may r e f l e c t an i d e a l order, actual succession involved a num-ber of s i t u a t i o n a l f a c t o r s , and there was often controversy among po-t e n t i a l successors. Rosman and Rubel (1971: 32-33) argue convincingly that "the lack of f i x e d rules of succession makes the a b i l i t y to pot-l a t c h f i r s t an important requirement for succession." Marriage The Tsimshian are s a i d to have a rule of preference f o r mar-riage with mother's brother's daughter (Boas> 1916: 440; G a r f i e l d , 1939: 231; 1966: 23). This was probably the case, but the evidence i s exceeding slim. Boas (1916: 440) says that t h i s was the "normal type of marriage, as described i n the t r a d i t i o n s , " but has no evidence other than myths to o f f e r . G a r f i e l d also bases her statement regard-ing marriage preference on myths, with a footnote reference to the same statement by Boas: "the normal marriage, according to the myths, i s with a cross-cousin" (1939: 231). In her l a t e r popular summary of Tsimshian cul t u r e , she again r e l i e s on the evidence from myths: "the i d e a l marriage, indicated time and again i n Tsimshian mythology, was between a man and the daughter of h i s mother's brother" ( G a r f i e l d , 1966: 23) . The only corroborative statement regarding cross-cousin mar-riage contained i n the Barbeau/Beynon f i e l d notes i s the following from Robert Stewart of the Niska to Beynon i n 1948-49: "the matches most sought were with f i r s t cousins ( l g u t 7 x a > ? / ) to r e t a i n the wealth of two 55 c l o s e l y r e l a t e d f a m i l i e s , e.g., brother and s i s t e r . " According to Sapir (1920: 263), the k i n term kwutxa' w*y" 'i denotes "my cross cousin, i.e.,. father's s i s t e r ' s c h i l d , mother's brother's c h i l d . " There are almost no data at a l l as to how frequent such marriages were. G a r f i e l d says only that " i n the genealogies obtained a very small number have been cousin marriages, even with the r e l a -tionships further removed than f i r s t cousins. Three f i r s t cousin mar-riages are known i n Port Simpson now" (1939: 232). Both myths and descriptions of marriage ceremonies recorded by Barbeau r e f e r to the bride accompanying her husband to h i s home. The problem with a m a t r i l i n e a l society such as the Tsimshian i s to determine where the husband's home was — with h i s parents or h i s maternal uncle. In 1916, Boas decided i t was with h i s parents: "the evidence of Tsimshian mythology shows that c h i l d r e n grew up i n the houses of t h e i r parents, and that the newly married couple l i v e d with the young husband's parents" (Boas, 1916: 499). However, by 1935 he had evidently reconsidered and chosen to avoid the issue altogether, although he hinted at an avunculocal s o l u t i o n : "cross-cousin marriage i s favored. The young couple l i v e i n the husband's home. There are many references to the r e l a t i o n between a man and h i s s i s t e r ' s son" 4 (Boas, 1966, 1935: 303). 4. Apparently, avunculocal post-marital residence patterns were often confusing to e a r l i e r generations of anthropologists. Referring to Birket-Smith and De Laguna's 1938 work on the Eyak, Murdock (1949: 220) wrote: "the authors, being unfamiliar with the category of avun-cu l o c a l residence, devote considerable space to the c o n f l i c t i n g s t a t e -ments of d i f f e r e n t informants as to whether the residence r u l e i s mat-r i l o c a l or p a t r i l o c a l . Since a s i m i l a r confusion i s usual for demon-st r a b l y avunculocal s o c i e t i e s , and since avunculocal residence i s common i n the region, being found for example among the T l i n g i t , i t seemed l i k e l y that the Eyak follow the same r u l e . " 56 Levi-Strauss also considered the problem of Tsimshian post-m a r i t a l residence i n h i s s t r u c t u r a l analysis of the myth of Asdiwal (1967a). Here he faced the s i t u a t i o n sometimes found i n the myths where residence following marriage with supernatural beings i s sometimes p a t r i l o c a l , and sometimes m a t r i l o c a l . He concluded that "mythical speculation about types of residence which are e x c l u s i v e l y p a t r i l o c a l or m a t r i l o c a l do not ... have anything to do with the r e a l i t y of the structure of Tsimshian society, but rather with i t s inherent p o s s i b i l -i t i e s and i t s latent p o t e n t i a l i t i e s " (1967a: 30). Reasonable enough. However, his further conclusions as to what the Tsimshian pattern actu-a l l y was seems unnecessarily complicated: In r e a l l i f e , the [male] children grew up i n the p a t r i -l o c a l home. Then they went to f i n i s h t h e i r education at t h e i r maternal uncle's home; a f t e r marrying, they returned to l i v e with t h e i r parents, bringing t h e i r wives with them, and they s e t t l e d i n t h e i r uncle's v i l l a g e only when they were c a l l e d upon to succeed him. Such, at any rate, was the case among the n o b i l i t y , whose mythology formed a r e a l "court l i t e r a t u r e " (Levi-Strauss, 1967a: 30). Although there are no e x p l i c i t ethnographic statements of Tsimshian post-marital residence preference, i t seems preferable to me to p o s i t a straightforward avunculocal residence pattern i n which the boy went to l i v e with h i s mother's brother as a c h i l d and continued to l i v e there, l a t e r bringing h i s wife to j o i n him, and eventually suc-ceeded to h i s uncle's p o s i t i o n while already resident with him and under h i s continued tutelage. In the case of preferred cross-cousin marriage, h i s wife, being h i s mother's brother's daughter, would a l -ready be resident i n the same household, and i t seems u n l i k e l y that the newly married couple would leave to l i v e with h i s parents, and then re-turn a f t e r h i s uncle (her father) had died. 57 This i s supported by Garfield's (1939: 277) d e s c r i p t i o n of the normal composition of the household: In the winter dwelling l i v e d the head, his mother i f she were a widow, h i s wife and small c h i l d r e n , h i s widowed or divorced s i s t e r s and t h e i r small c h i l d r e n , also h i s younger brothers and cousins (mother's s i s t e r ' s sons) and h i s grown nephews with t h e i r wives and chi l d r e n . There would also be other r e l a t i v e s , according to circumstances, but these were the usual ones represented. Avunculocal post-marital residence, as suggested above, pro-bably should, however, be considered a preferred rather than an absolute pattern or r u l e , leaving open the p o s s i b i l i t y that younger sons, i . e . , those not l i k e l y to succeed to an uncle's p o s i t i o n , perhaps did reside a f t e r marriage with t h e i r parents. This would mean, of course, that they would be l i v i n g i n dwellings owned and c o n t r o l l e d by men not of t h e i r clan. There i s some comparative data from the T l i n g i t which sup-5 ports an avunculocal post-marital residence pattern. Olson (1967: 5) says that T l i n g i t post-marital residence was p a t r i l o c a l , but also that since i t was customary to send boys of seven or eight years to l i v e with mother's brother, " a l l males of the household above that age be-longed to the clan 'owning' the house." He also describes the male composition of the household as co n s i s t i n g of "various combinations of brothers, married nephews of these, or men considered as 'children of s i s t e r s ' [men "who would normally go to l i v e with t h e i r mother's 5. The data for the Haida are not as c l e a r ; although Murdock (1949: 72) believes t h e i r residence pattern to be avunculocal. The Haida case was complicated by a period of m a t r i l o c a l bride service not shared by the T l i n g i t and Tsimshian (Swanton, 1905: 50). 58 brothers, who would often be members of the same household"] ( l o c .  c i t . ) . F i n a l l y , he reports that "when a house chief died, h i s suc-cessor was chosen from among h i s housemates on the basis of wealth and wisdom. But the nephew was usually the h e i r designate" ( i b i d . ; 6). More e x p l i c i t and repeatedly-confirmed by Tsimshian people's statements to Barbeau and Beynon was the rule among the Coast Tsimshian that marriages of the upper or c h i e f l y status l e v e l be endogamous. Among the Gitksan and Niska, who did not have as c l e a r l y defined status l e v e l s (see discussion below), i t was s p e c i f i e d that the ch i l d r e n of a chief marry into other c h i e f l y f a m i l i e s : " i t i s a very s t r i c t law that a chief i s not allowed to marry a common woman, h i s children could never become c h i e f s " (people of Kitwancool, i n Duff, 1959: 38). Accord-ing to one of Beynon's Coast Tsimshian teachers, " i f a r o y a l prince marries a woman of the common class-against the w i l l of h i s people, h i s c h i l d r e n are looked upon as wa ?ayin (lower class) children and he loses h i s chief standing." G a r f i e l d , however, says that the ch i l d r e n of such unions could be elevated to high rank: "a c h i l d who had one parent of c h i e f l y rank and one of common rank could not hope to secure recognition as a member of the higher class except through the most l a v i s h g iving of potlatches by h i s parent and himself" (Garfield,1939: 232) . One r e s u l t of the rule to marry within the upper status l e v e l was that marriages were continually contracted between the c h i e f l y houses of d i f f e r e n t t r i b e s , serving to a l l y them. Important chiefs were sa i d to 59 6 have married many women, often coming from several d i f f e r e n t t r i b e s . Descent Groups Houses; The basic s o c i a l unit i n a l l three d i v i s i o n s of the Tsimshian was a corporate matrilineage c a l l e d a "house" (wglp) and named a f t e r i t s highest-ranking c h i e f ' s name. Larger lineages f i s s i o n e d i n -to branch lineages, each named a f t e r i t s own house c h i e f , but subor-dinate to the highest-ranking chief of i t s major branch or segment. The house as a m a t r i l i n e a l descent group was not coterminous with the household, discussed above, i n which l i v e d members of the matrilineage plus t h e i r a f f i n e s and c h i l d r e n belonging to other lineages. The l a r g e r matrilineages often occupied more than one dwelling, which were named according to a d i f f e r e n t system: dwelling names were i n h e r i t e d as crests (ayuks), of which high-ranking lineages usually owned sev-e r a l . Branch lineages could draw upon the stock of house names owned by the parent house. The house was the p r i n c i p a l resource-owning corporation i n Tsimshian society. I t s resources included f i s h i n g , hunting, and gather-ing t e r r i t o r i e s or l o c a l i t i e s , which were exploited under the d i r e c t i o n of the house chief. The house also owned a stock of ceremonial p r i v i -leges: names (of several types), c r e s t s , myths, songs, and feast pre-rogatives, which were also under the control or stewardship of i t s c h i e f . We knowwthat houses f l u c t u r a t e d widely i n s i z e , at times r e -s o r t i n g to adoption to prevent e x t i n c t i o n , at other times growing so 6. The sections below on rank and status l e v e l s contain a d d i t i o n a l i n -formation on c h i e f l y marriages. 60 large that they f i s s i o n e d into two or more separate houses. But there are no descriptions i n the Barbeau notes, or i n the published l i t e r a -ture, of intra-house composition and dynamics. The l i n e s of f i s s i o n mentioned i n t r a d i t i o n a l narratives were between brothers. Clans: Each house and i n d i v i d u a l belonged to a l a r g e r , exo-gamic, s t i p u l a t e d , m a t r i l i n e a l k i n group which i s usually c a l l e d a 7 phratry i n the l i t e r a t u r e , but which I am c a l l i n g a clan. There were four, each represented i n a l l three d i v i s i o n s (the names of two of them change with the Gitksan), although not a l l four clans were to be found i n a l l t r i b e s . The names and p r i n c i p a l crests of each clan are as f o l -lows : 7. Boas v a c i l l a t e d on descent group terminology. He usually trans-l a t e d the Tsimshian word ptEx as " c l a n " but he also used " c l a n " to r e f e r to what I am c a l l i n g a "sub-clan" (see Boas, 1916: 488, 500 footnote). When he wished to avoid a l l ambiguity, he c a l l e d the l a r g e r groups simply "exogamic groups" ( i b i d . : 488). G a r f i e l d , whose f i e l d work was done under Boas' d i r e c t i o n , c a l l e d the l a r g e r groups "clans" i n her d i s s e r t a t i o n (1939), but l a t e r switched to "phratry" (1966). Barbeau con s i s t e n t l y used "phratry" i n h i s pub-l i c a t i o n s , and can probably be c r e d i t e d with i t s general popularity. I have abandoned the term "phratry" because of i t s misleading connotations. A phratry, by d e f i n i t i o n , i s a group of clans, c f . Bohanan (1963: 142): "a phratry i s a c o l l e c t i o n of clans (whatever that may mean) joined, usually on nonkinship p r i n c i p l e s . " The use of the term i s therefore inappropriate i n the Tsimshian instance, f o r the larger exogamic groups sometimes c a l l e d phratries are not, i n f a c t , groups of smaller u n i l i n e a l s t i p u l a t e d groups or clans. They do contain some c l e a r l y defined a l l i a n c e s of lineages which I am c a l l i n g sub-clans. These are not, however, uniformly d i s c e r n i b l e (see discussion of sub-clans i n the t e x t ) . 61 Clans Coast Tsimshian , anha'da* (?) laxk* i b u v ("on the wolf") g * isp^wudwa'da (?) laxsk* i ' k ("on the eagle") Niska " II II II ** Gitksan l a x s e . l (g.anha'da i n Kitsegukla) II g ' i s t . a / s t ("people of the fireweed") II Raven Frog Crests Wolf Bear Killerwhale G r i z z l y Eagle Beaver * g.anha'da may be derived from the T l i n g i t Raven clan name ganaxadi ("people of ganax") ( G a r f i e l d , 1966: 19). •** l a x s e / l may be derived from the Tsimshian word for the T l i n g i t v i l -lage at Cape Fox: l a x s e l j . Table I. Tsimshian Clans and Their P r i n c i p a l Crest Animals Clan exogamy was extended to the corresponding clans (moieties) of the neighbouring T l i n g i t and Haida, f or which the four Tsimshian clans were grouped i n two p a i r s , as follows: Haida Ravens T l i n g i t Wolves T l i n g i t Ravens Haida Eagles crests Tsimshian g r i z z l y raven g"ispjwudwa'&3 g' i s g . a.'st k i l l e r w h a l e frog laxk'ibu' wolf eagle bear beaver Tsimshian g.anha'dp laxse.'l l a x s k " i ' k Table I I . Tsimshian, Haida, and T l i n g i t Clans and Crests 62 In the preceding table (II) , which includes the correspondences of the eight major crest animals of the Tsimshian with those of the Haida and T l i n g i t , the Tsimshian g*ispawudwa7dd and laxk'ibu' are paired with the Haida Ravens and T l i n g i t Wolves, and the Tsimshian g.anhavda and laxsk*i?k with the Haida Eagles and T l i n g i t Ravens. This two-pair grouping of Tsimshian clans corresponding to Haida/ T l i n g i t moieties i s s i g n i f i c a n t , f or i t may r e f l e c t an o r i g i n a l moiety d i -v i s i o n shared by the Tsimshian with t h e i r two neighbours i n the Northern Province. In other words, the Tsimshian four-clan system may have developed from an e a r l i e r two-clan system. The av a i l a b l e evidence, i n addition to the crest correspondences themselves, i s f a r from conclusive. I t i s c e r t a i n l y suggestive, however, that the two largest and, on the basis of the t r a d i t i o n -a l n a r r a t i v e s , o r i g i n a l or indigenous Tsimshian clans are the untranslatable g.anha/dg and g'ispawudwa'da, and that most houses of the laxsk*I'.k ("on the eagle") and laxk*ibu/ ("on the wolf") consider themselves immigrants from the Haida, T l i n g i t , and Tahltan. A l s o , there i s the evidence of a sub=clan within the Niska la x s k * i . k whose members c a l l e d themselves the laxsgmE.*]_ix ("on the beaver"). The laxsjmE/lix could not use the dominant eagle crest of t h e i r c l a n , but used the beaver instead, and seem to have considered themselves quite d i s t i n c t from the other laxsk*I'.k. This case might be a recent example of the same process whereby the laxsk*i'.k and laxk*ibu^ separated from t h e i r two parent clans. S t i l l another pertinent example from the Niska i s the l a x t i y j q i : sub-clan which was laxk'ibu' at g' i t w i n k s i l k , g * ispg wudwa 'dp at g* i t l a x d a /inks, and c a l l e d themselves laxkibumg* i s g . a f s t ("wolf of fireweed") among the g* i t x a t * i.'n. One of the members of t h i s 63 group (Lazarus Moody, wihjn) described them as "more l a x k * i b u / than they are g* ispawudwa'dg. A man of t h i s family may marry a g* ispawudwa'd<9. From the beginning t h i s family could not marry a laxk*ibu ' . They are not r e a l g* ispgwudwa'd^." What seems to be happening here i s that a g* ispawudwa/d<9 sub-clan i s i n the process of redefining i t s e l f as laxk* i b u y . I t i s pro-bable that a d e t a i l e d analysis of the Tsimshian narratives might shed addi-t i o n a l l i g h t on t h i s postulated e a r l i e r two-clan system. Clan members shared a f e e l i n g of kinship and expected h o s p i t a l -i t y from each other. Among the Coast Tsimshian, they c a l l e d each other wulE^isk ( " r e l a t i v e s " ) . According to one person, clan l o y a l t i e s overrode t r i b a l ones: " i n case of warfare between the Tsimshian and the Haida, a Tsimshian would help one of the same crest among the Haida, and vice-versa. In times of peace, i t i s a law that one i n one crest w i l l help another i n the same crest. There i s a bond of r e l a t i o n s h i p between them a l l i n one 8 crest" (Herbert Wallace). Notice that the English word " c r e s t " i s used above to re f e r to the clan. The Tsimshian word i n the same context would have been ptEx. This i s usually translated as "clan" or "exogamic group," but i t i s also used to re f e r to the primary crest animal of a clan, which I beli e v e to be i t s p r i n c i p a l or o r i g i n a l referent (see discussion of t h i s point on page 113 below). The importance of a f f i n a l r e l a t i o n s h i p s between clans, and t h e i r continued renewal by new generations, was c l e a r l y recognized. On the 8. Dir e c t quotations from the Barbeau/Beynon notes are i d e n t i f i e d herein, wherever possible, by the name of the person supplying the information. 64 Nass, for example,"the l a x s k ' i f k chose t h e i r women from the kwaxcu (laxk* ibu/) group and the laxk* i b u x always married into the laxsk* i ' k group. So though at times there was open s t r i f e , they often l i v e d together i n peace; always the two groups were united by marriage" (Peter Calder) . Many crest and migration myths (ada.'ox) begin with two intermarrying clans l i v -ing i n separate v i l l a g e s , usually across a r i v e r from each other. Since a m a t r i l a t e r a l cross-cousin marriage system requires at l e a s t three i n t e r -marrying groups, i t i s probable that the other clan was divided into two groups, one of which included father's lineage and the other of which i n -cluded wife's lineage. A f f i n a l r e l a t i o n s h i p s were "honoured" i n potlatches by non-returnable contributions from in-marrying lineages to the host lineage's potlatch fund. A f f i n a l r e l a t i o n s h i p s between clans were also expressed i n a r e l a t i o n a l naming system unique to the Tsimshian on the Northwest Coast. This i s a system of children's names which were owned by the matrilineage but r e f e r r e d to p h y s i c a l and behavioural c h a r a c t e r i s t i c s ("various-ways and manners," according to one person) of the major crest animal or animals of the father's clan. The names have both a short or abbreviated form and a long form. Some examples owned by g* itxa'-fca houses are as follows: I. House of h E . l , g"ispgwudwa'dg (short form): g.amaya.m (long form): g.amhaiyaim g.Ex (only mocking) (raven) ( t r a n s l a t i o n ) : "mocking raven" (father's c l a n ) : g.anha'dj 65 I I . House of h E . l , g'ispjwudwa'dJ (short form): g.ayE' (long form): 'wat'i gayE.m g i p a i k j l x s k i f k (never) (zigzag) ( f l y i n g of) (eagle) ( t r a n s l a t i o n ) : "the eagle never f l i e s crooked (but f l i e s s t r a i g h t ) " (father's c l a n ) : laxsk'iCk I I I . House of 'ExiEwE.ls, g.anha'da (short form): dzag.am txE nE Ex (long form): dzag.am txE nE'Exl 'nE.xl (towards shore) ( f l a t ) ( f i n of) (killerwhale) ( t r a n s l a t i o n ) : "the k i l l e r w h a l e swims towards shore with f l a t f i n " (father's c l a n ) : g*isp3wudwa/da IV. House of 'wgkafs, g.anha'd^ (short form): ^ni.s'a'yin (long form): 'ayinl'na g.a'p^t xski'k (no) (of) (food) (eagle) ( t r a n s l a t i o n ) : "the eagle has nothing to eat" (father's c l a n ) : l a x s k ' i ' k V. House of 'aya'ig. ansk. g.anha'da (short form): lag.alEmdz^x (long form): lag.axlEmdzal ' n a k l k j l gibEo (at each end entering) ( o f f s p r i n g of) (wolf) ( t r a n s l a t i o n ) : "the o f f s p r i n g of the wolf enters from each end" (father's c l a n ) : laxk'ibu' Sapir (1915: 26) seems to have f i r s t understood and named these names, c a l l -ing them " c r o s s - p h r a t r i c " names; i n the context of t h i s paper, they would be properly c a l l e d "cross-clan" names. 66 Duff (1964b: 67-68) believes that the short form of the name could be completed by adding an appropriate reference to any of the other three clans to which the father might belong. He gives the example of a g.anha'dg name from Kitsegukla, ni.gamks ("on sunshine") which, i f father was of the g*isg.a/st clan, might be completed as "sunshine g l i n t i n g on the wet dorsal f i n of the emerging K i l l e r Whale"; i f father belonged to the lax s k * i ^ k , the same name might be completed as "sunshine g l i n t i n g on the white head of the Eagle" (loc. c i t . ) . This may have been true i n some cases, but does not accord with native theory, nor with my own reading of the evidence presented by the names themselves. According to several people, the names were o r i g i n a l l y bestowed by the father, a f t e r which they remained permanently i n the possession of the matrilineage, to be bestowed on successive children born to them from a father of the same clan: "the children's names were given by the father, according to the d i f f e r e n t crests of h i s house. And these names remain i n the children's family (mother's side) and are reapplied i n t h e i r own family. In t h i s manner the children's names spread i n a l l d i r e c t i o n s " (H. Wallace). This and s i m i l a r statements suggest to me that the e n t i r e name was formula-ted by the father with reference to h i s own crest. Given the assumption that marriages continued to unite the same two lineages or clans, there would be no d i f f i c u l t y i n reapplying them to l a t e r children. There are no ind i c a t i o n s i n the f i e l d notes as to what circumstances would e l i c i t the creation of a new name by the father, as opposed to giving the c h i l d a name already owned by the mother's house. 67 The above i n t e r p r e t a t i o n i s borne out by examination of the names themselves. There are simply too many short forms of the names on record that have d i r e c t and unmistakable reference to s p e c i f i c crest animals to have been applied equally w e l l to bears, frogs, eagles, wolves, f i r e -weeds, and ravens. The following short forms of Kispiox names r e f e r r i n g to frogs (major crest animals of the l a x s e ^ l clan) seem to me to bear t h i s out: "out from shore jumps" (the small f r o g ) , "on d r i f t l og" (on which s i t s the small f r o g ) , "to wrinkle up" (the hips of the small f r o g ) , "among grass" (the f r o g ) , " e n t i r e l y covered with green" (the f r o g ) , "drink throat" (the small frog; the small frog drinks water with the s o f t parts under the chin), "across flabby" (the b e l l y of the large f r o g ) , " i n t o water jumps" (the small f r o g ) , ( l i k e ) "red salmon" (the small frog; the o f f s p r i n g of the frog has a red b e l l y l i k e the salmon). S i m i l a r l y , the following short forms des-cribe wolves too w e l l to have been also applied, f o r example, to frogs: "together hunt" (the small wolves), "together attacking" (the wolves), " f a r away moving" (the wolf, whenever anybody comes to l i v e nearby), "taking away dog" (the large wolf), "prowl around" (the o f f s p r i n g of wolf), "under between i t s legs" (the wolf places i t s t a i l ) . D etailed l i n g u i s t i c and semantic analysis i s needed before the cross-clan naming system i s completely understood. I t would also be i n s t r u c -t i v e to analyze the names with reference to the past marriage preferences they record, i . e . , do c e r t a i n lineages own a preponderance of cross-clan names which r e f e r to t h e i r continually marrying i n t o one clan rather than the other two from which t h e i r marriage partners might also have been s e l -ected. 68 Sub-clans: Lineages which shared the same myth and crests con-sidered themselves to be more c l o s e l y r e l a t e d than other clan members, and form a d i s t i n c t intermediate category of Tsimshian descent groups. These are not, however, uniformly d i s t r i b u t e d or d i s c e r n i b l e , and seem to have been emerging only at c e r t a i n nodes i n the clan structure, probably as a r e s u l t of greater population growth at these points. Barbeau (1917a), Sapir (1915), and the l a t e r G a r f i e l d (1966: 18-22) c a l l e d these groups of lineages "clans"; Boas (1916: 488) considered them not " s u f f i c i e n t l y w e l l marked to be c a l l e d 'clans' i n contrast to the larger exogamic d i v i s i o n s " ; and I am c a l l i n g them sub-clans. A Coast Tsimshian c a l l e d other sub-clan members "my brothers" (wE /kyEyot) and "my s i s t e r s " (t^m'kti/tkw) and was s a i d to stand i n a "brother" (wE^kgt, sing.) r e l a t i o n s h i p to them or to "be wEkgt." That the sub-clans were r i t u a l l y , though not economically, corporate was explained i n a myth as follows: "at Temlaham l i v e d three brothers, tsibasE x, a l i m l a . x E / , and ni.shaiwaxs. Each was a chief, each had h i s own group, hi s own house, and h i s own hunting t e r r i t o r i e s . But a l l had the same myths, dirges, nur-sery songs, and crests i n common. Each had h i s own i n d i v i d u a l naxnj'x's [ s p i r i t names]" (Prevost and Aukland). Other people explained the d i f f e r e n t degrees of relatedness between sub-clan and clan members thus: "when many people have the same myth of o r i g i n they term each other wE'k^t ("brother"); i n other cases they term each other wulE.'isk ( " r e l a t i v e s , but not blood r e l a -t i v e s " ) " (Wallace and Ryan). The sub-clans stand out most c l e a r l y among the Coast Tsimshian g*ispgwudwa^da and the Niska laxsk*if.k and laxk* i b u / . Among the Coast 69 g* ispawudwa /dg > the houses which claimed the Temlaham myth of o r i g i n i n the i n t e r i o r , and those whose ancestors went to the undersea house of the mon-s t e r nagunaks, c a l l e d themselves g*itamlax'am and g*itnagunaks, r e s p e c t i v e l y . Among the Niska l a x s k ' i ' k , there were important sub-clans named sgmlaxsk'i'k ("real on the eagle") and laxs-jmE'lix ("on the beaver") . Important laxk' i b u 7 sub-clans were the g* isg.ansna.t ("people of the saskatoon bushes") and g*itwilnagE^el ("people by themselves"). Another Nass River sub-clan, the l a x t i y j q l ("on the tiy? /q i , " s a i d to be a place near Rivers I n l e t ) , has a l -ready been described as being laxk*ibu" i n one t r i b e , g* ispgwudwa'da i n ano-ther, and intermediate between the two clans i n a t h i r d . Only f o r the Niska did people consistently l i s t houses by sub-clans which, i n turn, were ranked within clans (see Appendix I and Barbeau, 1917a, f o r the names of other sub-clans of the Niska and Coast Tsimshian). The names of "clans" (my sub-clans) Barbeau reports for the Gitksan (see 1929: 18-22) are, i n f a c t , names he coined himself from myths shared by re l a t e d houses (hence, h i s "Frog-Woman," "Tongue-Licked," "Water-L i l y , " "Wild-Rice," etc., clans are not c a l l e d such by the people them-selves) . Gitksan people did spec i f y that c e r t a i n houses were more c l o s e l y r e l a t e d than others, notably those that shared the same myth and c r e s t s , but there are no ind i c a t i o n s i n the Barbeau/Beynon notes that they formed d i s t i n c t named sub-clans such as were reported f o r the Coast Tsimshian and Niska. The s t r u c t u r a l l y most s i g n i f i c a n t c h a r a c t e r i s t i c of the sub-clans named above i s that they potlatched jto rather than with each other. "When one of the l a x k * i b u " groups, such as the g*isg.ansna.t, gave a feast they would i n v i t e the other l a x k * i b u ^ groups as guests, but would be as s i s t e d 70-only by t h e i r own groups" (Peter Adams, g*isg.ansna.t ). The ni.yuks group of Coast Tsimshian g* isp-gwudwa'dd " i n feasts ... go together with suhalait [chief of the g ' i s p a x l ? " t s ] . The l a t t e r i s the only one (of the g'ispawudwa/dg) who contributes to h i s yE"jk [potlatch]. Temlaham and Gitnagunaks are i n -v i t e d as guests" (Johnson). "There were now two d i s t i n c t l a x k * i b u x groups. They did not intermarry, but neither did they ... contribute to each other's potlatches. At a funeral feast of the g* isg.ansna.t, only they paid dewgl or death duties; the g* itwilnagE *el came as guests, l i k e the laxsk* i ^ k and g.anha' da" (Matthew Gurney and Emma Wright) . Data such as the above d i r e c t quotations from Tsimshian people are few, but they leave no doubt that sub-clans existed which potlatched to each other i n the same way that clans d i d , i . e . , they stood to each other i n the r e l a t i o n s h i p of hosts to guests. Unlike clans, however, they did not intermarry. While the major premise of Rosman and Rubel's (1971: 204) model of the potlatch type society that "one potlatches to one's a f f i n e s " i s generally c o r r e c t , the potlatching of sub-clans within a clan shows that potlatching and exogamy are not n e c e s s a r i l y l i n k e d i n Tsimshian thought. Group r e l a t i o n s h i p s of opposition and economic exchange can be expressed independently of exogamy and exchange of women. I t may be argued, and i s probably true, that sub-clan potlatching functioned to define groups which were moving towards exogamic r e l a t i o n s h i p s and f u l l clan status. I t might also be argued, i n terms of Rosman and Rubel's model (1971: 26) that sub-clan potlatching "supports the d i s t i n c t i o n between ego's lineage, father's lineage, and spouse's lineage," i f i t could be shown that father's lineage and spouse's 71 lineage belonged to d i f f e r e n t sub-clans. There are not data s u f f i c i e n t to e s t a b l i s h this at present. It i s cl e a r , however, that sub-clan formation i s a co r r e l a t e of population s i z e . Of the houses (lineages and branch l i n e a g e s ) l i s t e d f o r the Coast Tsimshian, 128 are g*ispywudwa/dg, 119 g.anha'dg, 80 l a x s k * i 7 k , and 42 laxk'ibu'. Only among the largest clan, the g'ispgwudwa'dd, are sub-clans c l e a r l y emergent as potlatching groups, although they may have been forming among the g.anha'da. On the Nass, 49 houses are l a x k * i b u 7 , 48 lax s k ' i ' k, 22 g*ispgwudwa'da, and 20 g.anha''da. The two la r g e s t clans, the l a x s k * i ^ k and laxk*ibu', contained sub-clans. People s p e c i f i e d that the laxk* i b u ' sub-clans potlatched to each other; i t can be reasonably i n -ferred that the laxsk*i'k did too. I f the population size/sub-clan c o r r e l a t i o n i s a causal one, i t would suggest that the Tsimshian four-clan system developed from an e a r l i e r two-clan system as a r e s u l t of population growth. Chiefs and S o d a l i t i e s ; The roles of chiefs and t h e i r control over the h a l a ^ i t , or supernatural power r i t u a l s , were d i f f e r e n t i n the three d i v i s i o n s of the Tsimshian, and w i l l be discussed separately f o r each. Coast Tsimshian The Coast Tsimshian t r i b e s were characterized by a higher degree of s o c i a l i n t e g r a t i o n than was to be found elsewhere on the Northwest Coast. This was p r i n c i p a l l y due to a set of in t e g r a t i v e mechanisms which cut across the clan structure and were re l a t e d to the r i t u a l aspects of 72 the r o l e of t r i b a l c h i e f . There were no s u p r a - t r i b a l organizations or con-9 federacies of t r i b e s among the Tsimshian. The t r i b a l c h ief was the chief (headman) of the highest-ranking house i n the t r i b e , and a l l of the houses of a l l four clans were arranged i n a s i n g l e or continuous rank order under h i s . In most contexts, the t r i -b a l chief was c a l l e d sgmo'jig'Et (sgro: " r e a l " ; g'Et: "person"), as were other high-ranking chiefs of h i s and other upper status l e v e l houses. The chief had great prestige and influence i n h i s t r i b e and acted as i t s representative i n i n t e r - t r i b a l a f f a i r s . His family was expected to observe the most exemplary conduct and act as models of r i g h t behaviour f o r the people of the t r i b e . He was advised by a council of house c h i e f s , the lflkag'ig'Et, whose consent was necessary f o r important actions, such as h i s choice of successor. He had no p o l i c e force, but he could threaten supernatural re-t a l i a t i o n for non-participation i n dancing society o b l i g a t i o n s , and he con-t r o l l e d a group of a r t i s t s and stage managers, the g * i t > s ^ / n t k , which had some coercive powers i n h a l a ' i t a f f a i r s (see below). The t r i b a l chief ordered the annual movement to the Nass f o r eulachon f i s h i n g i n the spring, but seems to have had few other d i r e c t and i n s t i t u t i o n a l i z e d economic functions. legEx on the lower Skeena and sag.u >wE /n (Chief Mountain) on the Nass had middleman monopolies on trade 9. legEx, chief ot the g * i s p a x l j  y'ts, was generally acknowledged to be the leading or highest-ranking chief of the Lower Skeena River t r i b e s at Fort Simpson. However, this does not mean that he had any p a n - t r i b a l r o l e other than that which would normally accrue to one of such high p r e s t i g e . 73 with the Gitksan and the Tsetsaut, r e s p e c t i v e l y , but t h i s was not a regular function of t r i b a l c h i e f t a i n s h i p . Chief t s i b a s E f of g* i t x a ' i a was s a i d to have become wealthy from trading sea o t t e r and s e a l furs to white fur t r a -ders. The chief's main duties were r i t u a l i n nature (see G a r f i e l d , 1939: 182-184). He was expected to p o t l a t c h to chiefs of other t r i b e s , i n which he was supported by contributions from a l l of the houses i n h i s t r i b e , r e -gardless of h i s or t h e i r clans. He, i n turn, d i s t r i b u t e d food and wealth he received at potlatches to h i s t r i b e . G a r f i e l d (1939: 182) suggests that the chief had other (non-ritual) economic support from and obligations to the t r i b e , but i s not s p e c i f i c : While a chief can expect constant and l i b e r a l economic sup-port from h i s tribesmen, he does not contribute to potlatches given by them. He i s responsible for t h e i r economic welfare, must feed them when necessary and has to lay aside supplies f o r t h i s purpose. Narratives from g* i t x a ^ a report that chiefs received t r i b u t e i n the form of the f i r s t sea o t t e r and seal caught by each canoe of sea hunters and "other fur animals captured by hunters." S i m i l a r t r i b u t e may have been extracted by the chiefs of other t r i b e s . t s i b a s E y , a tyrant chief despised and feared by h i s t r i b e , was also said to have received wealth i n the form of ransoms from enslaving h i s own headmen. Unfortunately, for neither h i s r e d i s t r i b u t i o n of received pot-l a t c h goods, nor h i s r o l e as general provider for h i s t r i b e , are the data adequate to say to what extent the chief was " t r i b a l banker" or " c e n t r a l d i s t r i b u t i v e agent" i n Tsimshian society (see Sahlins, 1958: 3-5). I f one accepts Sahlin's thesis that "power, p r i v i l e g e , and prestige appear to be 74 generated p r i m a r i l y i n the process of goods d i s t r i b u t i o n " ( i b i d . : 3), one would expect to f i n d that considerable quantities of goods passed through the chief's hands. Perhaps the chief's strongest c o n t r o l over the t r i b e , and a source of considerable wealth, was h i s r o l e as w i h a l a . i t , or "great dancer." 11 When acting t h i s r o l e , he was addressed by a supernatural "power" name. The basic premise of the w i h a l a ^ i t r o l e was that the chief had greater supernatural power than others and could impart t h i s power to h i s people. Such great power was dangerous, and the chief was l i b e r a l l y compensated by xkF/t or "non-returnable" g i f t s for c o n t r o l l i n g i t to the benefit of his people. He was assisted and advised by a group c a l l e d the g * i t > s 3 "ntk ( g ' i t : "person"; son: " i n s e c l u s i o n " ) . This i s the group G a r f i e l d (1939: 304) described as "the p r o f e s s i o n a l group of a r t i s t s , song composers and organizers of the dramatizations [who] were a l l men who had received super-natural powers .... The a b i l i t y to carve, plan and operate novel mechani-c a l masks or other objects, or compose songs was considered a manifestation of the powers which the i n d i v i d u a l had received." According to one person (Heber C l i f t o n ) , they were a powerful group: "the g ' i t > s a / n t k were the 10. wi: "great"; h a l a / i t : "dancer." h a l a ' i t i s a very important word that cannot be e a s i l y translated, but which can be taken as a s i g n a l that supernatural beings or forces are involved. I t i s v a r i o u s l y translated as "dancer," "shaman," "dance," "power," "power dramatiza-t i o n , " and " i n i t i a t i o n . " I t i s also used as an adjective that can be roughly glossed as "sacred." 11. The l i s t of chief's supernatural power names i n Boas (1916: 513) i n -cludes some naxnp'x names, but seems also to include another category of names, those r e f e r r i n g to the heavens (laxha), which might have been s p e c i a l wihala. /it names. 75 song composers, the naxna /x makers, the makers of contrivances used by the i n i t i a t e s on t h e i r return from the sky. They were the advisors of the c h i e f s , h a l a f i t ' s , and were a most powerful group. Their influence was much greater than any other group i n the t r i b a l organization. They had powers of l i f e and death." Although t h e i r sources of wealth were not spe-c i f i e d , the g* i t * sa'ntk were paid handsomely for t h e i r services, n i . s l u ' t of g* i n a ? dp'iks was said to have been a wealthy and i n f l u e n t i a l man, p a r t l y because of h i s p o s i t i o n as a leading a r t i s t of the g* i t ; sj/ntk. He was of the common or c o u n c i l l o r (lflkag*ig'Et) status l e v e l , although there are no statements as to whether a l l of the g ' i ^ s j ' n t k were of t h i s status l e v e l . They probably were. Barbeau (1950: 780-790) describes i n some d e t a i l the r o l e of the g*it* 1 s j 7 n t k among the g * i t g . a ' a t l of the Southern Tsimshian. When young, they were selected from c e r t a i n families and s p e c i a l l y trained f o r t h e i r r o l e . They were " c o n t r o l l e d by various secret s o c i e t i e s with which they associated" as w e l l as being "employed by most of the c h i e f s " and h i r e d out to other t r i b e s ( i b i d . : 790. Barbeau's use of the word "em-ployed" i s probably misleading here. I t i s more probable that they were attached to the chief's household. I n t e r e s t i n g l y , I was t o l d by Norman T a i t , a Niska carver from K i n c o l i t h , that h i s father had s a i d i t was i l l -advised f o r a carver to marry since, i n the o l d days, he would have been continually moving from place to place. There were also said to be g* i t ' s a ''ntk among the B e l l a B e l l a , from whom the Tsimshian were said to have borrowed dancing s o c i e t i e s , strengthening the association between t h i s group and the dancing s o c i e t i e s headed by the c h i e f s . 76 Most s i g n i f i c a n t l y , the g ' i t ' s o 'ntk were c l e a r l y distinguished from the carvers of totem poles and crest a r t , who were c a l l e d the ukgihl9 and were not permitted to carve masks and halafit paraphernalia. They did not work i n secret as did the g ' i t ' s j ' n t k , and t h e i r status was lower ( i b i d . : 790). The g * i t , s ? / n t k ' s "power over l i f e and death" came from t h e i r r i g h t to k i l l or to force to j o i n a dancing s o c i e t y , any n o n - i n i t i a t e who witnessed t h e i r making or operating of h a l a ^ i t contrivances. This could be used to t h e i r advantage, of course, by f o r c i n g r e c a l c i t r a n t r e c r u i t s to buy t h e i r services for an i n i t i a t i o n . There seems to be a paradox here. On the one hand, the great secrecy of the g*it^sa'ntk's operations, and the threat of death to any n o n - i n i t i a t e intruding upon them or witnessing t h e i r malfunction, sug-gests that the supernatural power displays were believed indeed to be the r e s u l t of non-human actions or supernatural intervention i n human a f f a i r s . On the other hand, a l l members of the t r i b e were expected to be i n i t i a t e d into e i t h e r the Dog Eaters or Dancers dancing s o c i e t i e s , and hence to have p a r t i c i p a t e d In the dramatizations by which super-natural events, such as ascent into the heavens, were simulated. The paradox, then, i s that people can b e l i e v e to be true those same events which they knowingly simulate i n order to "deceive" others. This i s the same paradox Levi-Strauss investigates i n an ar-t i c l e on shamanism (1967b) , where he r e f e r s to a Koskimo shaman (des-cribed by Boas [1930, I I : 1-41]) who went mad i n h i s attempts to resolve i t . The shaman, who himself used t r i c k e r y i n h i s p r a c t i c e , t r i e d to get 77 another shaman, who has bested him and subjected him to ridicule, to admit that his healing performances were also trickery. When the sec-ond shaman refused to speak, the f i r s t shaman, unable to face his possible genuineness, went mad and shortly died (Levi-Strauss, 1967b: 169-176). Levi-Strauss sees the paradox related to the participation of the audience in the curing drama. As "actors" the audience can participate in emotional states and beliefs which they would reject in everyday l i f e . He also seems to be suggesting that the shamanistic performance offers the audience a chance to participate in some form of anti-structural exploration: " i t i s this v i t a l experience of a uni-verse of symbolic effusions which the patient, because he i s i l l , and 12 the sorcerer because he i s neurotic — in other words, both having types of experience which cannot otherwise be integrated — allow the public to glimpse as 'fireworks' from a safe distance" (ibid.: 176). It should be noted, in this connection, that the hala'it per-formance and idiom of expression are derived from shamanism. One mean-ing of the word halafit i s "shaman," and there are a number of corres-pondences in the use of cedar bark, red ochre, bearskin robes, songs, etc., used by the shaman and the dancing society i n i t i a t e . naxn?/x or s p i r i t name dramatizations are also called halafit's and include cer-tain performances in which chiefs ex p l i c i t l y act the role of swansk  hala'.it or "curing shaman." 12. Levi-Strauss acknowledged in a footnote written later (1967b: 180, footnote #19) that his use. of "neurotic" here was an oversimplifi-cation. 78 The h a l a f i t season, c a l l e d gwEndgsgm h a l a ' i t ( " a r r i v a l [on earth] h a l a ' i t " ) was o f f i c i a l l y declared open at a po t l a t c h by ni.swE'xs of g'inado ' i k s , whose p r i v i l e g e i t was, and observed by a l l of the Lower Skeena River Tsimshian at the same time ( d i f f e r e n t chiefs opened the season f o r the Southern Tsimshian). The season ran f o r the e n t i r e win-ter or time of tabu (haHrE^iks: "tabu") . Quite often i t was not com-pleted by the time the people moved to the Nass i n the early s p r i n g , and novices completed t h e i r i n i t i a t i o n s there. The ceremonies were planned by a council of a l l hala<it chiefs of the t r i b e , known c o l l e c t i v e l y as wutahala'it ( p i . of wihala.'it) , or "great dancers." "The council was held i n the greatest secrecy, i n a house completely surrounded by cedar bark rings(lu*'ix) , a warning to a l l that i t meant death to enter. naxno x whistles of a l l the secret s o c i e t i e s were sounded con t i n u a l l y while the council was going on. A l l the people stayed i n t h e i r own houses, such was the fear of the wutahalafit" (Joseph S t a r r ) . While people did not say so s p e c i f i c a l l y , the council of the wutalialafit must have been s i m i l a r i n composition to the council of headmen (lgkag'ig'Et ) , which advised the chief on secular matters. A l l members of the t r i b e p a r t i c i p a t e d i n the h a l a f i t system, which was described as a s e r i e s of "steps" or elevations i n supernatural power by means of i n i t i a t i o n ceremonies. The f i r s t step i n h a l a ' i t elevation was a "throwing ceremony," t s i . k (the word f o r dentalium) or samhala'it (ssm; " r e a l " ; h a l a ^ i t : "dance1') , which a l l c h i l d r e n were given and which could be sponsored by 79 13 either the father or the mother's brother. The English name of the ceremony refers to the chief's wrestling with and "throwing" his power Into the children, who were hidden under new cedarbark mats supplied by their father's sisters, who acted as their assistants. After this ceremony, the child was known as amg'Et (am: "good"; g'Et: "person") 14 and considered ready for initiation into a dancing society. The second step in supernatural elevation was to join either the Dog Eaters (nulim) or Dancers (miia) dancing societies (Boas' "sec-ret societies"). Everyone was expected to join one or the other society. If they didn't, they would be told by the wutahala'it that they would be ll'd n (killed at an early age by the powers of the hala'it). Pro-bably the only adult Tsimshian who were not dancing society members were those of the wa'ayin category (see below, under Status Levels). Membership in the dancing societies had the effect of divid-ing Coast Tsimshian tribes into two groups, cross-cutting clan member-13. There is some confusion about the name of the first power ceremony. Garfield (1939: 198) says that a second power ceremony was called sgmhala/it, which name Drucker (1940: 221) and Boas (1895: 575) give to the first power ceremony. The throwing ceremony of the Gitksan is called su-ha-lide (Duff, 1959: 40), which is perhaps related to sgmhala^it. A resolution of the problem might l i e in Drucker's equation of the sgmhalafit with the koxumhalait (g.?/ gom "nodding" hala.'it) or "head-shaking" dance (ibid.: 222) , which he also calls a "crest-display" dance. This is the chief's welcoming dance at potlatches, during which he shakes eagle down, a symbol of peace, from his headdress over the guests. The chief's attire and actions in the throwing ceremony and welcome dance are very similar (see Boas, 1916: 515 and 539-540), suggesting that an earlier sgmhala .'it became part of both the potlatch and hala". i t system. 14. Some people said that amg'Et was a second step after the throwing ceremony, but there are no recorded descriptions of a ceremony intermediate between i t and the dancing society initiations, and other informants said specifically that the word amg'Et referred to the status or condition of a person ready to enter a dancing society. 80 ship. The two groups were said to be about equal in membership. "The initiates enter either the society of father or maternal uncle, who-ever actually shouldered the whole expense, or that of the other, i f they both agreed, as i t was always to the advantage of the house to divide membership, but in no way had the clan or phraytral (sic) re-lationship anything to do with i t " (Beynon, field notes). This second step was called "ascending into the heavens" (hllaxE /), referring to the initiate's supposed ascent into the heavens while under possession by the spirits. The tribal chief was the leader or wihala'it of one of the dancing societies, and another high-ranking chief was wihala.'it of the other. Thus, for example, legEx was wihalafit for the Dog Eaters at g*ispaxl? / 3ts, and ni.swa?amak, another laxsk'i^k chief, was wihala.'it for the Dancers. (At g'ina'da'iks, the laxk*ibu' chiefly house was ex-tinct, and ni.swExs became wihalaCit for both dancing societies. This was regarded as an exception to the rule.) The general sequence of initiation, which took place in a special hala'it house set aside for each society, was the transfer of the spirit of the wihala^it into the initiate, for which the chief was compensated, and the disappearance of the initiate, who was now completely in the power or possession of the spirit. The initiate, who was sup-posed to have disappeared into the heavens, was secreted behind the par-tition (p 7toi) at the rear of the house. At this time, gifts were distributed to the members of the dancing society, who sang and witnessed the disappearance. After a time, the initiate returned, riding offshore 81 In a contrivance built to resemble his crest animal, and the power was removed from him by the wihala'it, who was again compensated. It took him several ritual procedures to restore the initiate to a normal state. Garfield (1939: 303) reports that "a f u l l fledged member of the Dog Eaters had to demonstrate contact with the spirit on two dif-ferent occasions, each accompanied by a distribution of wealth to the 15 guests invited to witness the event. Four such potlatches during a lifetime was reported as necessary before a person could retire with honors from the society. Older men who had fulfilled a l l the obliga-tions of feasts and property distribution could step out of the secret society and give their positions to younger persons. At the fourth ceremony the giver announced his intention of retiring and at the same time initiated a younger person to take his place." There are no con-firmations of retirement, transfer of "places," or the obligation to give four hala.'it feasts in the Barbeau/Beynon notes. However, there is confirmation in museum specimens and their accompanying documenta-tion that initiates gave up to four hala'it feasts or demonstrations. For example, there is a four-row dancing society headdress of red cedar-bark, collected by Barbeau from sqat* i/n, laxk'ibu' head-chief of g*itlaxdaImks, in the ROM whose "rows indicate that the owner had danced four times" (ROM, HN-765, Barbeau catalogue notes). We know from the 15. While hala/it's involved property distributions, i t is probably inaccurate to call them potlatches. See the discussion of pot-latches in Chapter Four. 82 notes on another specimen, a cedarbark c o l l a r (ROM,HN-765) , that sqat^i.'n was a member of the Dog Eaters, and i t can therefore be i n f e r r e d that t h i s was a headdress of the Dog Eater society. A three-row head-dress of the same type was c o l l e c t e d by Emmons, also from Gitlaxdamiks, and i s now i n the MAI (1/4209). Although he does not say from whom he co l l e c t e d i t , we can assume i t was someone who had "danced" i n the Dog Eaters three times. Both headdresses are i l l u s t r a t e d below. Plate 1. sqat'i.'n's four-row cedarbark headdress. ROM, HN-765. Collected by CM. Barbeau, Gitlaxdamiks, 1927. Probably Dog Eaters dancing society. 83 Plate 2. Three-row cedarbark headdress from Gitlaxdamiks. MAI, 1/4209. Collected ca. 1907 by G.T. Emmons. Probably Dog Eaters dancing society. At the end of the h a l a ' i t season, the w i h a l a / i t p u r i f i e d the h a l a ' i t house, which was once again used as a normal dwelling, and the t r i b e began i t s spring and summer food-gathering a c t i v i t i e s . Certain chiefs had the p r i v i l e g e of being i n i t i a t e d by other supernatural powers. These were the Cannibal s p i r i t (xg'Edt: x-"to partake or consume"; g'Et: "person"; also c a l l e d u * l a l a , from the Kwakiutl o^lala) and the Destroyer s p i r i t ( l u d z i s t > E / or wi^nanal; i n Niska, ho.nanai; from the Kwakiutl hawijnalal). These were not socie-t i e s , as among the Kwakiutl, but personal c h i e f l y p r i v i l e g e s , acquired d i r e c t l y from the northern Kwakiutl. There i s , however, no clear i n -d i c a t i o n of how they were acquired. Boas (1916: 510) says that they were acquired "through intermarriage," but does not elaborate, even though there are no provisions i n Tsimshian marriage rules f o r the 84 t r a n s f e r of p r i v i l e g e s between a f f i n e s . A Tsimshian version of the Kitimat myth of the o r i g i n of dancing s o c i e t i e s casts t h e i r a c q u i s i -t i o n i n t o t r a d i t i o n a l Tsimshian terms. Of the four young men who f i r s t saw the dances i n supernatural dwellings that rose out of a lake near Kitamat, two were chief's nephews, and two were of the c o u n c i l l o r (l^kag'ig'Et) status l e v e l . The f i r s t two gave the Cannibal and Des-troyer dances as g i f t s to t h e i r uncles, and these became the exclusive property of c h i e f s . Since the other two young men were not of the c h i e f l y status l e v e l , the Dog Eaters and Dancers were open to everyone (see one version of the myth i n G a r f i e l d , 1939: 293-294). One of Beynon's Tsimshian teachers said that sgagwe't, chief of the g ' i t ^ a n d ^ t r i b e , was i n i t i a t e d i n t o the Cannibals, contribu-tions came from the t r i b e as a whole, not j u s t from h i s r e l a t i v e s . G a r f i e l d (1939: 296) reports that i n i t i a t i o n s i n t o the Destroyers were financed by a l l of the members of the Dancers of a chief's t r i b e . These data suggest that c h i e f l y i n i t i a t i o n s into the Cannibals and Destroyers p a r a l l e l e d chief's potlatches as a f f a i r s supported by the e n t i r e t r i b e . The chief's two r o l e s , as s^mo^a ig' E t and w i h a l a ' i t are thus separable, and correspond to two s t r u c t u r a l orders of Coast Tsimshian s o c i e t y : one r i t u a l l y expressed i n h a l a ' i t symbolism; the other, the structure of descent groups and a f f i n a l t i e s , r i t u a l l y expressed i n the potlatch. Rank permeates both, but i s l e s s e x p l i c i t i n the h a l a ' i t order. Both parents contribute to the c h i l d ' s advancement i n both orders, but the father has a more d i r e c t r o l e i n sponsoring the c h i l d ' s h a l a ' i t 85 advancement: "a father may finance the e n t i r e hala.'it process himself. I t has not the same association of the ^jix (potlatch) f e a s t , wherein the whole r e s p o n s i b i l i t y i s maternal and the paternal side only as-s i s t s " (Gray and Johnson). Advancement i n both orders was necessary to properly assume one's place i n Tsimshian society. There are some in d i c a t i o n s that par-t i c i p a t i o n i n the h a l a f i t was beginning to overshadow potlatch p a r t i c i -pation: " I t was of more importance i n the s o c i a l standard to have been h a l a ' i t , than r e a l l y to have assumed a name" (Beynon, f i e l d notes). This i s e s p e c i a l l y i n t e r e s t i n g i n l i g h t of the general assumption, made by both anthropologists and Tsimshian, that the dancing s o c i e t i e s were recent borrowings of the Tsimshian from the B e l l a B e l l a . Boas dated the spread of the Cannibal dance to the Tsimshian and southern Kwakiutl as follows (1897: 664): The Kwakiutl state uniformly that the custom of devouring men was introduced among t h e i r t r i b e about s i x t y years ago [ca. 1835], and that i t was derived from the Heiltsuq. We also have conclusive evidence that the custom was acquired by the Tsimshian not more than seventy years ago [ca. 1825] and that they also obtained i t from the He i l t s u q (emphasis added). Unfortunately, Boas does not go on to say what t h i s "conclusive evidence" i s , and I cannot f i n d reference to i t i n any of h i s other writings on the Tsimshian. Given h i s famous caution i n drawing conclusions of any kind, one wants to be l i e v e that i t i s a well-founded statement, e s p e c i a l l y since there i s no d i r e c t evidence from other sources as to when the Tsim-16 shian acquired the dancing s o c i e t i e s . Boas was, of course, only r e f e r -16. E t h n o h i s t o r i c a l research may shed l i g h t on t h i s , although unfortu-nately there was l i t t l e contact by Europeans with the Tsimshian i n the early h i s t o r i c a l period. 86 r i n g to the adoption of the Cannibal dance, and i t i s quite possible that the other dancing s o c i e t i e s were acquired at a d i f f e r e n t time. Ex-cept f o r the myth of t h e i r o r i g i n at Kitamat r e f e r r e d to above, dancing s o c i e t i e s are not mentioned i n Coast Tsimshian myths, which would support t h e i r recency, and the use of Kwakiutl names has already been established. I t i s tempting to explore the implications of Boas' date of 1825 for the introduction of the Cannibal Dance to the Tsimshian (posi-t i n g , f o r the moment, that the Dog Eaters and Dancers were adopted i n the same period). Although I do not agree with Service (1962) that Northwest Coast s o c i e t i e s should be categorized as f u l l y developed chiefdoms as he defines them, h i s discussion of the changing nature of r e l i g i o u s s o d a l i t i e s as mechanisms of s o c i a l i n t e g r a t i o n at the c h i e f -dom l e v e l of s o c i o p o l i t i c a l development i s appropriate here: Rel i g i o n i n chiefdoms i s markedly d i f f e r e n t from that of t r i b e s and bands. I t i s not so much that the antecedent r e l i -gion i s a l t e r e d , but rather that i t i s augmented i n content and with new forms superimposed. The shamanistic practices and l o c a l l i f e - c y c l e r i t u a l s remain, but ceremonies and r i t u -a l s serving wider s o c i a l purposes become more numerous .... Related to a l l t h i s are new kinds of r e l i g i o u s p r a c t i -tioners who may be s a i d to form a priesthood. Whereas a sha-man achieves h i s p o s i t i o n by personal q u a l i f i c a t i o n s , a p r i e s t occupies a permanent o f f i c e i n the society. The differences between these two resembles the differences between the occa-s i o n a l leader of t r i b a l s o c i e t y and the true chief. C h i e f t a i n -ship and priesthood, i n f a c t , seem to a r i s e together as twin forms of authority, d i s t i n c t with respect to the contexts i n which the authority i s wielded but otherwise s i m i l a r , i f not i d e n t i c a l . O r d i n a r i l y p r i e s t l y o f f i c e s descend i n the same family l i n e as the secular o f f i c e s ; further, sometimes the p r i e s t and the chief are the same person (Service, 1962: 171). With t h i s model i n mind, one can view the dancing s o c i e t i e s as new r i t -uals "serving wider s o c i a l purposes," notably to cut across kinship t i e s ; 87 the chief i n h i s r o l e of w i h a l a f i t as p r i e s t ; and the g* i t * s ^'ntk as an i n c i p i e n t priesthood. Service (1962: 145) points out that "no one has observed the actual o r i g i n of a chiefdom," but perhaps i n the Tsimshian case we can see glimpses of the dynamics of c h i e f l y consolidation of secular and r e l i g i o u s authority i n a r e d i s t r i b u t i v e society. S i g n i f i -cantly, t h i s occurred, i f we accept Boas' date, a f t e r the beginnings of the fur trade (which began i n force a f t e r 1785) and about the same time that Fort Simpson was established, which suggests that the impetus toward c h i e f l y consolidation of power lay i n changes i n i t i a t e d by the i n t r u s i o n of the whiteman and r e s u l t i n g a l t e r a t i o n s i n a b o r i g i n a l s o c i a l r e l a t i o n s h i p s and sources of wealth. Niska: There are no ethnographic descriptions of the r o l e and functions of Nass River chiefs i n the published l i t e r a t u r e , and there are only scattered and occasional references to them i n the Barbeau/Beynon notes, which, unfortunately,are not s u f f i c i e n t to draw a c h i e f l y p o r t r a i t comparable to the preceding one f o r the Coast Tsimshian. I t i s c l e a r , however, that t h e i r o f f i c e i n s o c i e t y , t h e i r hegemony, was that of clan and sub-clan leaders within a t r i b e , and that they had no i n s t i t u t i o n a l i z e d o f f i c e of t r i b a l c h i e f , as did the Coast Tsimshian. The progression of c h i e f l y authority seems to be that the chief of the highest ranking sub-clan i n a clan was the head-chief of the clan, and that the head-chief of the l a r g e s t clan i n a t r i b e or winter v i l l a g e had prestige and rank comparable to a Coast Tsimshian t r i b a l c h i e f , although he did not hold t r i b a l " o f f i c e " as 88 17 such. Nor were a l l the houses of a l l the clans i n a t r i b e arranged i n a s i n g l e rank order, as among the Coast Tsimshian. Niska society was organized according to a clan p r i n c i p l e , with greater development and d i f f e r e n t i a t i o n of sub^-clans within a clan than was to be found among e i t h e r the Coast Tsimshian or the Gitksan. The differences i n the Coast Tsimshian and Niska rank and authority structures are r e f l e c t e d i n the ranked l i s t s of houses and clans i n Appendix I. Whereas Coast Tsimshian people l i s t e d houses of a l l four clans i n a si n g l e rank order under the t r i b a l c h i e f , Niska people l i s t e d houses i n rank order within clans or sub-clans, which were also ranked r e l a t i v e to each other. Non-clan or extra-clan leadership was developing on the Nass, however. At g* i t l a x d a finks , f o r example, the g* i s g . a.'st sub-clan under p i * 1 recognized the laxk* i b u ' chief sqat>' i."n as t h e i r c h i e f , and so were c l a s s i f i e d i n the ranked l i s t s under laxk* ibuf In other words, houses of one clan recognized the leadership of a chief belonging to another clan. There was also a tendency f o r head-chiefs of clans i n a t r i b e to claim f i r s t - r a n k i n g p o s i t i o n of t h e i r clan f o r the e n t i r e Nass ( s q a t ; i / n , f o r example, was one of those s a i d to be highest ranking laxk*ibu' on the Nass), and at l e a s t one ch i e f , kwaxcu, attempted to be recognized as "Number One Chief Nass River," although h i s claim was rejected. 17. There were two head-chiefs at g*itlaxdafmks: sqat* i / n of the, l a x k * i b u 7 and mfrnE'sk of the laxsk*i^k. In a potlatch order c o l l e c t e d by Beynon from Peter n i . s y a q l , sqat*i.'n ranked ahead of m>nE'sk, which accords with the general front-ranking p o s i t i o n of the laxk* ibu' over the other clans on the Nass. 89 Sapir (1915: 28) reports three dancing s o c i e t i e s as " i n h e r i -table" p r i v i l e g e s : Cannibals, Dog Eaters, and Destroyers; Boas (1897: 651) reports " s i x " : the sgmhala/it, Dancers, Dog Eaters, Cannibals, nanesta /t, and Destroyers, i n rank order from lowest to highest. He says that the Niska nanesta"t i s equivalent to the Kwakiutl n o n t s i s t a / l a L ( i b i d . : 652) which he l a t e r ( i b i d . : 654) says i s also equivalent to the Destroyers (Niska honan^aL), leaving us to deduce that nanesta yt and honan'aL are two d i f f e r e n t names f o r the Destroyer society among the Niska, and that he i s , i n f a c t , reporting f i v e s o c i e t i e s rather than " s i x . " In contrast to published information l i k e that i n the pre-ceding paragraph, museum specimens and accompanying documentation can be used to decipher the dancing society complex on the Nass with some p r e c i s i o n , e s p e c i a l l y with reference to the bet t e r known complex of the Coast Tsimshian. Accompanying a trumpet of the Dog Eaters society (ROM, HN-695) are Barbeau's notes that i t was "used when i n i t i a t i n g a boy or g i r l i n t o the lu-kim (Dog Eaters) ," t e l l i n g us that ch i l d r e n were i n i t i a t e d into these s o c i e t i e s as on the Coast. Quite a number of other pieces from the Nass, c o l l e c t e d by Barbeau f o r the ROM, are i d e n t i f i e d as being used i n the Dog Eaters and Dancers s o c i e t i e s , confirming t h e i r presence 18 on the Nass. There are also objects used i n the Cannibal dance c o l -18. Both of the Newcombes and Emmons c o l l e c t e d more objects from the Nass River than did Barbeau, but they neglected to record the names of the dancing s o c i e t i e s f o r the dancing society parapher-n a l i a they c o l l e c t e d so that, as usual, t h e i r documentation i s i n s u f f i c i e n t f o r ethnographic purposes. 90 l e c t e d from sqat'i.n (e.g., ROM, HN-771) and Dog Eater paraphernalia 19 c o l l e c t e d from both sqat'i.'n and mgnE 'sk (e.g. , ROM, HN-775 and HN-695) . Catalogue notes f o r a cedarbark c o l l a r (ROM, HN-798) report that sqat > i.'n was "manhala/it, the chief of the hala.'it," which most probably corres-ponds to the w i h a l a / i t r o l e of the Coast Tsimshian c h i e f s . F i n a l l y , a trumpet c o l l e c t e d from sqat* i.'n (ROM, HN-708) was described by Barbeau as follows: " i t was always kept secret, and was made by the sig * i d z y n : behind the scene, that i s , the people behind the scene." In other words, i t was made by the Nass River equivalent of the g ' i t * za'ntk. No specimens s a i d to belong to the Destroyers were encoun-tered. This does not nec e s s a r i l y contradict Sapir and Boas, but does suggest that membership i n that soci e t y was not as widespread as membership i n the Dog Eaters, Dancers, and Cannibals. The general p i c t u r e of dancing s o c i e t i e s which emerges from these fragmented data i s s i m i l a r to that of the Coast Tsimshian with at l e a s t one important d i f f e r e n c e : both sq at* i.'n and manEJsk, head-chiefs of g* itlaxda^mks, were members of the Dog Eaters. Among the Coast Tsim-shian, the two highest ranking chiefs i n a t r i b e would have belonged to d i f f e r e n t s o c i e t i e s of th i s grade. I n t e r e s t i n g l y enough, the myth of o r i g i n f o r the Niska dancing s o c i e t i e s i s d i f f e r e n t from that of the Coast Tsimshian, although i t i s also l o c a l i z e d at B e l l a B e l l a and Kitamat (Boas, 1897: 651-653). I t men-tions only the three s o c i e t i e s confirmed by the museum specimens: the 19. Cf., the two Nass River Dog Eater headdresses i l l u s t r a t e d on pages 82 and 83. 91 Dog Eaters, Dancers, and Cannibals, which were sa i d to have been given by a chief l i v i n g i n a cave to a hunter who had been led there by a bear. Gitksan: Gitksan chiefs were house c h i e f s , and the chief of the highest-ranking house i n a clan was clan chief for h i s t r i b e . There are no i n d i c a t i o n s that larger hegemonies were forming. Unlike the Niska, whose chiefs vied for s u p e r i o r i t y over one another, the Gitksan chiefs maintained an e g a l i t a r i a n posture. When the Kitsegukla were i n v i t e d to a p o t l a t c h , for example, "they would c a l l a chief from each clan i n turn, otherwise a difference i n standing would be implied, and one group would be offended" (Charles Mark). S i m i l a r l y , the arrangement of dwellings at both Kitwancool and Kitsegukla placed the two clan chiefs — the l a x k * i b u ' and l a x s e / l at Kitwancool and the g.anha'da and g*isg.a.st at Kitsegukla — side by side i n the middle of the v i l l a g e , with the other houses of t h e i r clans ranged out on e i t h e r s i d e , i d e a l l y i n descending rank order (the Kitwancool g'isg.a/st were newcomers and were not considered a clan equal to the other two). In recent years, presumably because neither one of the clan c h i e f s was a t r i b a l c h i e f , the people of Kitwancool have established an o f f i c e of v i l l a g e president, whose incumbent i s not a chief but who has the authority to deal with outside i n t e r e s t s and a u t h o r i t i e s for the t r i b e as a whole (see Duff, 1959: 12, 36). In 1938, " a l l the chiefs and the young men held a meeting i n Kitwancool. They created a pact or law of agreement among the chiefs of the v i l l a g e , and t h i s then formed a union between the Frog and Wolf clans. In other words, these clans united as 92 one, and under this agreement they swear to protect a l l the lands and natural resources belonging to the people of Kitwancool" (people of Kitwancool, i n Duff, 1959: 37). These two events, couched i n the thor-oughly Eurocanadian terms of "president" and "pact or law of agreement," permit the Kitwancool to present a united face to the outside world; there were no t r a d i t i o n a l Gitksan mechanisms they could have invoked to do the same. According to Drucker (1940: 222), whose b r i e f account i s the only one i n the l i t e r a t u r e , there were two dancing s o c i e t i e s among the Gitksan: the Dog Eaters (g. alu-kim) and Dancers (g.amila), although there were c h i e f l y r i g h t s to the Cannibal dance at Kitsegukla, Kitwan-cool , and perhaps Kitwanga and Kispayaks, and to the Destroyers (winalal, which Drucker c a l l s a "War Dance") at places not s p e c i f i e d . As among the Niska, the chief dancer (presumably of each society) i s c a l l e d mgnhala/it ( l o c . c i t . ) . The view of the Gitksan dancing s o c i e t i e s from the neighbouring C a r r i e r permits us to glimpse some of the dynamics of membership. Jen-ness (1943: 571) t e l l s of a r i c h but "low-born youth of Kispiox," who de-termined to enter the Cannibals ( w i l a l a ). "He made h i s ambition known, but the members of the w i l a l a and k a l u l l i m s o c i e t i e s , at a j o i n t meet-ing, decided that h i s low b i r t h debarred him from the w i l a l a s o c i e t y , but allowed him to enter the i n f e r i o r k a l u l l i m . " The l i n k i n g of the Cannibals and Dog Eaters implied above by t h e i r j o i n t meeting i s further s i g n i f i e d by Barbeau's catalogue documen-t a t i o n f or a b e a u t i f u l l y painted wooden screen from the house of wudaxhayEts, l a x s e f l , Kitwancool (NMC, VII-C-1130; see P l a t e 69). The 93 screen was named pt>lamwil E.q ( " p a r t i t i o n of dragon f l y " ) and showed a large dragon f l y i n the middle, which the "owner ...had not adopted ... as a c r e s t , but as a decoration," and which was flanked by crest figures named Half Exposed Person. The p a r t i t i o n was used by the g.alu-fcim hala'Tit f o r two occasions, f i r s t f o r the g.alulim and second 20 f o r the u l a l a . This suggests that those two s o c i e t i e s perhaps held j o i n t i n i t i a t i o n s (at l e a s t they used the same h a l a ^ i t dwelling), or were otherwise lin k e d i n native thought. R i t u a l dog eating can perhaps be viewed as a weaker or less intense form of cannibalism. Perhaps i t was only the Dog Eater chiefs who were allowed to become Cannibals. A s i m i l a r association f or the Dancers and Destroyers i s sug-gested by another museum specimen. I t i s a cedar club (NMC, VII-C-1073) c o l l e c t e d by Barbeau from the house of l u . l o g , l a x s e f l , Kitwanga, which was sa i d to be " f o r g.amita society o u t f i t " and used as follows: "the young halaCit brought t h i s club with him i n the feast house, h i t the door posts with i t before entering, and when he saw a good thing i n the house he h i t i t , and paid for i t afterwards." Thus, while the club belonged to a Dancers o u t f i t , i t s use was unquestionably that of the Destroyers society (see Boas, 1916: 551-553). Other museum specimens shed additional l i g h t on the dancing so c i e t y complex. At the ROM i s a whistle from Kispiox (HN-750) of which Barbeau says: "these whistles were always kept hidden. I f a man i by chance happened to see a whistle or anybody blowing a whistle he was 20. The complete documentation for t h i s specimen i s given on pages 246 and 247. 94 k i l l e d and no indemnity was paid. I t was acknowledged l e g a l to k i l l him. I f he had s u f f i c i e n t means he could buy himself out. His l i f e was then spared and he became a member of the secret society." Although there i s no mention of a g * i t > s ? 'ntk organization among the Gitksan, t h i s custom implies that the same sort of secrecy and control they maintained fo r the dancing s o c i e t i e s of the Coast Tsimshian was i n force here too. There are catalogue notes f o r a number of cedar bark head rings which t e l l us that dancing society members were expected to stage a number of power demonstrations during t h e i r careers, as among the Coast Tsimshian and Niska. For example, a Cannibal head r i n g belonging to nEqt, l a x s e f l , Kispiox, has three rings (NMC, VII-C-1173): "one r i n g was added to represent each time the hala.'it wasgLven; and neqt has given three" (Barbeau catalogue notes). The owner of a Cannibal whistle, Richard Morrison of Kispiox, t o l d Barbeau that he thought i t (ROM, HN-752) was made about 1866, at the time when the Great Western Union Telegraph l i n e was constructed. He added that the l a s t Cannibal r i t u a l was performed at Kispiox about 1892. F i n a l l y , to conclude t h i s s e c t i o n , the differences i n phrasing or i n emphasis regarding the Cannibals and Dog Eaters among the Gitksan and the C a r r i e r , who borrowed the r i t u a l s from them, are i n s t r u c t i v e . A C a r r i e r looked upon the onset of the s p i r i t as a supernatural possession or sickness, c a l l e d the kyan sickness, which he would gladly avoid; i t was p a i n f u l and dangerous, and could only be cured by exorcism by the kyanyuantan (society members), a f t e r which he joined t h e i r ranks. The Gitksan, however, regarded membership i n the s o c i e t i e s as highly desirable 95 and, as i n the case of the low-born Kispiox man referred to above, ac-t i v e l y sought out possession by the s p i r i t s and what appears to be sim-u l a t i o n of the "sickness": To the one people the society i s p r i m a r i l y a group of medi-cine men joined together to treat a p e c u l i a r and dangerous disease; to the other i t i s an organization for conferring prestige and influence on a l i m i t e d s ection of the community by means of a spectacular i n i t i a t i o n r i t e that invokes the sanction of the supernatural (Jenness, 1943: 571). Rank The Northwest Coast i n s t i t u t i o n of ranking i s apparently a d i f f i c u l t one for ethnologists to understand, and much has been wr i t t e n about i t , e s p e c i a l l y i n i t s r e l a t i o n s h i p to s o c i a l classes. In Appendix I, houses or lineages of a l l three divisions of the Tsimshian are l i s t e d i n rank order, as these were recorded from Indian t e l l e r s . The nature of the rank orders d i f f e r s : houses ranked by t r i b e (Coast Tsimshian), houses ranked by sub-clan or clan (Niska), and houses ranked by clan (Gitksan). Further, clans i n a t r i b e are ranked for the Gitksan, and sub-clans within a clan and clans within a t r i b e are ranked for the Niska. Although not included here, l i s t s of potlatch rankings were also c o l l e c t e d i n which chiefs (respresenting houses) from various t r i b e s were ranked. F i n a l l y , within a house, names are considered to be "high-" or "low-ranking." Ranking was obviously an important way of c l a s s i f y i n g i n d i v i -duals and s o c i a l groups to the Tsimshian. But i t was not an absolute determined by b i r t h , some kind of f i x e d "Chain of Being," as i s sometimes implied. Rank was an ideology, a l i n e a r pattern used by the Tsimshian to determine p r i o r i t i e s i n d i f f e r e n t 96 s o c i a l contexts. Conceivably, each concrete ranking one might observe might d i f f e r from every other, depending upon the context, the population units involved, and the p r i v i l e g e s being c a l l e d f o r t h f o r the event. This has been best expressed by Sapir (1967: 33): The ranking orders thus a r r i v e d at by seating, d i s t r i b u t i o n of g i f t s , i n v i t a t i o n s to f e a s t s , and i n various other ways ... might be expected to coincide. To a c e r t a i n extent they do tend to approximate, and the highest rank i n a community w i l l nearly always be found to head any such l i s t that might be constructed. In p r a c t i c e , however, one finds that the various orders do not nec e s s a r i l y s t r i c t l y correspond, i n other words, that a person might i n d i v i d u a l l y be of l e s s e r rank than another from the point of view of seating, but would have a p r i o r claim to be i n v i t e d , say. This curious state of a f f a i r s shows c l e a r l y enough that at l a s t analysis rank i s not a permanent status, which i s expressed i n a num-ber of absolutely f i x e d ways, but i s rather the resultant standing attained by the inheritance of a considerable number of t h e o r e t i c a l l y independent p r i v i l e g e s which do, indeed, tend i n most cases to be associated i n c e r t a i n ways, but may nevertheless be independently transmitted from generation to generation. Any ranked l i s t s , then, which were volunteered by Tsimshian people or wri t t e n down by ethnographers from t h e i r own observations of seating, g i f t s i z e , etc., are "true" f o r that moment or event only. My own con-cept of rank i s of a motion p i c t u r e of houses slowly r i s i n g and f a l l i n g along multiple v e r t i c a l axes. At any point i n time, say a potlatch, the operator of the camera might stop or freeze the action, permitting the anthropologist to make a l i s t of what he sees. But then the movie con-tinues, and i t would be a mistake f o r the observer to pack up h i s notebook and leave without n o t i c i n g that the action had begun again. I t would also be a mistake for him to assume that the order he had observed would be agreed upon by a l l of the actors involved. 97 We cannot expect, then, to f i n d symbolic expressions of minute rank gradations, say as between the houses numbered I I I and IV or X and XIII i n the l i s t s contained i n Appendix I. Such symbolizations could have been i n v a l i d a t e d at the next potlatch. What we can expect to f i n d , however, are symbolic expressions of wide rank d i f f e r e n c e s , as between houses ranked II and VIII i n a l i s t , or between those of high and low rank. Which brings us to cate- gories of rank. Here we f i n d another p r i n c i p l e of s o c i a l d i f f e r e n t i a -t i o n — in t o v e r t i c a l s t r a t a — which has symbolic expression. I t w i l l be l a t e r shown to be one of the p r i n c i p a l s o c i a l o rganizational fea-tures to be expressed or communicated by Tsimshian material culture. Status Levels The time i s probably long overdue f o r anthropologists to drop the word " c l a s s " from t h e i r descriptions of Northwest Coast s o c i e t i e s . This was, indeed, c a l l e d f o r by Drucker (1968 [1939]) as long ago as 1939, although f o r somewhat d i f f e r e n t reasons. The term " c l a s s " i s being applied with d e f i n i t i o n a l r i g o r by such t h e o r i s t s as Marshall Sahlins (1958, 1968), who re j e c t its a p p l i c a b i l i t y to s o c i e t i e s at t h i s l e v e l of s o c i o p o l i t i c a l i n t e g r a t i o n . A chiefdom i s not a class society. Although a stage beyond p r i m i t i v e equalitarianism, i t i s not divided i n t o a r u l i n g stratum i n command of the s t r a t e g i c means of production or p o l i t i c a l coercion and a disenfranchised lower c l a s s . I t i s a structure of degrees of i n t e r e s t : of graded f a m i l i a l p r i -o r i t i e s i n the control of wealth and force; i n claims to others' services, i n access to divine power, and i n material s t y l e s of l i f e — such that, i f a l l people are kinsmen and members of the society, s t i l l some are more members than others. For some are of superior descent. Yet, where rank i s thus l i n k e d to descent, status positions are often so 98 subtly d i f f e r e n t i a t e d that no one can say, or w i l l admit, where " c h i e f l i n e s s " leaves o f f and "commonality" begins (Sahlins, 1968: 24). Accordingly, I have adopted Sahlins' (1958: 3) term "status l e v e l s , " i n place of " c l a s s , " to designate categories of rank among the Tsimshian. Even so, there are d i f f i c u l t problems to be solved i n working out an accommodation between native or "conscious" models and the " s t a -t i s t i c a l " models suggested by other evidence (cf. Levi-Strauss, 1967c). There i s a strong ideology of mutually exclusive categories of rank among the Tsimshian and other Northwest Coast peoples which i s incompatible with the kinship-based s o c i a l organization of these s o c i e t i e s . A most revealing discussion of the d i f f i c u l t y of r e c o n c i l i n g a " c l a s s " ideology with s o c i a l r e a l i t y i s reported by Colson (1953: 202-226) for the Makah. While the people spoke of slave, common people, and c h i e f l y " c l a s s e s , " there was l i t t l e agreement as to who belonged i n each group or as to the c r i t e r i a f o r membership, e s p e c i a l l y f o r the c h i e f l y " c l a s s " : To some, ' c h i e f i s coterminous with household or family head and i s purely an ascribed status. They i n s i s t that only the f i r s t - b o r n son and daughter of the household head could i n h e r i t h i s status as ' c h i e f , and that t h e i r p o s i t i o n was not dependent upon i n d i v i d u a l merit. Others pr e f e r to beli e v e that a l l mem-bers of the 'ch i e f ' s ' family i n h e r i t e d h i s status .... To others, the ' c h i e f class i s l a r g e l y an achieved status, f o r they i n s i s t that the p o t e n t i a l ' c h i e f had to j u s t i f y h i s pos-i t i o n through great deeds ... or he dropped back into the class of commoners to be replaced by a younger brother or cousin who was more successful and thus was recognized as more worthy to lead the family. Others say that any person who could acquire s u f f i c i e n t wealth, by any means, to hold large potlatches i n h i s own name became a chief (Colson, 1953: 203). 99 Colson says that "indeed, i t seems as though each i n d i v i d u a l has received a p a r t i c u l a r r e v e l a t i o n with regard to what e a r l i e r Makah behaviour and b e l i e f were i n t h i s respect" ( i b i d . : 202). Her conclusions are that "the Makah today therefore do not possess a common coherent p i c t u r e of t h e i r former s o c i a l organization" and that "a class system t h e o r e t i c a l l y e x i s t s , but i t i s impossible for the observer to place any s i n g l e person i n h i s proper class because there are no generally accepted standards as to what constitutes a v a l i d claim to class status" ( i b i d . : 204). She further says the whole system i s "incomprehensible" to the observer ( l o c . c i t . ) . A key to understanding the contemporary Makah s i t u a t i o n , and the general problem of status l e v e l s on the Northwest Coast, i s perhaps contained i n the concept of p r e s t i g e , defined by F r i e d (1967: 32) as "the i d e o l o g i c a l component of status." He says, too,.that "I see no way to deal with prestige i n the absence of symbols, for prestige, stripped to i t s e s s e n t i a l s , depends upon shared evaluations of status" ( l o c . c i t . ) . When we ask what conferred prestige on the Northwest Coast the answer, of course, i s preeminently the pot l a t c h . I t can thus be seen as the medium through which Fried's "shared evaluations of status" were created and maintained. Among the Makah of the 1940s, which was when Colson did her fieldwork, the potlatch system was no longer operative. People could thus make claims to high or c h i e f l y status, but there was no i n s t r u -ment through which these claims could be l e g i t i m a t i z e d and community con-sensus obtained. While prestige was achieved through po t l a t c h i n g , the idiom with which i t was expressed was descent. There was an i m p l i c i t understanding that the a b i l i t y to potlatch was l i n k e d with high, usually c h i e f l y , descent. 100 The i n t e r l o c k i n g of descent and potlatching among the Tsimshian can be seen very c l e a r l y i n the case of c r e s t s , which indeed were the symbols F r i e d views as necessary i n dealing with prestige. Although one had to have an i n h e r i t e d r i g h t to a p a r t i c u l a r c r e s t , t h i s r i g h t had to be va l i d a t e d by a potlatch before the crest could be displayed. There was then a necessary, i f paradoxical (to us), r e l a t i o n s h i p between ascribed and achieved status, between descent, and wealth. Drucker says that "status ... was derived from kinship and expressed i n terms of wealth" (1958: 141). I am arguing the reverse: that status was derived from wealth, but expressed i n terms of kinship. Thus, the Coast Tsimshian categories of rank, which are l i s t e d below, were s a i d to be based on descent. Notice, however, that the category wa^ayin, the "Unhealed People," which i s usually glossed as "lower c l a s s , " i s a category of i n d i v i d u a l s and not lineages. Presumably, those people were born i n t o lineages which belonged to other categories of rank. The Coast Tsimshian had a strong ideology of v e r t i c a l s t r a t i -f i c a t i o n of t h e i r society into four groups: sgmg'ig'Et, l3kag*ig'Et, wa 'ayin, and t E l o . n g ' i t . Translations f o r these terms, and descriptions of t h e i r members, are as follows: 1. s^mg*ig'Et (s»m: " r e a l " ; g'ig'Et: "persons"): the Real People, also r e f e r r e d to as " r i p e . " The houses or lineages of c h i e f s , 21 c a l l e d " r o y a l houses" or " r o y a l t y " i n English by the Tsimshian. 21. The translations of Tsimshian terms i n t o English words such as " r o y a l , " " r o y a l t y , " "nobles," "prince," and "princess," are those of Tsimshian people themselves. 101 They are composed of the following groups: a) samo'*9ig*Et ("real person"): the chief himself; b) sigidgm hanax (-hanax: "wife"): Chief Woman, the ch i e f ' s w i f e ( s ) . c) 3:kuwE/ks3k (Iku- refers to "bright and s i l v e r y young salmon") : the "princes and princesses," c h i l d r e n of the chi e f ; d) g'ibawE'aksgk (translated by G a r f i e l d 1939: 177 as the " l i t -t l e n o b i l i t y " ) : the referent of th i s term i s n ' t c l e a r . H. Wallace indicated i n one place that i t ref e r r e d to the c h i e f l y status l e v e l as a whole; at another place i n the f i e l d notes he sa i d that i t referred to the men who were po-t e n t i a l successors to the chi e f , excepting h i s nephews or adopted h e i r s : "the g'ibawE'aksgk are those p r i m a r i l y e l i g i b l e .... I f the chief has no nephews or had adopted no children then these g* iba wE'3 ksg k come next i n rank f o r 22 chief. And those who e l e c t s e l e c t the one most f i t . " This i s consistent with Garfield's statement that i t i s "those people belonging to c h i e f s ' lineages who do not hold ranking names" (1939: 177). 2. Ijkag'ig'Et (meaning?): the c o u n c i l l o r s to a c h i e f , headmen of non-chiefly lineages. The name refer s both to the co u n c i l of headmen and to t h e i r status l e v e l as a whole. 22. David Aberle has pointed out to me the s i m i l a r i t y between t h i s d i s t i n c -t i o n and that made i n European roy a l t y between " h e i r apparent" and " h e i r presumptive" (see Webster's Seventh New C o l l e g i a t e D i c t i o n a r y ) . I t seems, on the basis of the European analogy, as i f the Tsimshian were d i v i d i n g the c h i e f l y group i n t o h e i r s apparent (one of which would nor-mally succeed) and p o t e n t i a l h e i r s presumptive (e.g., i n the absence of hei r s apparent, one of the g"ibe wE 'a ksj k would succeed). 102 3. wa >ayin (wa: "never"; 'ayin: "healed"): Unhealed People, also c a l l e d "green" (as opposed to the c h i e f l y group which was c a l l e d " r i p e " ) . "Honours have never been bestowed on them .... They have to go through ceremonies to be given a name, and then he (sic ) becomes a lflkag*ig'Et. The one who cannot do that remains i n the lower c l a s s . Once the name has been ra i s e d to the rank of Ifrkag*ig'Et i t i s i n h e r i t e d i n his family" (Mrs. Dudoward). My understanding i s that t h i s i s e s s e n t i a l l y a category of i n d i v i d u a l s , rather than lineages. I t i s composed of deviant re-l a t i v e s of the lfrkag*ig'Et status l e v e l (although i t may include former members of the c h i e f l y status l e v e l as w e l l ) , bastards, miscreants, and the children of slaves. "wa*ayin i s a class a free man reaches by coming from e i t h e r of the upper classes, simply by h i s conduct and behaviour he reaches that c l a s s . A person of continued bad character or bad temper can reach t h i s c l a s s . Sometimes a man i s placed i n t h i s class by h i s own r e l a -t i v e s to break his unruly temper. A person can r i s e again from the wa *ayin. A bastard c h i l d i s classed i n t h i s class and may never r i s e above wa'ayin. And i f a royal prince marries a woman of the common class against the w i l l of h i s people, h i s c h i l d r e n are looked upon as wa*ayin c h i l d r e n and he loses h i s chief standing. Children of slave parents, i f they get the consent of the chi e f , may become of the wa Jayin c l a s s . " (Joshua t s i b a s a ) . 4. iE-ko.ng*it ( p i . of T l i n g i t ) : slaves, war captives. 103 These four categories of rank, as described by the Tsimshian themselves, are not of equal order. The most stressed and fundamental d i v i s i o n i n Coast Tsimshian society was between the c h i e f l y category and the c o u n c i l l o r (lfrkag'ig'Et) category. In theory, they did not exchange members. "Nobody of the lgkag*ig'Et may become royal or pass into the upper house" (Mrs. Dudoward). "One of the lflkag'ig'Et may never pass i n t o the noble class or g'ibawE'aksak; and he may never assume the t i t l e of c h i e f . Never known of any such instance of passage from the lower [ i . e . , lakag*ig'Et] to higher c l a s s " (H. Wallace and Nelson). The a t t r i b u t i o n of c h i e f l y status or c o u n c i l l o r status was con s i s t e n t l y made by Coast Tsimshian people, e s p e c i a l l y when providing the ranked l i s t s of houses contained i n Appendix I (the d i v i s i o n between the two status l e v e l s i s indicated by a s o l i d h o r i z o n t a l l i n e i n the l i s t s ; i t i s missing only at g*itsamg.E'lam, where A. Stevens said "no l^kag*ig'Et . The system resembles the Gitksan i n that each head of the group was recognized as chief of h i s own group"). The wa*ayin are not represented i n the house l i s t s , confirming my impression that this was a category of i n d i v i d u a l s , and not of lineages. The " p u r i t y " of c h i e f l y status l e v e l houses or l i n e s was maintained, even when the lineage was threatened with e x t i n c t i o n , by the adoption or exchange of successors and women from c h i e f l y status l e v e l houses i n other t r i b e s . "In case of one house of the r o y a l family becoming e x t i n c t , we have to turn to the noble class of another t r i b e ; providing they are r e l a t i v e s and of the same crest. In case of absence of r e l a t i v e s anywhere of the ex t i n c t royal family, one of another house 104 and c r e s t , royal family, would be adopted. The g* ispaxlo 3 ts have so done for t h e i r present chief legEx, who was formerly of the royal family of the g*it"and?• l a x s k * i ' k " (Wallace and Nelson). Some of the c h i e f l y houses of the g*ispgwudwa'da had formalized t h e i r exchanges of women ("these houses have a system of providing each other with women to pre-vent t h e i r l i n e s from going e x t i n c t , " Beynon f i e l d notes, 1917). I f the house of wiceks of g'inax'ang'i'k were faced with going extinct through lack of women to bear a successor, seks of g* i t x a ' i a would send one of his nieces. The same re l a t i o n s h i p obtained between the houes of hE.1, g * i t x a ' l a , and ni.swE.xs, g*ina*d?'iks. These were c l o s e l y r e l a t e d houses, and the exchange of women re l a t i o n s h i p did not extend to the houses of ni.slk^mi.k of g* i l o d z a ^ a , or saxsa. axt of g* i t w i l g ? / J t s . People said that these two houses came to saltwater at a d i f f e r e n t time and did not reside for a time among the g ' i t x a ' l a , as did the wiceks group, n i . s l k y n i . k exchanged women with the Nass River g*ispawudwa 'd2 because they were more cl o s e l y r e l a t e d . Houses of the c o u n c i l l o r category were the bulk of the popu-l a t i o n . They were the s o l i d economic and moral support of the c h i e f s . In a potlatch speech, they were ref e r r e d to as "lakag*ig'Et, who are the strength of the chiefs of a l l the d i f f e r e n t t r i b e s . " "Whenever the head chief gives a potlatch, the l?kag*ig'Et have to contribute property; they give i t to the chief before the great potlatch i s given. The one who can contribute more to the chie f ' s p o t l a t c h i s the biggest man" (Mrs. Dudoward). The wa_jayj-n, the Unhealed People, were less a status l e v e l than a moral condition. There i s no way to know how many people i t represented, 105 but reference i s consistently made to wa'ayin i n d i v i d u a l s , not houses. I t seems to me to be of p r e c i s e l y the same character as the "worthless people" category of the Coast S a l i s h ( S u t t l e s , 1966). Suttles describes them as a "lower c l a s s , " which contained "those who, through t h e i r own or t h e i r forebears' misfortune or foolishness, had l o s t t h e i r l i n k s with the past and t h e i r knowledge of good conduct." Such people were d i f f i -c u l t to f i n d ; t h e i r existence was whispered and rumored, but no one would acknowledge to being of t h i s category. Hence, we might conclude that they were more of a "symbol" than a s o c i a l " f a c t . " Suttles decides that they were a small group, useful to maintain "the myth that morality i s the pr i v a t e property of the upper c l a s s " ( i b i d . : 175). "This myth made i t necessary (or at l e a s t useful) f o r a lower class to e x i s t as evidence f o r i t s t r u t h , but the myth probably also acted as a check on the growth of the lower c l a s s . I f the lower class grew too large, i t s existence would no longer be compatible with the myth; a large lower class would be seen by the upper class as a threat to society, and the at t i t u d e of the upper class would become i n t o l e r a b l e to the lower c l a s s " ( l o c . c i t . ) . Banishment to the category of the "unhealed" f o r the Tsim-shian, too, seems to have been used as an agent of s o c i a l c o n t r o l . I would go one step further than S u t t l e s , however, and say that i t was a category or residue of deviants — those who would not or could not accept the moral-i t y of Tsimshian society. I t was not, then, a "lower c l a s s " of Tsimshian s o c i e t y , but a deviant category of people who did not f i t elsewhere i n Tsimshian society. 106 The wa ayin were also described as being people of "doubtful o r i g i n . " They had assumed no honoured names, had no c r e s t s , no myth, and no past. In contrast to the c h i e f l y and c o u n c i l l o r categories, they almost seem to be non-persons, at l e a s t incomplete persons. This i s probably the sense of the expressions "unhealed" and "green" which were applied to them. One i s reminded of Levi-Strauss' (1969) concepts of "the raw and the cooked," with which he opposes the natural and c u l t u r a l orders. Because the wa*ayin did not p a r t i c i p a t e i n the two great s o c i a l and moral orders of Tsimshian society — the hala'fit order and the potlatch order — there was l i t t l e to define them as Tsimshian, or even human. A person who had no c r e s t s , no myth, no grandfather, and no knowledge of h i s ancestral home — undoubtedly a wa*ayin — was s a i d (by John Brown) to be " l i k e a w i l d animal" (see page 120 below). Slaves, at l e a s t , had the i d e n t i t i e s previously assumed i n the contexts of t h e i r own cultures. Coast Tsimshian categories of rank, i n sum, c l a s s i f i e d people (excluding slaves) into three groups on the basis of t h e i r being more and les s Tsimshian: 1) the Real People or c h i e f l y category (Tsimshian +), 2) the ordinary people or c o u n c i l l o r category (Tsimshian), and 3) the Unhealed People or deviant category (Tsimshian - ) . Membership i n the f i r s t two categories was expressed i n terms of descent, whereas the c r i t e r i a for membership i n the category of the Unhealed were e x p l i c i t l y moral q u a l i t i e s , or a condition of being deviant. These people could come from (be born into) e i t h e r of the other two categories. To c a l l them eit h e r a "lower c l a s s " or a "lower status l e v e l " would be misleading. 107 My conclusion i s that Coast Tsimshian society consisted of two status l e v e l s — a c h i e f l y status l e v e l and a c o u n c i l l o r status l e v e l — plus slaves and deviant Tsimshian, both of whom were e s s e n t i a l l y outside the society. At l e a s t i d e a l l y , marriages within status l e v e l s were endo-gamous. D i s t i n c t i o n s between the two status l e v e l s were maintained by, i f not created by, the potlatch system, one of i t s purposes being the manipulation of prestige symbols ( c r e s t s ) . The above conclusion, i t should be emphasized, contradicts the opinions of Tsimshian people and other anthropologists a l i k e , who have maintained that Tsimshian societ y was divided i n t o three v e r t i c a l s t r a t a or cla s s e s , usually said to be upper, middle, and lower classes, plus slaves. The Niska used the same categories of rank as the Coast Tsim-shian, but they maintained one important s t r u c t u r a l d i f f e r e n c e : Niska c o u n c i l l o r houses were "feeder" lineages for c h i e f l y houses and, when a c h i e f l y house was faced with e x t i n c t i o n , they supplied women to bear i t s successors. "The status of the l^kag* i g ' E t among the Coast Tsimshian and Niska d i f f e r e d i n t h i s way: among the Tsimshian, the l^kag'Et (sing.) can never hold r o y a l rank and may never become ch i e f s . Among the Niska, the lfrkag'Et's, as they may be c a l l e d , are i n r e a l i t y c hiefs of t h e i r own household and are recognized as princes but not of the high chief rank as chief of the group. But one of t h e i r duties i s when the house of kwaxcu, f o r example, i s threatened by e x t i n c t i o n through no more women, a woman may be taken from the lakag'Et ch i e f ' s household to become a f e -male member of the head chief's household, and her o f f s p r i n g are recognized as the head chief's own house, not as of the l^kag'Et house. The lfrkag'Et 108 has the same p r i v i l e g e ; a woman of the chief's household may become a member of h i s household, as do her child r e n . This has occurred many times between t h i s group" (Peter Calder). A c h i e f l y house may have several lflkag*ig'Et houses standing i n t h i s "feeder" r e l a t i o n s h i p to i t ; con-v e r s e l y , a 13kag*ig'Et house may stand as supplier of women to more than one c h i e f l y house. Being thus united by these exchanges of women and t h e i r pro-geny, c h i e f l y and c o u n c i l l o r houses of the,Niska were not d i f f e r e n t i a t e d i n t o status l e v e l s . My conclusion i s that Niska society was e s s e n t i a l l y a s i n g l e status l e v e l s o c i e t y , i n which differences between high and low rank had not yet resulted i n endogamous s t r a t a . Categories of rank for the Gitksan were not e x p l i c i t l y des-cribed i n the Barbeau/Beynon notes. However, we know from Jenness, who spent part of the winter of 1923 at Hazelton, that they c l a s s i f i e d people into c h i e f s , nobles, commoners, and slaves, categories borrowed by the Ca r r i e r . I n t e r e s t i n g l y , the C a r r i e r "now consider everyone a noble, at le a s t p o t e n t i a l l y " (Jenness, 1943: 571). The Gitksan also spoke of " c h i e f l y f a m i l i e s , " and of people being of "low" and "high" rank. A chief among the Gitksan was any lineage headman, although the chi e f s of the f i r s t - r a n k i n g house i n a clan i n a t r i b e or v i l l a g e had considerably greater prestige than the others. Extrapolating from these s l i m data, and using the Niska case as a model, I would hypothesize that the Gitksan had a s i n g l e status l e v e l s o c i e t y , plus slaves, i n which the extremes of rank were of considerable importance. Although they may not have had a named category of rank com-109 parable to the wa ayin among the Coast Tsimshian and Niska, Gitksan soc-i e t y must have also included a s t r u c t u r a l l y defined group of people who were outside of the potlatch and h a l a ^ i t systems. In h i s comparative study of Polynesian s o c i e t i e s , Sahlins (1958: 6) considered that "a society with a greater number of status l e v e l s than another can be considered more s t r a t i f i e d . " He then r e l a t e s the degree of s t r a t i f i c a t i o n i n Polynesian s o c i e t i e s to a d i s t i n c t i o n between " c h i e f and nonchief i n d i s t r i b u t i o n function," based upon produc-t i v i t y , so that: "the greater the p r o d u c t i v i t y , the greater the d i f -f e r e n t i a t i o n between d i s t r i b u t o r s (chiefs) and producers (nonchiefs) and the greater the tendency for t h i s d i s t i n c t i o n to extend i t s e l f into other aspects of culture" ( i b i d . : 249). I f Sahlin's t h e o r e t i c a l model of the r e l a t i o n s h i p s between p r o d u c t i v i t y , d i s t r i b u t i o n , and s t r a t i f i c a -t i o n i s correct, we should be able to predict that the i d e o l o g i c a l em-phasis upon status l e v e l d i s t i n c t i o n s i n Coast Tsimshian society was r e l a t e d to a greater d i f f e r e n t i a t i o n between d i s t r i b u t o r and producer than existed i n Niska and Gitksan s o c i e t i e s . To test t h i s , of course, would require another the s i s . Trade Relationships The greatest a b o r i g i n a l trading center on the northern North-west Coast was the mouth of the Nass River during eulachon f i s h i n g season i n the early spring. T l i n g i t , Haida, Gitksan, and people from the upper Nass converged each February on the area from Red B l u f f to Fishery Bay to trade with the Niska and Coast Tsimshian, who owned f i s h i n g grounds there. 110 Other trade r e l a t i o n s h i p s existed between the Coast Tsimshian and the Gitksan, the Niska and the Tsetsaut, the Niska and the C a r r i e r , and the Gitksan and the C a r r i e r , i n which foods from the Coast were traded for furs from the i n t e r i o r . A f t e r the fur trade began,two c h i e f s , legEx of the g * i s p a x l 3 / J t s and sagau*wE *n (Chief Mountain) of the lower Nass, seized monopolies over two of these trading patterns. Chief Mountain l i v e d f o r a while at kn^g.gli ("place of s c a l p s " ) , which was convenient for trade with the trading ships and the Hudson's Bay Company post at Fort Simpson. A predecessor of h i s had already es-ta b l i s h e d a trading r e l a t i o n s h i p with a group of Tsetsaut, who had no canoes, and whom he exploited, getting t h e i r furs " f o r almost nothing." This group, who were laxk*ibu^, are s t i l l r e f e r r e d to as xatks?m t**S3tsE/utgas  samSg'id^m sk*a'nis (slaves of Tsetsaut of Chief of Mountain) : "the Tsetsaut slaves of Chief Mountain." The missionary Tomlinson established a mission at K i n c o l i t h and gave the Tsetsaut better prices for t h e i r furs and i n v i t e d them to K i n c o l i t h . Chief Mountain was very b i t t e r over the los s of h i s trading p r i v i l e g e , but there was nothing he could do about i t , and he moved back to g * i t > i k s . legEx proclaimed an exclusive trading r i g h t f or himself and the g* ispaxlo / J t s with the Gitksan. In t h i s trade, h i s goods were the f i r s t s o l d , and he exacted a t r i b u t e from a f f i n a l r e l a t i v e s he permitted to accompany h i s expeditions. They usually went to the Gitksan. country three times a year, exchanging "saltwater foods" for b e r r i e s and furs. About 1850, he extended his trading expeditions to the C a r r i e r , eliminat-ing the Gitksan, who had been middlemen, and for several years held a I l l y e arly trading market at the junction of the Skeena and Bulkley r i v e r s (Jenness, 1943: 478). A n a r r a t i v e c o l l e c t e d from a g ' i t x a ^ a person (James Lewis) mentions a se r i e s of Coast Tsimshian trading r e l a t i o n s h i p s which may i n d i c a t e an a b o r i g i n a l precurser of legEx and the g* i s p a x l 0 "ts' trading p r i v i l e g e with the Gitksan. I f so, they had a p r i o r claim to t h e i r fur trade monopoly and did not seize the commerce a f t e r the fur trade period began, as has been assumed. The trading r e l a t i o n s h i p s r e f e r r e d to are as follows (the trading partners of the nine Lower Skeena t r i b e s were not more s p e c i f i c a l l y defined) 23 S t i k i n e T l i n g i t Haida of Prince of Wales Island Nass River Haida of Queen Charlotte Islands g'inax 'ang'i'k g'ilodza'ua g'itlE.'n g* i t w i l g ' a / t s g ' i t z a x l E ' l ) g'itsi.'s ) Those T l i n g i t with whom they had intermarried g * i s p a x l 3 ' ts g ' i t ' and?-'' g'itxa 'ia 24 Upper Skeena g ' i t s j l a ' s 3 (Canyon) Kitamaat and B e l l a B e l l a "Thus a l l had exclusive trading areas" (loc. c i t . ) . The same nar r a t i v e reports that the g' inax pang* i. yk secured t h e i r exclusive trading p r i v i l e g e 23. The n a r r a t i v e contains a long account of the purchase by wiseks, with the assistance of h i s g*inax'ang*i'k tribesmen, of a famous copper from Chief Shakes of the S t i k i n e . Because a l l of the t r i b e had con-t r i b u t e d to i t s purchase, the copper was considered t r i b a l property, although cared for by the c h i e f . 24. Swanton (1905: 79) reports that the g ' i t x a ' l a also had close associa-tions with the Haida of Skedans. Although he doesn't mention trade, he does say that the people of Skedans received "new c r e s t s , new s t o r i e s , and new features of the p o t l a t c h " from the g * i t x a ' t a . 112 with the S t i k i n e "by v i r t u e of having proclaimed i t as t h e i r r i g h t and by d i s t r i b u t i n g g i f t s to a l l the guests at a potla t c h . " There are i n d i -cations i n another n a r r a t i v e c o l l e c t e d from a g* ispaxl 3 " t s man (John Tate) that legEx's trading p r i v i l e g e with the Gitksan had also been pro-claimed and witnessed at a potla t c h . An important r e s u l t of these trading patterns, i n terms of t h i s and other museum studies, i s that a r t i f a c t s were often c o l l e c t e d from people who had not made them,and provenience records are thus often misleading. * * * * The most s i g n i f i c a n t aspect of Tsimshian s o c i a l organization described i n the preceding chapter, i n terms of i t s influence on material culture, i s the d i s t i n c t i o n developed between what was c a l l e d the potlatch order and the h a l a f i t order. Each finds r i t u a l expression i n contexts i n v o l v i n g formally and iconographically d i s t i n c t categories of material culture. Within the potlatch order, the predominant iconographic system i s that composed of " c r e s t s , " which are the subject of the following chap-te r s . 113 CHAPTER FOUR THE CREST CONTEXT This chapter i s included to put crests into ethnographic con-text, and to describe e s s e n t i a l r e l a t i o n s h i p s between the crest system and c e r t a i n r e l a t e d systems i n Tsimshian culture. Tsimshian crests are a series of named, totemic representa- t i o n s , usually of animals, applied to totem poles, house f r o n t s , cere- monial headdresses and robes, and c e r t a i n other objects of material c u l - ture. The Tsimshian use the English word " c r e s t " to replace three d i f f e r e n t l e x i c a l categories i n t h e i r own languages: ptEx, ayuks, and 1 dzepk. The differences between them are s i g n i f i c a n t . Boas (1916: 970), Sapir (1915: 3), and G a r f i e l d (1939: 336) tra n s l a t e the Tsimshian word ptEx to mean the clan i t s e l f , and i t i s often used by the Tsimshian i n t h i s way. But there i s also evidence that ano-ther meaning of the word i s the main totemic animal or animals used by the clan to generate cre s t s . Direct statements to Barbeau that "the k i l l e r -whale i s the ptEx of a l l the g*ispgwudwa'dz groups. The g r i z z l y would go together with the k i l l e r w h a l e as ptEx of the g'ispawudwa'dfr. The g* ispawudwa'd? have r e a l l y two ptEx. The other phratries ( s i c ) have only one ptEx." Or, "the g r i z z l y i s both a ptEx and a dzepk [material represen-t a t i o n of a c r e s t , see below]" and "eagle i s t h e i r ptEx and dzepk" suggest 1. Boas (1916: 500) mentions a fourth word for c r e s t s : sEnlai''duks, which he translates as "symbols," "marks," or "signs." I have found no other reference to t h i s word. 114 to me that the word ptEx p r i m a r i l y denotes the clan animal, and only by extension the clan i t s e l f . I t i s easy to see how the meaning of the word was extended to the clan, e s p e c i a l l y since laxsk * i fk means "on the eagle" and laxk* ibu^ "on the wolf" (g* isp^wudwa' d"a and g.anha'da are untranslat-able) . This use of the same word to r e f e r to crest animal and clan i s i n d i c a t i v e of the close symbolic association of the clan with the animal. ayuks i s the crest i t s e l f . I t i s the named, totemic e n t i t y that i s owned by a house and represented on c e r t a i n of i t s possessions i n material form. Whereas ptEx i s the animal species from which c e r t a i n p r i n c i p a l crests are derived, ayuks are those symbolic derivations them-selves. A dozen or more ayuks may be based on or derived from a s i n g l e ptEx animal. The crest l i s t s i n Appendix I I are l i s t s of ayuks. Barbeau 1s Tsimshian teachers r a r e l y used the word ayuks, pre-f e r r i n g to use the more concrete dzepk, which refe r s to the material rep-resentation of a c r e s t , the man-made thing or a r t i f a c t . According to Sapir (1917: 56): "dzapk i s c l e a r l y a d e r i v a t i v e of the verb dzab 'to make,' -k being a mediopassive s u f f i x ; dzab-k may thus be interpreted as 'what i s made' or 'what i s represented i n v i s i b l e form,' r e f e r r i n g probably to the carvings and other p l a s t i c representations of c r e s t s . " The differences i n the three terms are exemplified i n the f o l -lowing example: ptEx (clan) : laxsk*i^k ("on the eagle") ptEx (animal) : xsk.i<k ("eagle") ayuks : xski^ggm lo.b ("eagle of stone") 115 dzepk : Eagle form pecked i n sandstone boulder, belonging to the house of mgnEsk, laxsk * i . k , g* itlaxda'mks, Nass River (NMC, VII-C-1481) (see photo i n Barbeau, 1950: 37, specimen on the r i g h t ) . One of the concerns of t h i s study i s to investigate the r e l a -tionship between the ayuks — the named crest — and i t s dzepk's — i t s material representations. Another i s to search for the rules by which new crests (ayuks) were generated, since t h i s was an expanding system. These, w i l l be discussed i n the next chapter. Unless otherwise s p e c i f i e d , " c r e s t " w i l l be used here i n i t s ayuks sense. Crests were owned by houses, and were je a l o u s l y guarded posses-sions. They were a legacy from myth-time, acquired by the ancestors, and held i n perpetuity by t h e i r l i n e a l descendants. To display the crest of another house, without having secured the r i g h t to do so, was a challenge to the i n t e g r i t y and the very i d e n t i t y of the house. Crests were sometimes captured i n war and displayed by the v i c t o r s as humiliating reminders of defeat u n t i l purchased back by t h e i r owners. Some were not repurchased and show i n crest l i s t s as possessions of a l i e n houses. They were also taken as compensation for murder. Sometimes crests were loaned, usually to clan r e l a t i v e s , as gestures of generosity and s o l i d a r i t y and, on rare occasions, they were given away. When they had been f o r c i b l y seized, they were no longer displayed by t h e i r o r i g i n a l owners; when they were loaned or given away v o l u n t a r i l y , the o r i g i n a l owners continued to display them. Crests were never sold or transferred i n marriage by the Tsimshian: such actions would have been i n v i o l a t i o n of t h e i r very meaning. 116 The ownership of common crests implied a kinship r e l a t i o n s h i p , and membership i n the same clan or sub-clan. Clan kinship was extended on the basis of crest correspondences to the Haida, T l i n g i t , B e l l a B e l l a , and neighbouring Athapascans. Between the Gitksan and the Tsetsaut, t h i s was celebrated by a r i t u a l feast: "at the feast the Indians joined t h e i r crests. The Tsetsaut l a x s e f l became r e l a t i v e s of the Kisgegas laxsef1, and the laxk* i b u ' the same. Before they linked together i n t h i s 2 way there was always murder and treachery" (Arthur Hankin). Although crests were the property of the house,they • were vested i n i t s highest-ranking names, and under the c o n t r o l of i t s chief or headman. The house was perpetuated through i t s names, which were assumed or occupied by successive generations of m a t r i l i n e a l l y r e l a t e d men and women. Becoming Tsimshian, i n the sense of becoming a f u l l y adult, moral member of the community, was to p u b l i c l y assume these names i n a sequence of increasing importance and r e s p o n s i b i l i t y (see Duff, 1959: 40; G a r f i e l d , 1939: 224-226; Boas, 1916: 510-513). The chief or headman of the house was the man, or woman, who had assumed i t s most important name. Names were assumed at potlatches. A person who attempted to take a name without giving or p a r t i c i p a t i n g i n a potlatch was subject to r i d i c u l e . At the p o t l atch, the name was assumed i n context or association with crests owned by the house. Normally, the person assuming the name was also invested with the righ t to wear or display crests of the same general degree of importance as h i s new name. In the case of succession 2. This occurred a f t e r the s e t t l i n g of a boundary dispute. Hankin s a i d ( i n 1923) that the man who had brought the settlement about was s t i l l l i v i n g , i n d i c a t i n g that i t happened some time a f t e r ca. 1850. 117 to the name of a deceased c h i e f , the new incumbent erected a memorial totem pole which displayed important crests of the house. Crests were also transferred by a chief to h i s successor during h i s l i f e t i m e , i n which case he was making a pub l i c i n v e s t i t u r e i n his successor. I t i s important to d i s t i n g u i s h between kinds of potlatches and between a potlatch and what I c a l l a r i t u a l s e r i e s . The l a t t e r i s often c a l l e d a potlatch i n the l i t e r a t u r e , but i s a la r g e r r i t u a l event that includes a potlatch. Many c o l l e c t o r s f a i l e d to make t h i s important d i s t i n c t i o n , with the r e s u l t that masks, for example, can be found iden-i t i f i e d i n museum catalogues as "potlatch masks," when i n fact they were 1 used i n non-potlatch contexts i n a r i t u a l s e r i e s . The general term f o r potlatch i s y_aVk^ > a n c * i t s most obvious constant or defi n i n g features were 1) the d i v i s i o n of the people involved into two groups: hosts and guests, and 2) the pub l i c d i s t r i b u t i o n of wealth by the hosts to the guests. According to t h e i r purpose — the event the guests were i n v i t e d to witness — potlatches can be c l a s s i f i e d i n t o types. For example, there were house-building, marriage, and iygksa ("to cleanse"), or cleansing potlatches at which a mistake or i n d i g n i t y was "washed away." This i s the kind of potlatch often referred to i n the l i t e r a t u r e as a "face-saving" potlatch. The most important and frequent type of potlatch was what I am c a l l i n g the *sLx potlatch ( >>ix: "to proclaim," "to make known"). This was "the feast of assuming a name ... when assuming or e x h i b i t i n g a new cr e s t , or erecting a new totem pole" (Joshua t s i b a s a ) . ' j i x was ei t h e r the name of the potlatch or referred to i t s c e n t r a l event, or both; the 118 notes aren't c l e a r . R i v a l r i e s and challenges were t y p i c a l l y expressed through crest displays and crest assumptions within the j i x potlatch framework. R i t u a l s e r i e s were large-scale events which might include potlatches, throwing ceremonies, naxn o'x dramatizations, and dancing society i n i t i a t i o n s . The events diagrammed by Rosman and Rubel (1971: 24-25, based on G a r f i e l d , 1939: 198-201, 299-301, 305-309) as a t y p i c a l "Tsimshian pot l a t c h , " are i n fac t a r i t u a l s e r i e s , including a throwing ceremony, secret society i n i t i a t i o n s , and potlatch proper. The r i t u a l s e r i e s observed by Beynon i n 1945 at Kitsegukla included potlatches and naxn-a'x dramatizations. The d i s t i n c t i o n i s important. The Tsimshian themselves distinguished between hala.'it and potlatch (ya'j k u) events ( G a r f i e l d , 1939-: 192) , and i t was argued i n Chapter Three that these represent two separate orders or structures i n Tsimshian society. G a r f i e l d says that " i t i s often d i f f i c u l t to c l e a r l y separate the yaSkU from hala.'it a c t i v i t i e s , since the supernatural names and powers taken i n the h a l a ' i t are often as s t r i c t l y the property of p a r t i c u l a r lineages as the crests and house names and are dramatized i n much the same way and often at the same event" (1939: 192). I suggest that she was focussing on inappropriate features, and that attention to other features, i n c l u d i n g material culture, w i l l i n most cases make the d i s -t i n c t i o n r e a d i l y apparent. The more diagnostic of these are contrasted below: 119 h a l a ' i t features p o t l a t c h features chief addressed as wihala.'it chief addressed as sjmo* sig'Et no myth r e c i t e d myth r e c i t a t i o n prominent i n i t i a t e s wear cedar-bark neck r e c i p i e n t of name wears crest rings and head rings ( i n secret robe and headdress society) chief's costume: gwashalafit (Chilkat) chief's costume: crest robe and amhala^it ( f r o n t l e t ) headdress Raven Ra t t l e use of trumpets and whistles trumpets and whistles absent actors wear masks ( i n naxrij'x's) p r i n c i p a l s wear headdresses; t h e i r faces are not covered name i s dramatized ( i n iiaxno'x's) name never dramatized (although there may be a dramatization of a crest These d i s t i n c t i o n s w i l l be reinforced i n Chapter Six, where items of the material culture involved i n each order w i l l be i l l u s t r a t e d . also owned. The r e l a t i o n s h i p between the myth and the crest was a neces-sary one. This was recognized immediately by Barbeau and became one of the b a s i c themes i n a l l of h i s p u b l i c a t i o n s . In h i s review of Boas' Tsim- shian Mythology (1916), w r i t t e n during h i s f i r s t f i e l d season, Barbeau described t h i s r e l a t i o n s h i p more s u c c i n c t l y than he ever did l a t e r : "A crest without a myth to explain i t s o r i g i n and i t s connection with the owner was an i m p o s s i b i l i t y ; and such a myth was i n the patrimony of a clan 3 or a family" (Barbeau, 1917b: 560). 3. The s i g n i f i c a n c e of the crest-myth r e l a t i o n s h i p had been missed by Boas, an unfortunate r e s u l t of h i s r e l i a n c e on Henry Tate, who recorded the 1916 c o l l e c t i o n of myths f o r him. Whereas Boas believed that i t formed "the bulk of the important t r a d i t i o n s of the Tsimshian" (1916: 31), the fa c t of the matter was that the c o l l e c t i o n consisted of those myths which (continued next page ....) Crests were linked with houses by myths (ada.xox) , which were 120 The r e l a t i o n s h i p between crests and myths was a complex one. Persons who had no myth, probably those of the wa }ayin category, had no past, no c r e s t s , no i d e n t i t y : "a group that could not t e l l t h e i r t r a d i t i o n s would be r i d i c u l e d with the remark, 'What i s your ada/ox?' And i f you could not give i t , you were laughed at. 'What i s your grandfather's name? And where i s your crest? How do you know of your past, where you have lived? You have no grandfather. You cannot speak to me, because I have one. You have no ancestral home. You are l i k e a w i l d animal, you have (continued ...) were of the l e a s t (functional) importance to the Tsim-shian. Most of the myths Tate c o l l e c t e d were from what Boas c a l l e d The Raven Cycle, and were not only known by a l l Tsimshian but by t h e i r T l i n g i t and Haida neighbours as w e l l . Almost t o t a l l y absent from the c o l l e c t i o n were the numerous ada/ox or myths which t o l d of migrations and supernatural encounters during which crests were acquired. Since these l a t t e r number i n the many hundreds and form the bulk of the data c o l l e c t e d by Barbeau and Beynon over a period of 43 years, i t i s indeed a sobering thought that they were overlooked by Boas, the great c o l l e c t o r and publisher of myths. I t i s worthwhile quoting Barbeau's explanation i n f u l l : Why d i d Tate c o l l e c t general myths and t a l e s rather than l o c a l or s p e c i a l ones? The reasons for t h i s are f a i r l y c l e a r . The narratives of the f i r s t type are the property of a l l ; any informant at large may know and repeat them. Quite on the contrary, the second belong r e s t r i c t i v e l y to a clan, a house or a c h i e f . Not even the breakdown of the o l d order of things has yet abolished the deeply seated jealousy of the natives as to what formerly was t h e i r exclusive p r i v i -lege. No native, e s p e c i a l l y i n the presence of another, w i l l r e l a t e the t r a d i t i o n that concerns another; i t would be, to say the l e a s t , a breach of etiquette. We have no-t i c e d , moreover, that these are l i t t l e known, except by hearsay, to outsiders. Tate, who shared i n h i s compatriots' corrosive d i f f i d e n c e , does not seem to have overcome these b a r r i e r s . He i s not l i k e l y to have consulted many outside of h i s own family members. Hardly any of our twenty-five representative informants had been u t i l i z e d by him. The fact that he himself belonged to the lower class ... may not have made him persona grata with most of the chiefs — r o y a l or others (Barbeau, 1917b: 553). 121 no abode.' nyae" and adafox, grandfather and t r a d i t i o n , are p r a c t i c a l l y the same thing" (John Brown). In a manifest f u n c t i o n a l sense, the t e l l i n g of a myth at a potlatch validated one's r i g h t to claim and display the crests associated with i t . There were at l e a s t three, and probably four, necessary fea-tures involved i n such a v a l i d a t i o n : 1. The action must be r i t u a l l y "framed" i n a potlatch con-text. 2. The action must include some formal presentation of the crest or crests being validated. They may be represented on a new totem pole which i s being r a i s e d , painted on the front of a new dwelling which i s being dedicated, worn on the person of someone who i s r e c e i v i n g a name, or dramatically presented i n a staged performance. Such a pe r f o r -mance may be the occasion for the p o t l a t c h . That i s , the formal reason fo r g i v ing a potlatch may be to display a crest i n a dramatic and mem-orable (and therefore prestigious) fashion. This i s l i k e l y to be the case when a crest has been challenged, or to i t s e l f constitute a chal-lenge to the previously accepted owner of a c r e s t . More t y p i c a l l y , the reason f o r giving a potlatch i s to commemorate a dead c h i e f , assume a new name, or r a i s e a totem pole ( a l l three of which are usually involved i n a c h i e f ' s mortuary p o t l a t c h ) , and crest display i s included as a part of the s e r i e s of actions performed to achieve the more i n c l u s i v e end. 2a. I f the crests were to be represented on a totem pole, the carver had to be selected from the father's lineage of the owner of the c r e s t s : "... the p r i v i l e g e of carving the pole and 122 rendering s p e c i f i c ceremonial services f o r a l i b e r a l stipend f e l l to a smaller c i r c l e of strangers [people belonging to another c l a n ] , who may be termed a l l i e s or r e l a t i v e s by marriage. Not every ar-t i s t , though a stranger, could be i n v i t e d to carve a pole, as has often been supposed even among ethnologists. Far from i t . He must, indeed, be selected from among the 'fathers' of the deceased or h i s h e i r ; i n other words, he must be one of the 'fathers' of the members of t h i s family or one of t h e i r immediate r e l a t i v e s according to native computation" (Barbeau, 1929: 7). I f no carver stood i n the appropriate r e l a t i o n s h i p , one of the "fathers" appointed a substitute to do the actual carving while he "stood over him" ( l o c . c i t . ) . 3. I t i s the t e l l i n g of the associated myth that transforms a crest display i n t o a crest v a l i d a t i o n . The two above features must be present, however, f o r the simple t e l l i n g of a myth i n a non-potlatch context does not constitute crest v a l i d a t i o n . The types of myths i n -volved w i l l be examined i n more d e t a i l i n the chapter to follow. They must, however, include two d e t a i l s i n order to accomplish crest v a l i d a -t i o n : they must describe the events, usually i n v o l v i n g a journey, when the crest or the crest animal was encountered, and they must i n d i c a t e i n some way the r e l a t i o n s h i p of the t e l l e r of the myth to an ancestor i n -volved i n that encounter. 4. A formal speech by a guest acknowledging the myth t e l l e r ' s r i g h t to the myth and/or the congruence of the t o l d version with some previously heard and acknowledged version of the same myth (or, perhaps, 0 to some native myth model) , may or may not be necessary. The data are inadequate on t h i s point. Such a formal acknowledgement i s included i n 123 each of the crest v a l i d a t i o n scenes at the Kitsegukla potlatch s e r i e s , but i t also seems possible that the accepting of g i f t s d i s t r i b u t e d by the myth t e l l e r might constitute s u f f i c i e n t acknowledgement on the part of the guests. In a dispute over a crest at the Kitsegukla potlatches, the house opposing another house's right to display a p a r t i c u l a r crest on i t s new totem pole sang the following taunting song (Beynon, f i e l d notes): I gaze up to the sky I gaze up to the sky Where I see my uncle Who never l i e s about h i s myth. At the potlatch following the pole r a i s i n g , the chief explained h i s r i g h t to the crest i n question by r e c i t i n g h i s myth and the d e t a i l s of i t s ac-q u i s i t i o n , and the v i s i t i n g chiefs made speeches confirming h i s claim. The opposing house was forced to accept the s i t u a t i o n . The myth-crest r e l a t i o n s h i p also has an important, though normally l a t e n t , economic aspect, which helps to account for the func-t i o n a l s i g n i f i c a n c e of public crest v a l i d a t i o n . The same myth (ada.'ox) through which a crest i s v a l i d a t e d also expresses the house's claim to i t s t e r r i t o r i e s . T e r r i t o r i a l claims may not be expressed i n the version of the myth r e c i t e d at the crest v a l i d a t i o n , since the necessary elements for t h is version of the myth have to do with the ancestor's a c q u i s i t i o n of the c r e s t . But the f u l l myth contains, or can be expanded to contain, an account of the ancestor's, or the ancestral group's, migration to and/or possession of the t e r r i t o r i e s owned by the house, as w e l l as an enumeration of these t e r r i t o r i e s . Not many such f u l l t e r r i t o r i a l exten-sions of Tsimshian myths have been recorded. None have been published, 124 although even published versions of most Tsimshian myths contain hints 4 of t e r r i t o r i a l preoccupations. The c l e a r e s t , and most redundant, t e r r i t o r i a l expressions are to be found i n the myths of the migrations of the laxk*ibu* from the headwaters of the S t i k i n e , t h e i r s e t t l i n g f o r a time here and there with other l a x k * i b u 7 , who permitted the newcomers to e x p l o i t t h e i r t e r r i t o r i e s f o r a while before e x p e l l i n g them, and t h e i r f i n a l settlement on t e r r i -t o r i e s of t h e i r own on the Nass. The events dealing with crest a c q u i s i -t i o n during these migrations seem i n c i d e n t a l i n comparison to the search f o r land. The r e l a t i o n s h i p s involved i n the crest-myth-territory complex, as I hypothesize them, can be represented diagrammatically as follows: myth (1) / X . (2) crest / t e r r i t o r y (3) The house owns a myth which (1) valida t e s i t s r i g h t to display a c r e s t , and (2) expresses i t s claims to i t s t e r r i t o r i e s . Depending on to what end the myth i s being t o l d , the t e l l e r w i l l stress those events i n the myth which deal with e i t h e r (1) or (2). His audience, however, i s w e l l aware that i f they accept and acknowledge the events i n version (1), they are also accepting the events included i n version (2), thus creating 4. Barbeau, unfortunately, published only those versions of the myths he recorded which he judged to have " l i t e r a r y " merit, and which he could r e l a t e to such " u n i v e r s a l " themes as Orpheus, Samson, and Jonah (see Barbeau, 1950: 269-272 for an example), or those which dealt with such dramatic Northwest Coast themes as the Bear Husband. Hence, h i s published versions of Tsimshian myths give an unbalanced p i c t u r e of the t o t a l corpus he c o l l e c t e d . 125 the symbolic r e l a t i o n s h i p (3) between crests and t e r r i t o r y . Therefore, i n the shorthand of r i t u a l action, the crest becomes a v i s u a l symbol of the economic resources of the house that i s di s p l a y i n g i t . This must be what the people of Kitwancool meant when they wrote i n t h e i r h i s t o r y , "when a clan raises a totem-pole and puts t h e i r r i g h t f u l crests on the pole, i t means a great deal to them, as every pole has a hunting-ground" ( i n Duff, 1959: 37). I think that they were r e f e r r i n g to this t e r r i t o r y -symbolling function of c r e s t s , one that I do not be l i e v e has been ade-quately recognized i n the anthropological l i t e r a t u r e on the Northwest Coast. The function commonly ascribed to crests i n the l i t e r a t u r e i s that they express the s o c i a l i d e n t i t y of t h e i r owners. To be sure, they do. By d i s p l a y i n g his crests the owner expresses h i s clan a f f i l -i a t i o n , i n some cases h i s house membership, and, among the Coast Tsim-shian, some clues to h i s status l e v e l and h i s rank therein. I t w i l l be a major burden of the following analysis to explain how these q u a l i t i e s are expressed. But here i t must be pointed out that these are q u a l i t i e s of h i s s o c i a l p o s i t i o n that are already known to h i s fellow actors. They do not need to "read" a pole or a crest robe i n order to know how 5 to act towards i t s owner. Nor are crest emblems worn i n everyday s o c i a l i n t e r a c t i o n . Sapir (1915: 6) reports that "one cannot even pay a neigh-bour a v i s i t and wear a garment decorated with a minor crest without j u s t i f y i n g the use of such r e g a l i a by the expenditure of property at the 5. A possible exception might be the wearing of clan headdresses i n warfare, where quick clan i d e n t i f i c a t i o n may be necessary. 126 house v i s i t e d . " Barbeau's Tsimshian teachers were quite s p e c i f i c that crest-bearing costume items were worn at potlatches. Crests, therefore, are worn and displayed i n r i t u a l contexts where they express features of s o c i a l p o s i t i o n which are already known. Which suggests that we are dealing with the kind of r i t u a l function des-cribed by Leach (1965: 15-16): In sum, then, my view here i s that r i t u a l action and be-l i e f are to be understood as forms of symbolic statement about the s o c i a l order. Although I do not claim that anthropologists are always i n a p o s i t i o n to i n t e r p r e t such symbolism, I hold nevertheless that the main task of s o c i a l anthropology i s to attempt such i n t e r p r e t a t i o n .... R i t u a l i n i t s c u l t u r a l context i s a pattern of symbols; the words into which I i n t e r p r e t i t are another pattern of symbols composed l a r g e l y of t e c h n i c a l terms devised by anth-ropologists — words l i k e lineage, rank, status and so on. The two symbol systems have something i n common, namely a common structure .... This i s what I mean when I say that r i t u a l makes e x p l i c i t the s o c i a l structure. The structure which i s symbolised i n r i t u a l i s the sys-tem of s o c i a l l y approved "proper" r e l a t i o n s between i n d i v i d u a l s and groups. These r e l a t i o n s are not formally recognised at a l l v times .... Indeed I am prepared to argue that ... neglect of formal structure i s e s s e n t i a l i f ordinary s o c i a l a c t i v i t i e s are to be pursued at a l l . Nevertheless i f anarchy i s to be avoided, the i n d i v i d u a l s who make up a society must from time to time be reminded, at l e a s t i n symbol, of the underlying order that i s supposed to guide t h e i r s o c i a l a c t i v i t i e s . R i t u a l performances have t h i s function f o r the p a r t i c i p a t i n g group as a whole; they momen-t a r i l y make e x p l i c i t what i s otherwise a f i c t i o n . Following Leach, the main task of the following chapters w i l l be to i n t e r -pret crests as a system of r i t u a l statements about the Tsimshian s o c i a l order. A secondary purpose w i l l be to i n t e r p r e t c e r t a i n items of Tsim-shian material culture as cres t s . * * * * 127 The naxna'x System: Before turning to the crests themselves, i t i s i n s t r u c t i v e to contrast crests and crest displays withnanother iconographic and r i t u a l system of the Tsimshian: naxno''x dramatizations. naxn?'x i s the Tsim-shian word for " s p i r i t " or "supernatural being," and i n t h i s context i t r e f e r s p r i m a r i l y to lineage-owned names which were dramatized i n p u b l i c performances. The Tsimshian c l a s s i f y naxna"" x dramatizations as h a l a f i t ' s — dances or power demonstrations — which are both conceptually and as event sequences d i s t i n c t from y a 9 3 k U or potlatches. However, there are many concrete instances i n the crest l i s t s (Appendix II) where a named e n t i t y i s claimed as both a crest and a naxnV x. I t i s therefore neces-sary to look at naxna'x's i n order to understand c r e s t s . The naxny'x naming system has not previously been described, although i t was b r i e f l y characterized by Duff (1964b: 68-69) i n h i s des-c r i p t i o n of the Barbeau/Beynon f i e l d notes. While I have extracted con-siderable data about i t from the notes and have found numerous w e l l -documented examples of naxn?'x masks i n museum c o l l e c t i o n s , a f u l l pre-sentation of the system must await a l a t e r study. What foJJows i s only an o u t l i n e , included here i n order to ill u m i n a t e an area of confusion or over-lap between naxno^x's and c r e s t s . Two types of naxna'x performances can be distinguished: 1) the name dramatizations, and 2) c h i e f s ' power demonstrations. There does not, however, seem to be a terminological d i s t i n c t i o n — both are re f e r r e d to by Tsimshian people simply as naxna'x. The other common usage of the 128 term i s to r e f e r to whistles used i n these performances and i n secret society i n i t i a t i o n s , and believed by spectators to be the voices of s p i r i t s . The use of naxno'x whistles i n these events i s s i g n i f i c a n t of a basic conceptual d i f f e r e n c e between name dramatizations and crest d i splays: s p i r i t s or supernatural beings were believed to be present when t h e i r voices were heard. Their immediate and continuing power was demonstrated i n the event. Crests, on the other hand, were bestowed by or taken from supernatural beings i n mythtime, and the crest display was i n commemoration of that past supernatural event. I f one were to invoke the sacred/secular dichotomy which i s t r a d i t i o n a l i n Northwest Coast studies, h a l a f i t ' s (including naxn-a'x's) were sacred; potlatches 6 (including crest displays) profane or secular. Here, I w i l l separate the two kinds of naxno'x's and discuss the naming system f i r s t . Like c r e s t s , naxnj' x names were owned by houses and i n h e r i t e d m a t r i l i n e a l l y . Unlike c r e s t s , myths associated with them to explain t h e i r o r i g i n s are conspicuously absent from the a v a i l a b l e data, nor are there any myths, n a r r a t i v e s , or speculations by Tsimshian 6. I f , however, we look at the potlatch c a r e f u l l y i n terms of many con-temporary d e f i n i t i o n s of r e l i g i o n , i t i s a sacred or r e l i g i o u s event. Consider, for example, Geertz's (1968: 4) well-known d e f i n i t i o n of r e l i g i o n as "(1) a system of symbols which acts to (2) e s t a b l i s h powerful, pervasive, and l o n g - l a s t i n g moods and motivations i n men by (3) formulating conceptions of a general order to existence and (4) c l o t h i n g these conceptions with such an aura of f a c t u a l i t y that (5) the moods and motivations seem uniquely r e l i g i o u s . " The a p p l i c a t i o n of such a d e f i n i t i o n of r e l i g i o n to the r i t u a l and b e l i e f systems of the Northwest Coast would also free us of such ethnocentric judgments as "the culture i s p e c u l i a r l y lacking i n the usual forms of r e l i g i o u s worship, e s p e c i a l l y among the more complex northern groups where s o c i a l structure and i t s implications formed the core of c u l t u r a l a c t i v i t y " (Gunther, 1966: 69). 129 people i n the Barbeau/Beynon data concerning the o r i g i n of the naming system. Since i t was not found i n neighbouring groups, as were other kinds of h a l a / i t ' s , the conclusion i s suggested that the naxno'x naming 7 system was an old and uniquely Tsimshian complex. 7. Beynon thought naxno^x's corresponded to the lower ranked dances of the Northern Kwakiutl, which were also family-owned s p i r i t dramati-zations or impersonations. In a l e t t e r to Barbeau dated February 25, 1955, he referred to the Owikeno lower ranked dances l i s t e d i n Drucker (1940: 202, nos. 9-23) and wrote that these "to my mind correspond to the Tsimshian naxnj'x's or p r i v i l e g e s which i n Tsimshian dramatizes c e r t a i n p r i v i l e g e s , the various i n d i v i d u a l s have each p r i v i l e g e having i t s own song or breath as i t i s termed (k,sgmE3:k) and these p r i v i l e g e s are usually dramatized at a ya°ok feast when each guest w i l l show and dramatize h i s own s p e c i a l p r i v i l e g e . I do not remember whether you dealt with t h i s to any great extent excepting where the naxng'x was assumed as a name, such as,  7al. 'p = stone thrower; l i ' luks = t h i e f ; hukbi*k = l i a r ; rio'tiist = crazy person, etc. I have seen some of the dramas i n action, some of which I think I wrote i n my recordings of the totem pole r a i s i n g s at gidzagu'kla some years ago." I t should be pointed out that naxny'x's were not dramatized i n the ya*jk , as Beynon says. At the Kitsegukla ser i e s to which he r e f e r s , they were given i n the days preceding the potlatches and pole r a i s i n g s . While there are indeed some s i m i l a r i t i e s i n the Northern Kwakiutl lower ranked dances and naxna'x dramatiziations, the two seem to me to be quite d i f f e r e n t . The s i m i l a r i t i e s include a c e r t a i n suspense hinging on the r e v e l a t i o n of the name of the s p i r i t being impersonated. Of a kumogwa performance, Drucker (1940: 208) writes: "The attendants have been pretending great t e r r o r , because they could not l e a r n what the s p i r i t i s . As soon as the master of ceremonies reveals the s p i r i t ' s name, they become assured, and assume a pompous I-told-you-so a t t i t u d e . " There i s also i n both an aspect of what Drucker ( l o c . c i t . ) c a l l s "buffoonery." However, there i s no suggestion i n the l i t e r a t u r e on the Northern Kwakiutl that the s p i r i t s involved i n these dances were also taken as personal names, nor was the range of s p i r i t s involved anywhere near as large as i n the naxnp'x system. Further, those who p a r t i c i p a t e d i n the Northern Kwakiutl dances were organized into s o c i e t i e s and the dancers were i n i t i a t e s who were possessed, disap-peared, and returned with supernatural powers, much as i n the higher ranked Shaman's Series (Drucker, 1940: 202-^203). S t i l l , Beynon's suggestion i s i n t r i g u i n g , and the two r i t u a l s should be c a r e f u l l y compared before i t can be rejected. 130 The names and t h e i r dramatizations were described by Duff (1964b: 68) as follows: They are assumed e x c l u s i v e l y by adults, and are somewhat more common among the Gitksan than the other d i v i s i o n s . These names tran s l a t e into such terms as " l i a r , " "person of long ago," "always sleeping," "trouble-maker," "propped up," " a l -ways begging," "conceited woman." When the name i s assumed, and on other occasions when the "house" wants to entertain guests with.;-; a display of some of i t s prerogatives, members of the lineage stage a performance which i n some way drama-t i z e s the name, or more exactly, the supposedly supernatural being to which the name r e f e r s . These performances make use of a great v a r i e t y of masks and t r i c k s of stagecraft. Once assumed, the name i s not reserved for ceremonial occasions but i s used as an everyday secular name. For the present study, Duff's d e s c r i p t i o n needs elaboration i n s everal respects. F i r s t , there was a status l e v e l difference i n the use of naxno'x names among the Coast Tsimshian. "In the r o y a l f a m i l i e s they don't adopt the naxno'x as a rule for t h e i r regular names, but the lakag'ig'Et do. There are exceptions when Inkag*ig'Et don't adopt t h e i r naxn^x as a regular name" (Mrs. Johnson, g'inax 1 ang'i'k). Then, too, the kinds of names involved were more varied than Duff i n d i c a t e s . In a ddition to those r e f e r r i n g to human q u a l i t i e s or a t t r i b u t e s , such as he l i s t s , there were also names r e f e r r i n g to animals. These are the naxnj'x's which overlap most commonly with c r e s t s , and were included by people i n the crests l i s t s . Names r e f e r r i n g to humans, however, outnumber those r e f e r r i n g to animals about f i v e to one. Some examples of each type are l i s t e d below. When the animal name was claimed by someone i n a clan which d i d not have that p a r t i c u l a r animal as a c r e s t , the clan of the naxna'x owner i s indi c a t e d i n parentheses: 131 Proud Lowly Slave Quarrelsome Crying Without Reason Deaf Mannish Woman Covetous Person Dumb Person Throwing Stones Stupid T l i n g i t Eager to Dance Mocking Others Wrinkled Old (Person) White Man Continually Nodding Lazy Causing Fights Always Crying Sassy Man Gr i z z l y Invader G r i z z l y (laxsk*i^k) Homeless Bear G r i z z l y Man Black Bear Man (laxsef1) Big Mouth G r i z z l y ( l a x s e f l ) G r i z z l y of the Large Rat Gr i z z l y Eating Salmonberries Raiding Wolf Wolf Man (laxse^!) Migrating Wolf Small Raven (g'isg.afst) Frog F l y i n g Frog Owl Great Thunderbird Eagle of the Sky ( l a x s e f l ) Sea Lion (laxsk*i'Ck; laxk' ibu' ) Sculpin (laxsk* iCk) Barbeau's hypothesis f o r the overlap between crests and naxn^'x's, which can be r e a d i l y i n f e r r e d from numerous references i n Totem Poles of the Gitksan (1929), was that some e n t i t i e s which were o r i -g i n a l l y naxnVx's l a t e r became elevated to the status of crest s . For ex-ample, when r e f e r r i n g to the laxk'ieu' crest of Migrating Wolf ( i b i d . : 123), he says that " t h i s crest was used as a narhnawk i n the f i r s t place, that i s , a personal s p i r i t name. I t i s s t i l l dramatized as a narhnawk, 132 although i t has grown i n t o a family crest as w e l l . " A f a r more s a t i s -f y i n g explanation, e s p e c i a l l y for those instances i n which the naxnj'x name refer s to a crest of the clan of i t s owner, i s suggested by Sapir (1915: 5): Thus, while the k i t w i l ^ n a . k ? i 31 clan of the k i t ' a n w i / l / > k c t r i b e , the second clan i n the rank of the Wolf phratry as represented i n the t r i b e , does not possess the r i g h t to use the wolf as a r e a l c r e s t , i t nevertheless can show i t i n a potlatch " f o r fun," as i t i s t h e i r p h r a t r i c emblem; the point i s that they may not use the wolf crest to increase t h e i r p r e s t i g e , as by the giving away of property i n connexion with i t . Sapir's references to the use of the wolf as other than a " r e a l c r e s t " and " f o r fun" leave no doubt that naxnVx's are meant. The c r i t i c a l d i f f e r e n c e , then, between crests and naxnVx's i s that naxns'x name dramatizations were not prestige enhancers or accompanied by property d i s t r i b u t i o n s . Since these potlatch features were not involved, the normally s t r i c t rules could be relaxed. The absence of property d i s t r i b u t i o n s i n connec-t i o n with naxn a'x dramatizations i s confirmed by Beynon's account of some two dozen naxnj'x's at Kitsegukla i n 1945. The naxnj'x system also permitted houses to dramatize animals used as crests by other clans, as i n d i c a t e d i n the above l i s t . This was-confirmed by a Coast Tsimshian informant as follows: "A man of one crest may use the animal of another crest [as a naxn>' x] . Thus n i . s b ' t [g.anha'dp] takes a g r i z z l y f o r h i s naxna' x; legEx [laxsk* i'k] has a g r i z z l y [ a l s o ] . " The same person offered i n explanation cnly that "a naxny'x and a dzEpk [crest] are very d i f f e r e n t things." I t i s p o s s i b l e that these animal naxnj'x's were o r i g i n a l l y father's crests ( c f . , the use of father's crest i n cross-clan names). One of the naxnj'x's dramatized at Kitsegukla i n 133 1945 was a lotk, or armour, made of g r i z z l y skin. While t h i s was a l a x k * i b u 7 c r e s t , the owner of the name was g.axsqabax , a g.anha'da. Beynon explained that he was p r i v i l e g e d to wear the laxk* ibu'' crest "as h i s paternal o r i g i n was wolf and t h i s was not a crest but rather a nam? x." In the passage quoted above, Sapir mentioned a "fun" aspect of n a x n J ' X performances, which i s i n d i r e c t contrast to the solemn f o r -mality of the potla t c h . What Bateson would c a l l the "ethoses" (1959: 2, footnote) of the two r i t u a l s were e n t i r e l y d i f f e r e n t . Beynon attempted to summarize one aspect of the difference a f t e r witnessing the name dra-matizations at Kitsegukla: "strange to say, speech-making does not enter i n t o t h i s and only very l i t t l e i s a c t u a l l y s a i d , the actions and the sing-ing and the announcing i s the only method one can f i n d out what i s a c t u a l l y happening." Unlike potlatches, which have the nature of l e g a l testimony i n our own soci e t y , the analogous Western form to the naxn:/x name dra-matization i s the charade. They were short dramas i n which the performer presented various clues to the audience, by h i s mask and costume as we l l as h i s actions, u n t i l i t was able to guess, or the master of ceremonies announced, the name or s p i r i t he was impersonating. They were suspenseful, f r i g h t e n i n g , funny, and ambiguous. I n t e r e s t i n g l y , naxno'x performances gave the peqle an opportunity to express antagonism to or resentment of the c h i e f s . There are several cases recorded i n which the performer re-v i l e d or p h y s i c a l l y abused the c h i e f s . This was apparently done with impunity, although the chiefs had to be paid small compensations afterwards. One imagines that these incidents were greatly enjoyed by the audience. 134 A sub-type of naxn t'x name dramatization was c a l l e d the sedu'lsj? or "restore to l i f e " naxnj'x. In these, the performer dramatizing a name "died" or was " k i l l e d , " and the chiefs present were c a l l e d upon — i n order of rank — to t r y and restore him to l i f e . Each i n turn acted the r o l e of a curing shaman over the body, unsuccessfully, u n t i l the l a s t one to perform succeeded. This, of course, was another s i t u a t i o n , reminiscent of h i s w i h a l a ' i t r o l e , i n which the chief could demonstrate hi s c o n t r o l of supernatural power to the people. While a s t r u c t u r a l analysis could doubtless show s i g n i f i c a n t oppositions and reversals between potlatches, with t h e i r celebration of the p r e v a i l i n g authority structure, and naxn?' x dramatizations, the mani-f e s t purpose of the l a t t e r to both actors and audience was entertainment. One of the most important r e s u l t s of the discovery of the Tsim-shian naxn •>'x name dramatizations i s that we can now explain the function of Tsimshian masks, which are found i n great numbers i n museum c o l l e c t i o n s . They include both human face masks, the largest category, and animal masks 8 and were used p r i n c i p a l l y , i f not e x c l u s i v e l y , i n naxn3 ~x dramatizations. The importance of t h i s association i s indicated by the following two quotations: I t i s probable that the T l i n g i t , Haida and Tsimshian acquired t h e i r masks along with the Dancing Society performances which they adopted from the Kwakiutl .... (Drucker, 1955b: 74). 8. Some masks may also have been worn i n dramatizations of myths. This i s suggested only by some c r y p t i c references i n museum catalogues and i s not very convincing. S t i l l , I prefer to leave the p o s s i b i l i t y open u n t i l more d e f i n i t i v e data appear. 135 There are no i n d i c a t i o n s i n the Barbeau/Beynon data, nor r e l i a b l e sug-9 gestions i n the published l i t e r a t u r e , that masks were used i n Tsimshian dancing s o c i e t i e s . [Regarding Tsimshian human face masks i n the Rasmussen c o l l e c -tion] whether these masks were used for ceremonial dancing or for shamanistic performances cannot be determined from Mr. Rasmussen's notes, and the meager l i t e r a t u r e gives no clues (Gunther, 1966: 134). Tsimshian shamans did not wear masks, and Gunther's reference to "cere-monial dancing" i s too vague to be useful . The major hypothesis underlying my i n t e r p r e t a t i o n of Tsimshian masks and headdresses i s that when the wearer's face i s v i s i b l e — i . e . , when the object i n question i s a headdress — i t s iconography ref e r s to a c r e s t , and that when the wearer's face i s hidden — i . e . , when the object i s a mask — i t s iconography ref e r s to a naxn3 /x. This derives from the d i s t i n c t i o n s made i n t h i s chapter between crest displays and naxna'x name dramatizations. The crest headdress was worn to enhance the prestige and rank of the wearer, and to l e g i t i m i z e him, as an i n d i -v i d u a l , as the successor to a name-title and i t s associated cr e s t s . The naxno'x mask, on the other hand, was worn i n order to disguise the person who was impersonating a s p i r i t , and whose own i n d i v i d u a l i t y would have i n t e r f e r e d i n the effectiveness of his performance. S i g n i f i c a n t l y , c hiefs sometimes hi r e d other people to perform t h e i r naxna'x's for them, confirming that the i d e n t i t y of the person was not important. This would have been inconceivable i n the display of a crest headdress. This 9. Tate ( i n Boas, 1916:548) mentions the use of b i r d masks i n halaCit's at K i t k a t l a . He does not elaborate, and I suspect that they may have been i n fa c t worn by Northern Kwakiutl who were v i s i t i n g there. 136 d i s t i n c t i o n between crest headdresses and naxno'x masks becomes e s p e c i a l l y c r i t i c a l i n the case of those masks representing animals also used as cres t s . I t w i l l be tested i n Chapter Six. The other kind of naxna'x's are those I r e f e r r e d to above as c h i e f s ' power demonstrations. These were not name dramatizations, but p r i v i l e g e s of a d i f f e r e n t s o r t , which are somewhat d i f f i c u l t to conceptu-a l i z e . Although they were c a l l e d naxnj/x's or " s p i r i t s " and were per-formed i n hala.'it and not potlatch contexts, they do not seem to have re-ferred to s p e c i f i c or i d e n t i f i a b l e supernatural beings, nor did they seem to be predicated on the spectators b e l i e v i n g that supernatural power was n e c e s s a r i l y involved i n t h e i r performance. Rather, they were elaborate t r i c k s of stagecraft, usually dependent upon the kinds of mechanical contrivances s a i d to be the s p e c i a l t y of the g' i t ' s ^ ' n t k . Like naxnp'xnames, c h i e f s ' power demonstrations could display animals used as crests by other clans. A naxn?'x of t s i b a s E g ' i s p ^ w u d w a ' da chief of g ' i t x a ' l a , was a hollow pole with a raven on top, set up at the rear of the house. As t s i b a s E ' and the v i s i t i n g c h iefs walked around i t and the people sang, the raven flapped i t s wings and the pole grew u n t i l i t touched the c e i l i n g , and then i t disappeared i n t o the ground. The naxn3'x was named sg.anhagu /hE (sg.an: "wooden"; hagu /hE: a T l i n g i t word, meaning unknown) and was said to be of T l i n g i t o r i g i n . Another naxna'x of t s i b a s E 7 , named Revolving Steps, was used to humiliate legEx, the two chiefs having had a long-standing r i v a l r y : " i n the h a l a / i t ' s , each would always t r y to outdo the other i n having the best naxna'x's and ... the most modern devices" (Henry Watt, g ' i t x a ' i a ) . 137 As he was preparing for the event, t s i b a s E 7 was reported (by Watt) to have s a i d : "I w i l l now use and i n s t a l l one of my most d i f f i c u l t and se-cret naxn->'x's: my Revolving Steps. Over these a l l the guests must ar-r i v e and only a few w i l l be able to enter over them without humiliation. As t h i s i s a naxns'x and my s p e c i a l p r i v i l e g e to use i n my own fe a s t , no one must take offense. But I w i l l humiliate them .... I want e s p e c i a l l y to humiliate legEx and to show h i s t r i b e that we are as b i g as they are and as clever." The naxn:/x was e s p e c i a l l y constructed steps going into the house which could be made to revolve; anyone not knowing about them was l i k e l y to be thrown o f f balance and to tumble ignobly i n t o the house i n front of those already i n s i d e . Since one of legEx's s p e c i a l p r i v i l e g e s was the right to come l a s t to any fe a s t , t h i s was an e s p e c i a l l y e f f e c t i v e ploy i n tsibasE's part. The use of words such as "modern devices" and " c l e v e r , " plus the t r i c k e r y involved i n the Revolving Steps, suggests that these c o n t r i -vances should be regarded more as demonstrations of technological power than supernatural power. However, they were used i n the h a l a ' i t context, and as such undoubtedly had connotations of the supernatural basis or sanction of the c h i e f ' s authority. They can also be viewed as i n t e g r a t i v e s t r a t e g i e s used by r i v a l c h i e f s , to which t h e i r e n t i r e t r i b e s could r a l l y or respond, since they made no reference to clan c r e s t s . This theme of i n t e g r a t i v e c h i e f l y symbols w i l l be further developed i n the next chap-t e r s . 138 CHAPTER FIVE THE CREST SYSTEM I: GENERAL STRUCTURE In 1962, Levi-Strauss surveyed the l i t e r a t u r e on "the problem of totemism" of the previous half-century, noting the demise of. various e m p i r i c i s t , psychological, and f u n c t i o n a l totemic hypotheses, and r e -s u r r e c t i n g what we might c a l l "neo-totemism" with his now famous s t a t e -ment that natural species were chosen as totemic c l a s s i f i e r s "not be-cause they are 'good to eat' but because they are 'good to think'" (1963: 89). "The passage from a concrete to a formal d e f i n i t i o n of totemism," he notes (1963: 10)5 " a c t u a l l y goes back to Boas." The work he r e f e r s to i s Tsimshian Mythology (1916) , i n which Boas stated that totemism was the a p p l i c a t i o n of a r u l e of homology between a system of denotation and a s o c i a l system which was being denoted, and that "the homology of d i s t i n -guishing marks of s o c i a l d i v i s i o n s of a t r i b e i s proof that they are due to a c l a s s i f i c a t o r y tendency" of the human mind (1916: 519). Boas be-l i e v e d that the use of natural species as the basis f o r a totemic system of denotation was a r b i t r a r y , and that to consider i t otherwise was to take on the e n t i r e l y d i f f e r e n t problem of "the r e l a t i o n s h i p of man to nature, which i s obviously quite d i s t i n c t from that of the c h a r a c t e r i z a t i o n of kinship groups. The only connection between the two problems i s that the concepts r e f e r r i n g to the r e l a t i o n of man to nature are applied for the purpose of characterizing s o c i a l , more p a r t i c u l a r l y kinship groups" (Boas, 1916: 517). 139 On the contrary, according to Levi-Strauss, there are neces-sary, l o g i c a l r e l a t i o n s between the system of denotation and the system that i s denoted. "The animal world and plant l i f e are u t i l i z e d not merely because they are there, but because they suggest a mode of thought" ( L e v i -Strauss, 1963: 13). The " r e a l l i n k between the two orders i s i n d i r e c t , passing through the mind" (loc. c i t . ) . This mode of thought was the subject of Levi-Strauss' next book, La Pensee Sauvage (1962) , translated i n t o English as The Savage  Mind (1966). Here he defined "the science of the concrete," of which totemism was j u s t one expression. "As I showed i n an e a r l i e r book and am continuing to e s t a b l i s h here, s o - c a l l e d totemism i s i n fac t only a p a r t i c u l a r case of the general problem of c l a s s i f i c a t i o n and one of many examples of the part which s p e c i f i c terms often play i n the working out of a s o c i a l c l a s s i f i c a t i o n " (1966: 62). An important concept i n the science of the concrete, or the science b u i l t upon the c l a s s i f i c a t i o n of concrete images drawn from na-ture, i s what Levi-Strauss c a l l s "the species notion" or "totemic oper-ator." Plant and animal species are admirably suited as l o g i c a l vehicles due to t h e i r "intermediate p o s i t i o n as l o g i c a l l y equidistant from the extreme forms of c l a s s i f i c a t i o n : c a t e g o r i c a l and sing u l a r " ( i b i d . : 149). Being thus a "medial c l a s s i f i e r , " the species concept can widen i t s r e -ferent upwards i n the d i r e c t i o n of associated elements (e.g., sky/earth) or categories (e.g., high/low), or contract i t downward, i n the d i r e c t i o n 1 of proper names. I t thus allows a basic twofold (or v e r t i c a l ) movement 1. Tsimshian cross-clan names are an excellent example. 140 between "the unity of a m u l t i p l i c i t y [and] the d i v e r s i t y of a unity" ( i b i d . ; 136). In simpler terms, the species concept suggests l o g i c a l movement or connections between the concrete and i n d i v i d u a l on the one hand and the abstract and c a t e g o r i c a l on the other. As always, Levi-Strauss 1 argument i n The Savage Mind defies paraphrasing, and that w i l l not be attempted here. However, i t i s impor-tant to e s t a b l i s h that i t i s h i s conception of totemism as developed there that stimulated and gave d i r e c t i o n to the present inquiry. The key concept i s that "the differences between animals, which man can extract from nature and t r a n s f e r to culture ... are adopted as emblems by groups of men i n order to do away with t h e i r own resemblances" ( i b i d . : 107). At c r i t i c a l stages i n the presentation of the evidence, reference w i l l be made to Levi-Strauss' argument i n order to e s t a b l i s h the congruity of the Tsimshian system with h i s general model of totemic systems, as w e l l as to r e f e r the reader to the comparative data he has assembled. Change Change i s i m p l i c i t i n a totemic system since i t i s , i n L e v i -Strauss' words, a " l i v e d " c l a s s i f i c a t i o n , subject to the pressures of demographic determinism: Unlike other systems of c l a s s i f i c a t i o n , which are p r i m a r i l y conceived ( l i k e myths) or acted ( l i k e r i t e s ) , totemism i s a l -ways l i v e d , that i s to say, i t attaches to concrete groups and concrete i n d i v i d u a l s because i t i s an hereditary system  of c l a s s i f i c a t i o n (author's emphasis) (Levi-Strauss, 1966: 232). A very c l e a r example of demographically induced change occurred i n the l a x k * i b u 7 house of mali at Kitwancool. The house owned a grizzly-bear c r e s t , v a l i d a t e d by a variant of the Bear Husband myth. Feuding broke 141 out between mall's nephews over t h e i r wives, and a group of them moved to Kispiox to avoid overt c o n f l i c t . There was a question over who was to r e t a i n the g r i z z l y c r e s t , which was f i n a l l y solved by s p l i t t i n g i t i n two: mali kept the bear's hindquarters, while the f i s s i o n i n g group took 2 i t s front part to Kispiox (Barbeau, 1929: 111). I t can be seen on sqabE^x's totem pole, j u t t i n g out from near the bottom ( i b i d . : PI. XXI, f i g . 4). We can also assume that the crest system was expanding i n the 19th century as a d i r e c t r e s u l t of white economic input, which enabled the Tsimshian to invest more wealth i n potlatches and concomitant crest d i splays. Totem poles, f or example, grew t a l l e r and more prevalent, creating a growing demand f o r crests as motifs f o r the carvers. Both Newcombe (1907) and Deans (1891: 286) report that the Haida were contin-u a l l y short of crests f o r t h e i r totem poles, and used non-crest motifs as " f i l l e r s . " The s t o r i e s of the Raven cycle were also used on "story" poles by both the Haida and T l i n g i t (see Barbeau, 1950: 352-361), although not by the Tsimshian. The reason was the unique capacity of the Tsimshian system to continually generate new crests derived from a s i n g l e crest animal (to be explained below). Even though they were incorporating white man's houses, c l o t h -ing, and money as prestige symbols, the Tsimshian crest system remained an e x c l u s i v e l y Tsimshian symbol system and expanded according to i t s own r u l e s . Out of the hundreds of crests recorded, there are only two which 2. According to Barbeau (1929: 109), the house of mali had o r i g i n a l l y moved to Kitwancool from Kispiox as a r e s u l t of an e a r l i e r feud. 142 make any reference at a l l to the white man. These are the crests named "Mr.-Ross's-dog," and the "palisade," owned by wa'ig'Et (g'isg.a^st) and malu.laq ( l a x s e ' l ) , r e s p e c t i v e l y , both of Kisgegas. The crests were adopted during a Gitksan expedition against the Tsetsaut, dating from a f t e r 1827, and r e f e r to an outpost of Fort St. James at Bear Lake run by a Mr. Ross (see Barbeau, 1929: 103-104). This was probably the Gitksan's f i r s t encounter with the white man and, being a non-natural event, was 3 properly generative of crests. But what i s even more s i g n i f i c a n t i s that both the Coast Tsimshian g*ispwudwa'dg and g.anha'da already had a dog c r e s t , and the Gitksan g'isg.a/st already had a f o r t i f i c a t i o n c r e s t , so that a l l the references to Mr. Ross' dog and the palisade around h i s house do i s to p a r t i c u l a r i z e the crests for these two houses. History was thus embedded i n the system, continually threaten-ing to undermine i t . I t s form at any given moment was a balance of s t r u c -ture and event. The Tsimshian narratives abound with accounts of wars, i n t r i g u e s , f u g i t i v e s , migrations, murders. Reading them one has the impression of Tsimshian h i s t o r y as a continuously f i s s i o n i n g and fusing movement of people up and down the Nass and Skeena Rivers, out onto the 3. I t should be mentioned that the people of Hazelton erected a totem pole i n 1970 at 'Ksan, a reconstructed Gitksan v i l l a g e and t o u r i s t a t t r a c t i o n , with a representation of W.A.C. Bennett, the then Premier of B r i t i s h Columbia, on top. "When t h i s was c a l l e d a " s a c r i l e g e " and a " p r o s t i t u t i o n " of Indian art by an o f f i c e r of the Union of B.C. In-dian Chiefs, one of the carvers of the pole " s a i d the carving repre-sents the government — not Bennett — and symbolizes non-Indian i n -volvement i n the 'Ksan V i l l a g e c u l t u r a l project" (Vancouver Sun, August 15, 1970). In any event, the carving does not mean that Mr. Bennett or the p r o v i n c i a l government were adopted as crests. 143 i s l a n d s , and back again. What i s amazing i s that the structure survived i n the clear and vigorous form that i t did. The General Structure of the Tsimshian Crest System The most s i g n i f i c a n t s t r u c t u r a l feature of the Tsimshian crest system, and one which distinguishes i t from other Northwest Coast crest systems, i s what might be c a l l e d i t s i n t e r n a l d i f f e r e n t a t i o n , r e s u l t i n g from i t s capacity to generate a number of crests from a s i n g l e crest a n i -mal. This, i n turn, i s based upon a fundamental differ e n c e i n conceptua-l i z i n g crests between the Tsimshian and t h e i r neighbours. Among the Haida and T l i n g i t , f o r example, i t was the crest animal — the bear, k i l l e r w h a l e , raven, eagle, etc. — which was owned, and i t s various representations or manifestations were simply matters of i n d i v i d u a l and a r t i s t i c preference. Among the Tsimshian, on the other hand, i t was a s p e c i f i c manifestation or way of representing the animal which was owned. Thus, the Haida had a s i n g l e raven crest, while the Tsimshian had over a dozen: Supernatural Raven, Raven of the Sky, Raven of Copper, A l l Abalone Raven, S p l i t Raven, Raven on Top of Raven, Raven Eating Salmon L i v e r , Chief Raven, Raven Hang-ing by One Claw, Prince Raven, White Raven, Raven with S t a r f i s h i n i t s Beak, Raven's Nest, Raven S i t t i n g Quietly, Soaring Raven, Raven of the Water. I f crests were v i s u a l expressions of Tsimshian s o c i a l organiza-t i o n , there should be some s i g n i f i c a n t s t r u c t u r a l d i f f e r e n c e between Tsim-shian and H a i d a / T l i n g i t s o c i a l organizations r e l a t e d to t h e i r d i f f e r e n t kinds of crest systems. 144 The pertinent difference i s i n group ranking: the Tsimshian ranked lineages, sub-clans, and clans i n d e f i n i t e order, whereas "we f i n d that among the Haida and T l i n g i t rank order of groups i s hardly present ...." (Rosman and Rubel, 1971: 47). The Haida characterized f a m i l i e s as being of high or low rank and there were fam i l i e s known as those that "stood f i r s t " i n a v i l l a g e (Swanton, 1905: 70). S i m i l a r l y , among the T l i n g i t " c e r t a i n clans were regarded as 'high' whereas others were gener-a l l y regarded as low caste for varying reasons," although "there was no sharp d i s t i n c t i o n between commoners and nobles" (Olson, 1967: 47). In every T l i n g i t v i l l a g e the highest chief of the "leading clan" was c o n s i -dered a sort of "town chief" ( i b i d . : 49), as there was also a town chief or "town mother" among the Haida. There were, however, no ranking systems equivalent to that of the Tsimshian anongGelther. As a r e s u l t , the Haida and T l i n g i t crest systems are what might be c a l l e d s i n g l e axis systems — they express only descent group membership (although some crests were ex c l u s i v e l y owned and considered prerogatives of e s p e c i a l l y high rank), while the Tsimshian crest system i s a double axis system — i t expresses both descent group membership and rank. The Tsimshian system, thus having a greater symbolic load, needed a greater number of contrastive units to express i t . Along the f i r s t or h o r i z o n t a l axis of the Tsimshian system, and comprising those of the Haida and T l i n g i t , are the units and i n t e r v a l s representing descent groups: the d i s c o n t i n u i t i e s between bear, wolf, eagle, raven, e t c . For i t i s the differences between natural species that totemic systems u t i l i z e i n order to symbolize s o c i a l d i s t i n c t i o n s : "the homology [totemic systems] evoke i s not between s o c i a l groups and natural 145 species but between the differences which manifest themselves on the l e v e l of groups on the on=ihand and on that of species on the other. They are thus based on the postulate of a homology between two systems of d i f f e r e n c e s , one of which occurs i n nature and the other i n cul t u r e " (Le'vi-Strauss, 1966: 115, h i s emphasis). In other words, the lax s k ' i ' k d i f f e r from the laxk*ibu' j u s t as the eagle d i f f e r s from the wolf. There i s no sense of i d e n t i f i c a t i o n or " r e l a t i o n s h i p of substance" ( i b i d . : 135) between the animal species and human group: people do not think of themselves as e a g l e - l i k e or w o l f - l i k e . I f they did, argues Levi-Strauss, exogamy, by which l i n k s between groups are maintained, would be d i f f i c u l t : "the more each group t r i e s to define i t s e l f by the image which i t draws from a na t u r a l model, the more d i f f i c u l t w i l l i t become f o r i t to maintain i t s l i n k s with other s o c i a l groups and, 4 i n p a r t i c u l a r , to exchange i t s s i s t e r s and daughters with them" ( i b i d . : 117). Although each Tsimshian clan i s p r i n c i p a l l y associated by both anthropologists and Tsimshian with i t s ptEx (clan) animal, there are i n fact two animals f o r each clan which are f u n c t i o n a l l y equivalent as primary clan symbols (the g*ispflwudwa'da have two ptEx). These are the animals I have c l a s s i f i e d as "primary crest animals" i n Appendix I I , and are as f o l -lows (the ptEx animals are underlined): laxk' i b u ' g*isppwudwa'da g. anha'd g laxsk* i ' k wolf g r i z z l y raven eagle bear k i l l e r w h a l e frog beaver 4. Although the T l i n g i t c a l l e d crest animals "ancestors" and believed that people i n the descent group owning a p a r t i c u l a r animal as a crest had some s p e c i a l a f f i n i t i e s with that animal, there i s no i n d i c a t i o n that they thought of themselves as having animal c h a r a c t e r i s t i c s (see Olson, 1967: 117-118). 146 These eight animals are the primary b u i l d i n g blocks i n the crest system of the Tsimshian, as w e l l as those of the Haida and T l i n g i t . I t i s p r i -marily on the basis of correspondences of these crests that exogamic r e -l a t i o n s h i p s are maintained throughout the Northern Province. Other a n i -mal species, humans, and some plant species and natural phenomena, which would also be placed along t h i s h o r i z o n t a l a x i s , are owned by sub-clans and lineages and do not have clan-wide d i s t r i b u t i o n (excepting f o r some "secondary crest animals" of the Coast Tsimshian). These various cate-gories of crests w i l l be discussed more f u l l y below. In the s t r i c t e s t sense, these natural species are the "images drawn from nature" which constitute Tsimshian totemism. I t i s a system of the same order and kind, and includes most of the same natural spe-c i e s , as the T l i n g i t and Haida crest systems but, as ind i c a t e d above, constitutes only one axis of the more elaborate Tsimshian crest system. While d i s c o n t i n u i t i e s between forms e x i s t i n nature, rank does not. The second or v e r t i c a l axis of the Tsimshian crest system adds ano-ther dimension by which rank discriminations can be expressed. I t does th i s by applying a group of "operators" which transform the general crest' animal i n t o a s e r i e s of p a r t i c u l a r i z e d forms. The operators are a t t r i -butes that I have abstracted from the names and descriptions of p a r t i c u -l a r i z e d crests as given by Barbeau's informants (and summarized i n Appendix I I ) . They were not l i s t e d or described, nor necess a r i l y conceived of, as separate or separable a t t r i b u t e s by the Tsimshian. They are, however, quite e a s i l y i s o l a t e d . 147 Below are three pa i r s of crest names (ayuks) i n which the a n i -mals d i f f e r , but the modifiers "Prince," "Real," and "Of the Sky," are shared: •frkwE'lksam mgdi^k : Prince of G r i z z l i e s •fckwE'lksjm haCots : Prince of Cormorants s^mnE'x3: : Real K i l l e r w h a l e s mg.anaVo : Real Frog g.ag.um l a x E ' : Raven of the Sky madi.ggm l a x E ' : G r i z z l y of the Sky The operators are p a r t i c u l a r i z i n g a t t r i b u t e s described or implied by the modifiers contained i n the crest names. In most cases, the modifier i s i n fact a de s c r i p t i o n of the a t t r i b u t e , as i n Standing Bear, S p l i t Person, White Marten, G r i z z l y of the Sea ( i . e . , with f i n ) . In other cases, how-ever, the l o g i c connecting the crest name and i t s operator i s more obscure. The l a x k * i b u / bear crest Without Knowledge, f or example, refe r s to the operator "young." A bear cub, of course, i s "without knowledge" r e l a t i v e to the adult. Further, when we know that t h i s c rest i s v a l i d a t e d by the Bear Husband myth, i n which a woman marries a bear and has h a l f -human, half-bear chi l d r e n , who are unable to adapt to human society and who therefore return to the world of the bears, we can perhaps i n f e r that Without Knowledge i s a reference to the i n a b i l i t y of the bear cubs to ac-quire human ( i . e . , c u l t u r a l ) knowledge. Without Knowledge i s a bear crest only. Two more general crest names are Prince of (Animal Species) arid Supernatural (Animal Species) , which have i n t e r e s t i n g l o g i c a l connections with the operators "abalone" and "human faces." The word translated by Tsimshian people as "prince" i s lkuwE/_ks3k. Now, the p r e f i x Iku- also r e f e r s to "bright and s i l v e r y young 148 salmon." Crests named Prince of (Animal Species) were consistently des-cribed as being decorated with abalone, leading to the discovery of a logical association of the irridescence of abalone pearl with the similar irridescence of the young salmon's skin. IkuwE^ksdk i s , moreover, the t i t l e applied to the sons of a chief, who are as young and promising of wealth and plenty as the young salmon on i t s way down the river to the sea. We might carry the association even further by noting that just as the young salmon i s only temporarily in the river moving downstream, the young prince i s only temporarily resident in the house of his father, and that both w i l l change directions when adult, the salmon to return upstream to spawn, the young prince to succeed his uncle as chief of his own matrilineage. Both are, in addition, returning to places of origin: the salmon to the stream where i t was spawned, the prince to his matri-lineage. The association of the operator "human faces" with crests named Supernatural (Animal Species) would seem to refer to the beliefs and practices explored at the end of the last chapter. The word for supernatural being or s p i r i t is naxna /x. Its association here with human faces (which are added to or represented on the crest animal) might refer to similar aspects or kinds of spi r i t s as the human face masks used to dramatize naxno'x or s p i r i t names. There is a marked tendency for the operators to be alternatives from a series of contrast sets, so that they not only define particularized crests but contrast them with others from the same set. Some of the more' widely distributed sets of operators are as follows: 149 1. colour (black, white, red) s p e c i f i e d ; colour not s p e c i f i e d . 2. whole, s p l i t ( v e r t i c a l , horizontal) 3. with abalone, without abalone 4. adult, young 5. (of the)water, land, mountains, sky a. with f i n s , without f i n s b. with wings, without wings 6. head (ears, muzzle, teeth), body ( t a i l , paws, b e l l y ) 7. standing, s i t t i n g 8. with copper, without copper 9. s i n g l e form, multiple forms 10. with human faces, without human faces 11. headdress (head, whole animal), robe (skin, whole animal) 12. s p e c i f i e d numbers of parts (one, two, four, ten ...) In the table on the following page,a serie s of f i v e operators (abalone, s p l i t , young, human faces, white) has been applied to the eight primary crest animals i n order to produce twenty of the p a r t i c u l a r i z e d named crests reported by Barbeau 1s teachers. In the terminology of eth-noscience, such arrangements or " c l a s s i f i c a t i o n events" are c a l l e d para-digms, which are defined as "multidimensional forms of arrangement or-ganized by class i n t e r s e c t i o n " (Conklin, 1969: 107). Perfect paradigms ( i . e . , those i n which a l l of the spaces representing possible combinations of components are f i l l e d ) are rare i n f o l k taxonomies. I s o l a t i o n of the operators permits the i d e n t i f i c a t i o n of v i s u a l representations of the same animal-and-attribute r e l a t i o n s h i p s i n material cul t u r e . A carving or painting of a g r i z z l y with a f i n can now be i n -terpreted with reasonable and s a t i s f y i n g c e r t a i n t y as the crest named P a r t i c u l a r i z e d Crests OPERATORS abalone s p l i t young human faces white PRIMARY CREST ANIMALS Prince of Prince of Prince of Prince of Prince of Prince of Prince of Wolves Bears G r i z z l i e s Killerwhales Ravens Eagles- Beavers S p l i t S p l i t S p l i t Bear Killerwhale Eagle Without Children of Knowledge Chief Frog Supernatural Supernatural Supernatural Supernatural G r i z z l y Frog Raven Beaver White White White White Wolf Bear G r i z z l y Raven Wolf Bear G r i z z l y Killerwhale Frog Raven Eagle Beaver Table I I I . I n t e r s e c t i o n of Operators and Primary Crest Animals to Form P a r t i c u l a r i z e d Crests 151 G r i z z l y of the Sea. Although there are s t i l l many gaps and ambiguities to be faced i n working out such correspondences, the crest l i s t s i n Appendix II are now a v a i l a b l e as r i c h primary data upon which r e f i n e d iconographic i n t e r p r e t a t i o n s can be based. The l i s t s themselves can no longer be s u b s t a n t i a l l y improved. However, as more crest representa-tions i n museums are i d e n t i f i e d and brought i n t o t h i s framework, we should be able to analyse, rather than merely i d e n t i f y , the r e l a t i o n -ships between the semantic and v i s u a l systems. Such an analysis might well follow the model developed by l i n g u i s t William Watt, i n that the crest names and t h e i r v i s u a l repre-sentations seem to be s i m i l a r i n nature to the cattlebrand and blazon systems which he has so elegantly described (Watt, 1966, 1967). The s i m i l a r i t y was i n fa c t noticed by .,Watt, although he was r e f e r r i n g to European "totemic" or h e r a l d i c systems: ... the i d e n t i f i c a t i o n and categorization of elements [ i n the v i s u a l or i c o n i c system] i s aided by the brands' asso-ciated "names," or "blazons" ... which by t h e i r nature spot-l i g h t the p r i m i t i v e s and some of the rules of augmentation and combination. In f a c t , the brand-and-blazon system may be s i m i l a r i n t h i s respect only to one other system: that of the h e r a l d i c arms ("coats of arms") (Watt, 1967: 22). Watt sees the r e l a t i o n s h i p between the i c o n i c or v i s u a l system — the cattlebrands themselves — and the verbal system — t h e i r blazons — as so close or congruent that "a quite simple algorithm can be devised for t r a n s l a t i n g from one to the other" ( i b i d . : 25). Elsewhere, he says that the cattlebrands and t h e i r blazons are "covariant" systems: "to vary the p i c t u r e i s to vary the d e s c r i p t i o n , and vice-versa" (1966: 15). Tsimshian crest names (ayuks) and crest representations (dzEpk) are c l e a r l y also covariant systems. Watt reports (1966: 16) that he and 152 R.W. Hsu are working on a grammar of heraldry which, when i t i s a v a i l -able, should suggest methodological leads f o r a possible grammar of Tsimshian c r e s t s . The major burden of t h i s and the following chapter w i l l be to show that the Tsimshian were using the axis of d i f f e r e n t i a t i o n pro-duced by the a p p l i c a t i o n of operators to natural species i n order to express differences i n rank. I t w i l l also be argued that the develop-ment of symbols of rank was r e l a t e d to an expansion of power and hege-mony by the c h i e f s ; more s p e c i f i c a l l y , to t h e i r assuming a new kind of power base i n the r o l e of w i h a l a ^ i t or Great Dancer. The argument w i l l , then, seek to enlarge Levi-Strauss' model of a totemic system by showing how the Tsimshian were transforming totemism i n t o an enlarged symbol system, one capable of expressing new and non-kin-based forms of s o c i a l r e l a t i o n s h i p s as w e l l as the structure of descent groups. Since t h i s transformation was incomplete and i n process, i t w i l l be necessary to examine Tsimshian crests i n considerable d e t a i l i n order to i s o l a t e new forms and tendencies i n the system. Categories of Crests In Appendix I I , I have divided Tsimshian crests into seven categories: primary animal, secondary animal, human, monster, plant, n a t u r a l phenomenon, and a r t i f a c t . While these are my own categories,-, and do not represent d i s t i n c t i o n s made by the Tsimshian, the crests i n them are s i g n i f i c a n t l y d i f f e r e n t . Primary animal c r e s t s : The two primary animals of each clan were l i s t e d on page 145 above, where they were described as the basic 153 b u i l d i n g blocks of the Tsimshian crest system. As such,they share a number of q u a l i t i e s that set them apart from the other categories of c r e s t s : 1) they are the p r i n c i p a l clan i d e n t i f i c a t i o n symbols. In those cases of lineages or sub-clans which do not have the r i g h t to use the ptEx animal, the other primary animal functions to i d e n t i f y clan membership; 2) they are each the source of multiple p a r t i c u l a r i z e d c r e s t s , f a r more than any other animals i n the system; 3) they can be displayed i n t h e i r general ( i . e . , non-particular) form by clan members without a v a l i d a t i n g myth or d i s t r i b u t i o n of wealth; indeed, there do not seem to be myths accounting f o r the o r i g i n s of the primary animals as general cre s t s . V a l i d a t i n g myths were, however, reported and ob-vi o u s l y required to display them as p a r t i c u l a r i z e d c r e s t s . In most of these c h a r a c t e r i s t i c s , the primary animals d i f f e r markedly from a l l other categories of crests,and I have concluded that these eight were the o r i g i n a l crest animals of the Tsimshian, and probably of the T l i n g i t and Haida as w e l l . This conclusion, i n part, follows a suggestion of Sapir's that crests can be s t r a t i f i e d by f r e -quency of occurrence: "the older the cr e s t , the greater number of times 5 i t i s found i n the various clans; on the other hand, a crest found i n only one clan may be suspected to be of recent o r i g i n " (Sapir, 1966: 44). Secondary animal c r e s t s : The secondary animals are a l l other animals used by the Tsimshian as crests. Except f or a serie s of eight secondary animals of the Coast Tsimshian, they were a l l claimed by l i n -eages and sub-clans rather than having clan-wide d i s t r i b u t i o n . These 5. Sapir's " c l a n " would correspond to my sub-clan. 154 Coast Tsimshian exceptions, which are also found i n l i m i t e d d i s t r i b u t i o n i n the other two d i v i s i o n s , are the following: laxk* i b u ' g'ispgwudwa'd? g. anha 'd? l a x s k * i ' k crane grouse sc u l p i n h a l i b u t mosquito owl s t a r f i s h octopus The eight secondary animals were s a i d s p e c i f i c a l l y to have been used by a l l (crane, mosquito, grouse, owl, halibut) or many ( s t a r f i s h , s c u l p i n , octopus) houses i n a clan, and seem to have functioned as minor clan symbols. Two of them, the octopus and s c u l p i n , were also sources of multiple p a r t i c u l a r i z e d c r e s t s , and resemble primary animals i n t h i s respect. I t i s also probable that general forms of these secondary animals could be represented without v a l i d a t i n g myths. A l l other secondary animal crests were claimed by houses or sub-clans only and required v a l i d a t i n g myths to be displayed. In most cases, they were reported i n one form only. Human cr e s t s : A b a f f l i n g category of crests are those de-r i v i n g from the human being, forms of which were claimed by houses i n a l l four clans i n a l l three d i v i s i o n s of the Tsimshian. At l e a s t three i d e n t i c a l human crests were claimed by d i f f e r e n t clans: the Two-Headed Man by the laxk*ibu', laxsk* i'k, and g'ispawudwa'd3; the Whole Being by the laxsk*i 7.k and g.anha'dg; and the Robe of Scalps by the laxk*ibu 7 and g.anha'dg. This, of course, runs counter to the basic p r i n c i p l e of totemic systems: that differences or d i s c o n t i n u i t i e s between n a t u r a l  species are used to characterize differences between human groups. Use of the human as crests by a l l four clans obviously i n v a l i d a t e d crests of t h i s category from functioning as descent group symbols l i k e the 155 c l e a r l y d i f f e r e n t i a t e d animal cre s t s . Further, since human crests were claimed by houses of a l l ranks, they could not have functioned as symbols of rank. They were, i n f a c t , so widely d i s t r i b u t e d through-out the crest system as to perhaps be considered some kind of pan-Tsimshian symbol. A p a r t i a l explanation f o r the widespread d i s t r i b u t i o n of human crests was suggested to me by the crest representations on Gitksan totem poles (see plates i n Barbeau, 1929). The mixing on the great majority of these poles of both animal and human f i g u r e s , rather than running counter to the basic totemic idea, seems instead to support i t . I f we exclude f o r the moment the crest iconography of the human representations, and regard them only as the generalized humans they are to the eye — human representations characterized by a bland and almost boring sameness — they can then only be distinguished by t h e i r r e l a t i o n s h i p s to d i f f e r e n t sets of animal f i g u r e s . In other words, the humans on the poles have l i t t l e or no difference or d i s t i n c t i o n i n the absence of the totemic animals with which they are associated. Simi-l a r l y , according to the totemic p r i n c i p l e , the sameness of human beings i n Tsimshian society was d i f f e r e n t i a t e d by the association of people and descent groups with d i f f e r e n t animal species. This explanation i s supported by the fac t that, indeed, some of the human crests do r e f e r to s l a i n emenies or p a r t i c u l a r ancestors of the descent group, who are mentioned i n the narratives as r e a l ( h i s -t o r i c a l ) people. The majority, however, are more properly described as supernatural beings i n human form who were encountered i n myth-time and assumed as crests i n the same fashion as animal cre s t s . I can o f f e r 156 no further explanation f o r t h e i r ambiguity- as descent group symbols i n an otherwise c l e a r l y f u n c t i o n a l totemic system. I do, however, as-sume the ambiguity to be a fu n c t i o n a l one, since other ambiguities i n the system, such as the monster crests discussed below, can be shown to express features of Tsimshian s o c i a l organization. Monster cre s t s : An unexpectedly large number of crests f a l l i n t o the monster category, defined as crests which combine the a t t r i -butes of two or more animals, including human beings. The monster crests can be further separated i n t o two sub-categories: simple and complex. Simple monsters are those i n which the b a s i c animal i s c l e a r l y i d e n t i f i a b l e as a primary, secondary, or human crest animal, such as Supernatural Sculpin, Mountain Goat of the Sea, S t a r f i s h Person, F l y i n g Frog, or G r i z z l y of the Sea. Complex monsters are those based on some sort of "problematical" or composite animal form not found i n nature. I t i s l i k e l y that crests of t h i s sub-category represent a further dev-elopment of the basic monster idea responsible for crests of the f i r s t sub-category. They are most often described as b i r d - l i k e creatures, often with recurved beaks, which are subject to transformations i n t o other forms, such as sea monsters and human-like creatures. S i g n i f i -c a n t l y, Tsimshian people often had d i f f i c u l t y i n providing English glosses for the native terms, r e s o r t i n g to "hawk-like" or " l i k e an eagle, but not an eagle," confirming the suggestion that there are no n a t u r a l prototypes f o r these creatures. Descriptions of important ( i . e . , widespread) complex monster crests are l i s t e d below. They are composite descriptions based on the data reported i n Appendix I I . References are made i n parentheses to a 157 ser i e s of plates which i l l u s t r a t e p ossible material representations of some of the complex monsters. I t should be c l e a r from the verbal d e s c r i p t i o n s , however, that t h e i r most important shared c h a r a c t e r i s t i c i s that they can be transformed from one form i n t o another. I t i s therefore i n v i o l a t i o n of t h e i r very nature, and perhaps t h e i r function, to suggest that t h e i r forms had congealed or s o l i d i f i e d i n t o only those shown i n the p l a t e s . One of the major weaknesses of Barbeau's monograph on Gitksan totem poles (1929) and the Coast Tsimshian and Niska sections of h i s broader totem pole survey (1950), i s that he attempted to freeze these complex monsters i n t o constant forms. Hence, the sgmg*i'k, f o r example, became a k i n g f i s h e r or a woodpecker. Also, when the Tsimshian could not i d e n t i f y one of these forms on a totem pole, Barbeau would often l i s t what were obviously h i s own guesses as f i r m a t t r i b u t i o n s . To h i s c r e d i t , he did l i s t a large number of the Tsimshian people's alternate or contradictory i d e n t i f i c a t i o n s i n footnotes, thus recording f o r us t h e i r own lack of agreement about the monsters' phy s i c a l c h a r a c t e r i s -t i c s . Regrettably though, f a r too many of h i s a t t r i b u t i o n s are ques-tionable for v a l i d conclusions to be based on h i s published i d e n t i f i c a -tions of totem pole f i g u r e s . They should be used as guides only. Complex monsters: g'i'balk: '(Plate 70a, b,c). Said to be a large monster with a head l i k e an eagle and a large f i n protruding from i t s back; also s a i d to have wings and human forms around i t s face and on i t s back; said to be r e -l a t e d to the naxn^gom g'i'k (Supernatural Mosquito) ; also said to rer-158 semble a human being, laxsk* i'.k. (A g* isp^wudwa'da g* i'bglk was said to be l i k e a large eagle.) log.3m g'i'balk: (Rotten g ' i ' b a l k ) . Said to be a winged creature with human faces on i t ; i n the myth i t was claimed by the l a x k * i b u ' i n a rotten or decayed condition a f t e r i t was taken as a crest by the laxsk * i ' k . laxk* ibu'. samg'i'k: (Plate 71a,b,c). A supernatural b i r d , sometimes sa i d to be a c woodpecker (Barbeau also c a l l e d i t a k i n g f i s h e r ) ; i t has a long, s t r a i g h t , pointed beak. In an important g.anha'dg myth, two s^mg'i'k's, one large and one small, are associated and the large one has a human on i t s back, g.anha'd9; also claimed by la x s k ' i ' k . g ' i l ^ a ' d a l : A b i r d l i k e an eagle with many small human beings on i t s hea'd; also s a i d to be l i k e a raven, and to make great noises l i k e a thunderbird; the same creature, under the name of g'e'mgxgm was., sai d to be l i k e the xske'msam with a crooked beak (see xske'ms?m i n Plate 72) . g. anha' dg. xske'msam: (Plate 72a,b,c). A b i r d l i k e an eagle but with a more re-curved beak; also s a i d to be an " e x t i n c t " b i r d l i k e an eagle; said by one person not to be a hawk, g'ispgwudwa'dfl. Also claimed by laxsk*i.'k and laxk* i b u / . mjdzgks: (Plate 73a,b). A b i r d usually s a i d to be l i k e a hawk or 6 chicken hawk with a curved beak; also s a i d to be represented i n 6. The hawk i s not a Tsimshian crest animal. I am therefore suspicious of the great many "hawk" i d e n t i f i c a t i o n s given to Tsimshian pieces i n published catalogues and museum records. The recurved beak, said by Boas (1955: 190) and others to be a hawk a t t r i b u t e , i s an at t r i b u t e of the myizftks and other complex b i r d monsters i n Tsim-shian iconography. I t may indeed be a hawk a t t r i b u t e f o r other groups. 159 human form; also s a i d to be an eagle under a d i f f e r e n t name (be-cause the eagle was a crest of another clan). g.anha /d3. Thunderbirds (by various names) (Plate 74a,b): tsag.ao'xlp: (Hooked Nose). A b i r d with a recurved beak, with c h a r a c t e r i s t i c s of thunderbird; also s a i d to be a human with long, hooked, recurved nose (see human monster of same name below). g* ispgwudwa'd^. g.algbli'bam l a x E y ; (Thunder of the A i r ) . A thunderbird with long, curved beak, s a i d to resemble both an eagle and the xske'msam. g * isppwudwa 7dg; also g.anha'dg. lax*j'm; (On Top). A supernatural b i r d with a long, s t r a i g h t beak; also a human with a long nose; also s a i d to have a re-curved beak, g* ispgwudwa'dg. gwasdzEdEmti: (Lightning Robe)-?, A huge thunderbird with g r i z z l y feet, g'ispawudwa'd?. A g.anha'da Lightning Robe was-said to be a huge b i r d with a raven i n each wing and to have c h i l -dren on i t s back. x t s i . tiyE^'tux: (Thunder) . A b i r d with curved beak; also a winged person, g* i s g . a / s t . ha ^clli'': (To Cause a S l i d e ) . A large b i r d with a long, hooked, recurved nose, g'ispawudwa'dg. t^sag.ao'xlo: (Hooked Nose, Barbeau's Glass Nose) (Plate 75). A human monster with a large b e l l y (also c a l l e d l a ^ a s p a ' n c or Large Belly) and sharp or g l a s s - l i k e nose, laxk* ibu /. mag.amba'laq, or madzilu/lgq; (Moth; -lu. h j q means "corpse"). A l i t t l e b i r d with a long recurved beak, laxk*ibu. 160 lag.jmba/lag: (Decayed Corpse). (Plate 76a,b,c). Said to be beaked l i k e a b i r d or to be represented as a human corpse, g* ispgwudwa'da. g'Edamt^sam sqanifs: (Man of the Mountains). A short human being with wings, with a nose about a foot long, g'ispjwudwa' dg. hagwalox; (Plate 77). A sea monster, usually r e f e r r e d to as a monster k i l l e r w h a l e , but also said to have aspects of g r i z z l y . I t s f i n was sa i d to be a dangerous snag, g*ispgwudwa'da. A la x s k ' i ^ k  hagwalox was said to have humans along i t s back; a laxk*ibu' house claimed a hagwalax of the lake, s a i d to be a large box f u l l of human beings, decorated with human h a i r and a f i n on i t s back, which swam as though a l i v e . wilmi'c: (Where Spawns). A g r i z z l y - l i k e monster shown with c h i l d r e n on i t s back; also said to be a whale-like creature with the eyes of i t s young protruding from along i t s backbone; also said to be small humans "spawning." g'ispawudwa'da . max-takpi/l; (Over Ten) a. A monster with the head of a scu l p i n with 10 human beings on i t ; also said to be a raven with 10 human heads i n a row over i t s head; also s a i d to be a supernatural whale (hagwgl^x?) with 10 l i t t l e forms of humans; also said to be a large human with 10 small humans, g.anha'dd. b. A large monster, s a i d to be a hagwglax, also s a i d to have a horned beaver head; the name Over Ten refers to 10 eagles s i t -t i n g a l l over i t or one large eagle s i t t i n g on i t s head with 9 others along i t s back; i n another form Over Ten was 10 abalone s h e l l s worn i n a row from forehead over the head, laxsk'i'k. 161 At f i r s t , i n addition to the echoes of primary and secondary crest animals ( s c u l p i n , raven, g r i z z l y , e t c . ) , these monsters present a bewildering complex of parts and transformations: one b i r d becomes another, b i r d s become men, moths become corpses. There are patterns to be discerned, however, which seem to r e l a t e to a ce n t r a l theme of cannibalism. The u n i f y i n g clues come from several myths. That d i f f e r e n t clans own the myths i s probably not s i g n i f i c a n t , since most of these monsters were claimed by more than one clan, a l b e i t at times under d i f f e r e n t names. Barbeau (1929: 93) gives a symopsis of a myth owned by sev-e r a l related g * i s g . a f s t houses. The people were st a r v i n g , l i v i n g on roots, b e r r i e s , and salmon bones, when they f i n a l l y caught two mountain goats, "on the bodies of which they saw a ghost-like monster, with a beak almost l i k e a b i r d ' s , the Moth, feeding greedily. They s a i d , 'It must be the ghost of one of our dead r e l a t i v e s partaking of food." They k i l l e d the monster and took i t as a crest — the lpgomba^laq (or l^g.-?mlu/la q) , or Decayed Corpse. On one totem pole (Plate 76b) , i t i s shown as a human figu r e i n a flexed p o s i t i o n , ready f o r b u r i a l or 7 cremation. On another pole (Plate 76a), i t appears as a human fi g u r e with an opening f o r a p r o j e c t i n g beak or nose that has f a l l e n o f f . These two totem pole representations confirm that both aspects of the monster — moth and corpse — are part of the crest. S i g n i f i c a n t l y , i t i s both eater (moth; ghost r e l a t i v e ) and eaten (corpse). 7. "The body was then doubled up, the chin r e s t i n g on the knees, hands folded over the chest" ( G a r f i e l d , 1939: 239). 162 The crests of the g*lib"3Ik and Rotten or Decayed (1> g.3 m-) g*iCb3lk again suggests a corpse motif. The g'i^bglk, a complex mon-st e r with aspects of b i r d , sea monster, and human, was sa i d to be re-la t e d to the naxn^g.om g'i^k, or Supernatural Mosquito. The mosquito, i n turn, i s somewhat rela t e d to the samg'i'k ( l i t e r a l l y , " r e a l mosquito": sgm-: " r e a l , " i n the sense of high or noble; -g'i'k: "mosquito"). In a l a x s e ' l myth, a missing boy was taken by a samg'i'k and l a t e r the people saw him s i t t i n g on i t s back i n a nest. In one version of t h i s myth, there were many frogs and insects around the tree; i n other ver-sions they f i r s t noticed the boy because of human excrement at the base of the tree. The boy's stomach burst a f t e r he was rescued and h i s body was taken back to the v i l l a g e . Once again, the myth associates a supernatural b i r d (or mos-quito) with a corpse, which had died from overeating. The next myth weaves these themes together i n a more e x p l i -c i t l y human context. I t i s the l a x k ' i b u 7 myth of t*sag. a'ox-tj or Hooked Nose (t^sag.a 'o: "nose"; - x l ^ : " r e t r i e v i n g hook"). He was a human monster (although i n g'ispfrwudwa'dg crest l i s t s the creature of th i s name i s a thunderbird with a recurved beak) with a b i g b e l l y ( c f . the corpse above) whose nose was long and sharp, l i k e glass (see P l a t e 75). With h i s nose, he k i l l e d c hildren by s p l i t t i n g them down the middle, as people do salmon, and hung them to dry ( i . e . , the chi l d r e n become food). He then turns i n t o a woman who i s t r i c k e d by some sur-v i v i n g c h i l d r e n i n t o swallowing hot stones ("false" food). "Her nose came out to a great length, her stomach exploded, and she died" ( c f . again, the boy on the samg'i^k). "As the l a s t spark rose, they heard a 163 voice say, 'You people w i l l always s u f f e r from my nose.' That i s the o r i g i n of mosquitos" (Mr. and Mrs. Isaac Tens). The monster's ashes became the f i r s t mosquitos. In another version of t h i s myth (Barbeau, 1950: 262-264), the Chief of the Sky was offended by the chil d r e n and i t was h i s totem pole which k i l l e d them and s p l i t them open. "This pole was very br i g h t and had a long shining nose, l i k e a very sharp k n i f e . The glassy nose moved up and down, s p l i t t i n g the body of the boy open. Women walked out of the house, took the body, and spreading i t open as they do with a salmon, cleaned i t and hung i t up to dry i n the a i r " ( i b i d . : 263). The bird-man, beak-nose-belly, corpse-food associations i n these myths need not be elaborated. I t i s perhaps not going too f a r a f i e l d to mention that T'sonoqua, the Cannibal Women of the Kwakiutl, also eats chi l d r e n , and that the Cannibal-at-the-North-End-of-the-World i s the source of mosquitos (Boas, 1897: 401), as i s a Cannibal giant of the T l i n g i t (Barbeau, 1950: 378). The l o g i c of these associations i s complete when we r e a l i z e that mosquitos also "eat" people. One other a t t r i b u t e of the monster crests remains to be d i s -cussed — the frequent appearance of small human forms or faces on t h e i r heads, backs, and wings. This, I think, i s an iconographic symbol of th e i r supernatural q u a l i t y . I t was noted above that the operator "human faces" was consistently associated with crests named Supernatural (Animal Species). The small humans and human faces on the complex mon-sters seem to be of the same type. Also, many of the monsters were ex-p l i c i t l y said to be supernatural. 164 Why monsters? There were s t i l l many unused animal forms i n the natural environment of the Tsimshian with which they might have extended the h o r i z o n t a l axis of the crest system. Instead, they chose to a c t u a l l y b l u r the d i s t i n c t i o n s between the animals already used by t r a n s f e r r i n g a t t r i b u t e s between them. Some kind of reverse process to the o r i g i n a l development of totemic d i s t i n c t i o n s was taking place, one which was creating crests of a new order. An explanation l i e s i n what Levi-Strauss :would c a l l an up-ward l o g i c a l movement from the species notion or l e v e l to an environ-mental l e v e l . I f we examine the crests of the simple monster category, we see that many of the transforming features (operators) added to the o r i g i n a l crest animal are a t t r i b u t e s which s i g n a l a movement into a new and unnatural environment. So, a g r i z z l y (land) acquires a f i n and moves in t o the sea, a frog (land, lake) acquires wings and takes to the a i r , and a mountain goat (land, mountains) assumes a f i n and swims i n the sea. The new or transformed creature then becomes more symbolic of an environment (land, sea, a i r ) than a natural species. The mind, on apprehending a winged frog, immediately t r i e s to grapple with the presence of the wing and i t s s i g n a l l i n g of f l i g h t and a i r . According to V i c t o r Turner (1964: 14), the mental process involved i n the creation of monsters can be explained by William James' "law of d i s s o c i a t i o n , " which he states as follows: "when a. and b_ have occurred as parts of the same t o t a l object, without being discriminated, the occurrence of one of these, _a, i n a new combination ax, favors the * disc r i m i n a t i o n of a., b_, and x from one another. As James himself puts i t , 'What i s associated now with one thing and now with another, tends 165 to become dissociated from e i t h e r , and to grow into an object of ab-s t r a c t contemplation by the mind." Turner then relates t h i s e x p l i c i t l y to the creation of monsters: "elements are withdrawn from t h e i r usual settings and combined with one another i n a t o t a l l y unique configura-t i o n , the monster or dragon. Monsters s t a r t l e people i n t o thinking about objects, persons, r e l a t i o n s h i p s , and features of t h e i r environ-ment" ( l o c . c i t . ) . The " d i s s o c i a t e d " elements of f i n , wings, beak, ears, etc., thus become separate components of the Tsimshian r e a l i t y which s t a r t l e or encourage abstract thought by t h e i r recombination i n t o monstrous patterns. What must next be understood i s the sense o f the recombina-tion s . I do not think that the purpose of the d i s s o c i a t e d element i s to encourage thought about i t s e l f , or that a f i n on a bear only s i g n a l s i t s associated environment of water. What i t draws attention to are the r e l a t i o n s h i p s involved i n the new pattern: the observer i s stimulated to speculate about the f u l l range of meaning of an undersea bear. This p a r t i c u l a r monster, f o r example, i s a mediation of the deep and pervasive dichotomy i n Tsimshian culture between land and sea. I t i s stated quite e x p l i c i t l y by the Tsimshian themselves. In t h e i r t r a d i t i o n a l h i s t o r y , many people l i v e d o r i g i n a l l y i n the large and r i c h v i l l a g e of t*a mlax*am, said to be i n the i n t e r i o r . A f t e r an offense to the animal s p i r i t s , the vil]qge was struck by a disastrous but " l o c a l " b l i z z a r d . When the people f i n a l l y r e a l i z e d that the snow and s t a r v a t i o n were confined to the environs of t ^ mlax'am, they l e f t to seek new homes. According to one version of the t r a d i t i o n (from informant Swanson), the 166 crests the g*ispgwudwa'da group took with them when they l e f t were the following: G r i z z l y with Abalone on i t s Breast, Mountain Goat Hat, Red Leggings, Red Leather Armour, Hat of G r i z z l y Paws, Hat of Freshwater Duck, Groundhog Robe, Lynx Robe, Red Fox Robe, and Mink Robe — i . e . , a l l representing animals found i n the i n t e r i o r . Other people s p e c i f i e d that the G r i z z l y was the "main crest" and the " o r i g i n a l c r e s t " of the t ;a mlax^am g'ispawudwa' dd. The people l e f t t'} mlax 'am and traveled down the Skeena R i -ver to the coast, where they amalgamated with other groups of g*ispawudwa'da. These people had the k i l l e r w h a l e as t h e i r main cr e s t , and the two groups exchanged crests as symbols of t h e i r newly discovered relatedness. According to several Coast Tsimshian people, t h i s i s why the g'ispgwudwa'dg now have two ptEx, the Killerwhale and the G r i z z l y . Neither the amal-gamation nor the exchange of crests was t o t a l , however, and the two groups retained symbols and consciousness of t h e i r differences. The seacoast people transformed the G r i z z l y crest i n t o G r i z z l y of the Sea, and the t*amlax>am people adopted the K i l l e r w h a l e as K i l l e r w h a l e of the H i l l s (or the Lake). "Then the people applied t h e i r own Killerwhale of the H i l l s i n the same way as the G r i z z l y of the Sea i s applied" (Bradley). "The r e l a t i o n s h i p of the g'itnagunaks to the g'itVmlax 'am i s c a l l e d lekswulE'isk ("strange relation,?" or "stranger r e l a t e d " ) . They don't use the G r i z z l y , but only G r i z z l y of the Sea." Monster crests l i k e the G r i z z l y of the Sea, then, become crests  of i n t e g r a t i o n , rather than crests of d i f f e r e n t i a t i o n , as those of L e v i -Strauss' totemic model might be c a l l e d . The recombination i n t o monster crests of diss o c i a t e d a t t r i b u t e s expresses the s o c i a l i n t e g r a t i o n of 167 formerly separate groups. This i s supported by Barbeau's (1917a) analysis of Tsimshian n a r r a t i v e s , which t e l l a d e t a i l e d h i s t o r y of con-federation and amalgamation of l o c a l groups. "The clans [sub-clans] i n each phratry [clan] are e i t h e r of more or l e s s remote foreign o r i -g i n, or ancient l o c a l bodies. Their respective t r a d i t i o n s , myths, cre s t s , p r i v i l e g e s , and duties a l l tend to perpetuate the memory of t h e i r independent o r i g i n " (Barbeau, 1917a: 405). Monster crests help to bind these conglomerate groups together, by providing symbolic s t a t e -ments which combine t h e i r disparate o r i g i n s into new unity. Complex monster crests are also crests of i n t e g r a t i o n , but at the t r i b a l rather than the clan l e v e l . This idea w i l l be developed i n the next chapter, but can be b r i e f l y stated here. The growing hege-mony and power of the Tsimshian c h i e f s , most noticeable i n the o f f i c e of t r i b a l chief of the Coast Tsimshian, was incompatible with t r a d i -t i o n a l clan organization and i t s resultant clan jealousies and r i v a l -r i e s . These would only have been exacerbated had a t r i b a l chief or an ambitious clan chief displayed clan crests — crests of d i f f e r e n t i a -t i o n — with the t r a d i t i o n a l f l a u n t i n g and pomp. However, a monster crest of the complex category, one which could be represented i n a v a r i e t y of ambiguous forms, could be displayed and g l o r i f i e d by the chief without offense to the other clans of h i s t r i b e . Plant c r e s t s : Plants form a very small category of c r e s t s , of l i m i t e d d i s t r i b u t i o n . Even among the Gitksan, whose clan name g'isg.a/st translates as "people of the fireweed," a Fireweed crest i s claimed by only two houses. The most important plant crests are a 168 se r i e s of seaweed crests claimed by houses of the Coast Tsimshian g*ispawudwaMg. Seaweed crests are not claimed by any other clan or d i v i s i o n . Natural phenomenon cr e s t s : This i s also a r e l a t i v e l y small category of c r e s t s , although i t includes some very important ones-.. Most of the crests are sky phenomena: moon, sun, s t a r s , Big Dipper, r a i n , clouds, snow, rainbow, mirage, thunder ( b i r d ) , hole i n the sky, l i g h t , and red sky of morning. There are also water c r e s t s : w h i r l -pool and r i p t i d e ; and v a r i e t i e s of f i r e s . One of the more i n t e r e s t i n g aspects of the natural phenomenon crests i s that many of them are s a i d to be represented by forms of the human being or the human face. The most s i g n i f i c a n t natural phenomenon crests are those claimed by a sub-clan of the g*ispflwudwa'd3 (Barbeau's Sky Clan) and v a l i d a t e d by the well-known and widespread gau'S myth (see summaries i n Boas, 1916: 850-855). I t i s too long and complicated a myth to r e l a t e here, but i t includes as a c e n t r a l episode a young woman's being taken to the sky by the shining son of the Sky Chief, by whom she has several c h i l d r e n , who return to earth with crests given to them by t h e i r grandfather. The number of children and the s p e c i f i c crests vary, but they are always sky phenomenon, usually sun, moon, s t a r s , and rainbow, and sometimes a form of thunderbird. These crests m a t e r i a l i z e on earth as l i v i n g house front paintings on (usually) four large shining houses. A l l four clans claim the moon, but only those g*isp^wudwa'd& who have the above myth seem to have a proper claim to i t . The others treat i t as a naxno'x. 169 A r t i f a c t c r e s t s : A r t i f a c t s claimed as crests are l a r g e l y the same types -of a r t i f a c t s on which the other crests are said to be rep-resented: items of costume and personal decoration, some weapons, house parts, house types, and house names, and a s c a t t e r i n g of la d l e s and feast dishes. One copper and several supernatural canoes are l i s t e d , although these are not mentioned as crest-bearing objects i n other contexts. An important a r t i f a c t claimed by houses i n three clans i s the woven spruce root hat topped with woven discs or rings — the langmg.Eit. It s crest aspect i s the number of discs i t s owner i s permitted to show, which varies from three to nineteen according to the l i s t s . The grea-ter the number of d i s c s , the more prestigious the hat. In the l i t e r a -ture, such discs have been said to r e f l e c t the number of potlatches the wearer has given, but there i s no confirmation of t h i s i n the Tsimshian case. The number i s part of the c r e s t , an i n h e r i t e d p r i v i l i ege. Most a r t i f a c t crests are an instance of the crest system running away with i t s e l f . Items once prestigious because of the crests applied to them, have now become crests i n and of themselves. In places of crest representations, many of them are distinguished by dec-oration and material: colours, abalone, copper, dentalium, porcupine q u i l l s , bone barbs, deer hooves, " g l i t t e r i n g things," and glass. At-trib u t e s of some r a r i t y , but devoid otherwise of s p e c i f i c iconographic meaning. In other words, although they are crest s , they are not part of a totemic system. Prestige has replaced iconographic meaning. 170 This chapter has described general categories of the Tsim-shian crests l i s t e d i n Appendix I I . I t has also discussed an analy-t i c a l separation of the Tsimshian crest system into h o r i z o n t a l and v e r t i c a l axes, the l a t t e r comprising those crests and aspects of crests unique to the Tsimshian. Considerable attention was given to crests i n a monster category and t h e i r r e l a t i o n s h i p to an underlying Cannibal theme i n Tsimshian thought. This led to a d i s t i n c t i o n , which goes beyond Le'vi-Strauss' totemic model, of crests of d i f f e r e n t i a t i o n and crests of i n t e g r a t i o n . The following Chapter w i l l explore these d i s -t i n c t i o n s i n more d e t a i l . However, i t w i l l do so i n terms of concrete examples -of Tsimshian a r t i f a c t s i n museum c o l l e c t i o n s and shown i n f i e l d photographs. This, too, w i l l represent an extension of L e v i -Strauss' model — from names and ideas to things. In a sense i t i s the c r i t i c a l chapter of the d i s s e r t a t i o n , f or i t i s the a p p l i c a t i o n of the meanings discovered i n the Barbeau f i e l d notes to the motifs on museum specimens. 171 CHAPTER SIX THE CREST SYSTEM I I : ICONOGRAPHY OF MUSEUM. SPECIMENS Rules of crest use were an unexpected category i n the Barbeau/ Beynon f i e l d data. A general assumption i n Northwest Coast studies has been that a crest could be represented on any kind of a r t i f a c t and, con-versely, that any a r t i f a c t decorated with a representation of a crest animal was therefore to be classed as " c r e s t a r t . " This understanding must now be modified, for the Tsimshian at l e a s t , f or there i s a d i s t i n c -t i o n to be made between representations of crests and other representa-tions of animals. The d i s t i n c t i o n depends on the type of a r t i f a c t i n -volved. When l i s t i n g crests and t h e i r owners, the Tsimshian almost i n v a r i a b l y mentioned c e r t a i n types of a r t i f a c t s on which the crests could, and more r a r e l y could not, be represented. I t i s impossible to know now whether t h i s information was s o l i c i t e d by Barbeau or volunteered by the Tsimshian, although I think i t was the l a t t e r . The t o t a l range of crest objects mentioned i s quite large and includes totem poles, house posts, house front and w a l l paintings, headdresses, amhala/it 's 1 ( f r o n t l e t headdresses), robes, l a d l e s , drums, feast dishes, and masks. 2 However, i t does not include h a l i b u t hooks, r a t t l e s , coppers, boxes, 3 or c h i l k a t s , items of complicated iconography often thought to be crest s . 1. Masks are a s p e c i a l case i n which the crest animal was also a naxn? x name. 2. One copper was included i n the l i s t s , but as a crest i t s e l f ; there was no reference to another crest being represented on i t . 3. There are two exceptions; they are discussed l a t e r i n the chapter. 172 I investigated this problem with halibut hooks collected by Barbeau for the NMC. Typically, they were the only halibut hooks I encountered in museum collections with documentation adequate to the purposes of iconographic investigation. Of a series of fourteen hooks made and used by Albert Wesley, a g.anha'da of g'it 'and?; and purchased from his widow, Barbeau wrote: "the carvings on such hooks are not meant to represent crests; they are only intended to bring good luck in fishing. The halibut is supposed to pick out the most attractive hook to be caught on" (Barbeau catalogue notes, 1915 collection; the series begins with No. VII-C-556). This, however, was contradicted by Barbeau's own catalogue notes for other halibut hooks which were said by some people to represent crests. It is necessary to look more closely at particular specimens in order to decide the issue. The series of hooks collected from Wesley's widow includes the following three, one of which was said by another person to "look like" the g*ispflwudwa'd? crest Grizzly of the Water. The others repre-sent a shark, which is a laxsk* i.;k crest, and the lag.ax'wE' sa (a double-headed sea monster), another g'ispywudwa'dg crest. The captions are based on Barbeau's catalogue notes. Plate 3. NMC. VII-C-565. Halibut hook. Made and used by Albert Wesley, g.anha'd^, g'it'and?'. Looks like Grizzly of the Water according to Peter Denny. Collected by CM. Barbeau, 1915. 173 Plate 4. NMC. VTI-C-566. Halibut hook. Made and used by Albert Wesley, g.anha'da, g'it'and> / . Carving rep-resents the lag.ax 1wE /s3 (a double-headed sea monster). According to Peter Denny i t i s a yE.aklak ("bird of under the water"), not used as a crest. Collected by CM. Barbeau, 1915. I P l a t e 5. NMC VII-C-562. Halibut hook. Made and used by Albert Wesley, g.anha'dj, g ' i t ' a n d s ' . The carving represents a shark swallowing a f i s h . Not a crest. Collected by CM. Barbeau, 1915. 174 Although the preceding are three h a l i b u t hooks with representations resembling one l a x s k ' i l k and two g'ispjwudwa'dg cr e s t s , they were made and used by a g.anha/da. Furthermore, i t was stated, apparently by Peter Denny, that the representations were not used as crests. Another hook c o l l e c t e d by Barbeau from K i n c o l i t h reinforces the i n t e r p r e t a t i o n that representations resembling the crests of one clan could be used i n non-crest contexts by persons of another clan. I Plage 6. NMC. VTI-B-1332. Halibut hook. Collected from Frank Bolton, laxsk'i'k, K i n c o l i t h . Made by yelna.o, Haida of Skidegate. The carving i s a bullhead or sc u l p i n with two eagles i n i t s mouth and was named " b i t i n g two eagles." This was not a c r e s t , but was intended to make thi s hook catch a l l kinds of f i s h . The bullhead i s supposed to be able to catch any kind of f i s h and even eagles. This hook was given to Bolton by yelna.o to show frien d s h i p ; Bolton gave him eulachon grease i n exchange. Collected by CM. Barbeau, 1927. While the eagles are a laxsk*i.'k crest (and Bolton was laxsk* iJ.k) , the bullhead or sc u l p i n i s a g. anha 'dg crest. Another hook (shown on the following page) c o l l e c t e d at Kin-c o l i t h suggests a shamanistic basis f o r h a l i b u t hook iconography: 175 Plate 7. NMC. VII-C-1398. Halibut hook. Collected from Albert A l l e n , g.anha'dg, K i n c o l i t h . Made by nag.adzu't, Haida of Masset, who died before ca. 1877. The carving represents a medicine man holding an octopus. The medicine man i s himself h a l f - h a l i b u t and the carving was used on the hook for good luck. 5 blankets (equivalent to $5.00) was paid f o r the hook. Collected by CM. Barbeau, 1927. The owner of this hook was g.anha xd9; the octopus was a laxsk' i.'k crest. Barbeau c o l l e c t e d another hook with a representation s p e c i -f i c a l l y s a i d to be the crest of two g'ispgwudwa 'd; houses, although he does not record whether t h i s information came from the owner of the hook or i f the owner was a member of eit h e r house. However, no such crest was claimed by the g'ispawudwa'dg i n the crest l i s t s . P l ate 8. NMC. VII-C-685. Halibut hook. From g ' i s p a x l a " t s . Carving s a i d to be Man of the Sea, a crest of the house of lag.axni'tsk (XI) and t'a.mks (IV), g*ispjwudwa 'Ad. Owner of the hook not s p e c i f i e d . Collected by CM. Barbeau, 1915. I 176 The burden of the preceding evidence i s that h a l i b u t hooks were carved with representations that were considered to be "lucky" or a t t r a c t i v e to the h a l i b u t ( i . e . , to have c e r t a i n magical a t t r a c t i o n s f o r the h a l i b u t ) , and that while these may be the same animals used as c r e s t s , they are not n e c e s s a r i l y the crest animals of the users of the hooks. Round r a t t l e s are shamans' r a t t l e s , although the animal rep-resentations on them are often interpreted as crests. Unfortunately, iconographic evidence i s meager; while a great many round r a t t l e s have been c o l l e c t e d , I found only four specimens with animal representations whose c o l l e c t o r s included data which could be used f o r iconographic purposes. Barbeau c o l l e c t e d a crudely carved round r a t t l e from Fanny Johnson, ha^namux, g* i s g . a / s t , Kitsegukla, i n 1924 (NMC, VII-C-1151), which was said to be carved with a representation of the owl, one of the crests of her house. While i t was not claimed by t h i s house i n the crest l i s t s , the owl was a Gitksan g*isg.a/st crest and i t i s more than probable that i t did belong to ha}namux's house. The use of the r a t t l e was not s p e c i f i e d but, being round, i t i s l i k e l y to be a shaman's r a t t l e . Another round r a t t l e c o l l e c t e d by Barbeau i n 1920 (NMC, VII-C-945) belonged to Alexander Mott of Hazelton, a laxse.'l shaman. This r a t t l e also contains a representation of the owl on one side and a human face on the other. The representation of the owl i s a r e a l i s t i c rendering of the whole body, and i s unmistakable. Now, the owl was not a l a x s e ' l c r e s t animal, Fortunately, Barbeau recorded that "the owner of t h i s had seen i t i n h i s dream and used i t on h i s r a t t l e ; whatever a 'doctor' 177 dreamt of could be used in this way although the object or animal might be the crest of another phratry [clan]." There seems to be a significantly large proportion of owl designs on round rattles in museum collections, suggesting that i t was a common shaman's s p i r i t , and suggesting also that Fanny Johnson might have been a shaman who also happened to be a g* isg.afst, and therefore that the owl on her rattle might have been there as a s p i r i t helper rather than a crest. Emmons collected a round rattle in 1909 from the Kitwanga laxsk*i'k chief samadi^k, who was also a shaman (MAI, 9/7998). The rattle has representations on i t of a frog and a beaver. While the beaver was a crest of samadi.'k's house, the frog was not. Once again, the represen-tation of a crest animal is li k e l y to be of secondary significance, since this i s a shaman's rattle. I t i s illustrated in Dockstader (1961: PI. 84). The fin a l example i s another round rattle purchased by Barbeau from Fanny Johnson in 1924 (below). This one was named Gnawing Marten, which was said to be a naxnj'x in ha7namux's house. When using the rat-t l e , the performer moved about imitating a marten and wearing a mask Plate 9. NMC. VII-C-1150. Rattle purchased from Fanny Johnson, ha'namux, g'isg.a.'st, Kitsegukla. Said to represent naxnp'x named Gnawing Marten. In using this rattle the performer moved about imitating a marten and wore a copper clawed headdress and a mask. Carved by g'itx^'n, laxsk'i'k, g*itsjla's£ in ca. 1889. Collected by CM. Barbeau, 1924. I 178 and crown of copper claws. There i s a form of naxnj'x known as a "curing" naxn;> /x, i n which the performer imitated a curing r i t u a l . This pro-bably explains the use of a round r a t t l e i n t h i s naxn^'x performance, e s p e c i a l l y since a crown of bear claws was the standard shamanistic headdress. My p r i n c i p a l point, however, i s that the marten was not used here as a crest representation. The evidence from these few, though well-documented, h a l i b u t hooks and round r a t t l e s i n museum c o l l e c t i o n s supports the conclusion based on the rules of use i n the crests l i s t s that animal representations on these a r t i f a c t s are not c r e s t s , even when they are s i m i l a r to those representations which are used as crests on other categories of material culture. There i s probably a tendency i n the case of shaman's r a t t l e s f o r the shaman to dream of and acquire as s p i r i t helpers the same a n i -mals to which he i s e n t i t l e d as cre s t s . The l a r g e r point being made here i s that crest representations are a l i m i t e d category of ceremonial a r t , one not to be confused with animal representations on f i s h hooks, shamans' r a t t l e s and, to extrapo-l a t e from these, other domestic, u t i l i t a r i a n , and shamanistic objects. The use of crest representations on war headdresses, armour, and weapons suggests that war on the Northwest Coast had ceremonial or r i t u a l aspects, which i t c l e a r l y did. Some writers (e.g., Codere, 1950) have suggested that the potlatch became a functional s u b s t i t u t e for War. Raven r a t t l e s , c h i l k a t blankets, and c h i e f s ' chests f a l l i n t o a s p e c i a l category of c h i e f l y prestige items, usually of unknown iconography, and w i l l be discussed below i n a section on C h i e f l y Symbols. 179 When we examine the crest l i s t s f o r the categories of ma t e r i a l culture which are s p e c i f i e d f o r c r e s t s , i t i s immediately obvious that they are intimately r e l a t e d to the potla t c h . The two categories most often s p e c i f i e d , and c l e a r l y the most important to the Tsimshian, are 4 1) a r c h i t e c t u r a l features: totem poles, including house entrance poles, house posts, house front paintings, beams, r a f t e r s , and ceremonial entrances; and 2) costume features: robes and headdresses. In other words, the most important objects f o r crest representations were those worn on the person i n the potlatch (and i n war) and represented on the house, which was where potlatches occurred. The focus of climax of these crest expressions were the person and dwelling of the c h i e f , who was the embodiment of h i s t r i b e ' s or l o c a l c l a n segment's power and prestige. While totem poles are generally considered to be the dominant and most pre s t i g i o u s form of a r c h i t e c t u r a l crest display on the Northwest Coast, Barbeau's Coast Tsimshian teachers evidently regarded house front paintings as being of the same class or order, the two forms being s p e c i f i e d together i n the majority of instances. In marked contrast, the Niska s p e c i f i e d house front paintings only 6 times, and the Gitksan only 10; totem poles were overwhelmingly preferred as a r c h i t e c t u r a l c r e s t displays i n these two d i v i s i o n s . I t should be remembered, however, that the Niska had many standing poles as l a t e as 1918 and the Gitksan s t i l l have, which were used as memory aids by people i n d i c t a t i n g crest l i s t s to Barbeau. Also, according to Barbeau's estimates (1929: 4; 1950: 4-12), 4. I consider the Tsimshian totem pole, i n both i t s free-standing and house-entrance forms, as part of or extension of the house i t s e l f . 180 most recorded Niska and Gitksan poles were carved a f t e r 1862, when many of the Coast Tsimshian converted to C h r i s t i a n i t y and stopped making them. There i s s l i g h t evidence that house front paintings were even more s i g n i f i c a n t o r i g i n a l l y than totem poles, and indeed may have preceded them as a r c h i t e c t u r a l crest expressions. Barbeau quotes h i s best tea-cher, Herbert Wallace, as saying that "of the two kinds, the house-front paintings (neksugyet) were the most important; they were the r e a l crest boards. The poles (ptsoen) were merely commemorative" (Barbeau, 1929: 15). This i s somewhat confirmed by the myths. There are several myth instances of supernatural o r i g i n s of house front paintings (the " l i v i n g " house fronts received from the Sky Chief have already been mentioned), but none of totem poles. Indeed, totem poles are almost never mentioned i n myth contexts. Boas (1966: 301) noted only two references to totem poles i n Tsimshian myths, one of which was a stone pole. There are a few other i n c i d e n t a l mentions i n the Barbeau c o l l e c t i o n of myths, but only a few. House front designs are sometimes said i n the myths to be carved, rather than painted. Tsimshian totem poles are w e l l known from Barbeau's two summaries (1929, 1950), and need not be discussed here. I l l u s t r a t i o n s of house front paintings are rare. We know of only three, and possibly four, Tsimshian examples. 1. A house front painting at g*idgstsu ; i s shown i n a photo-graph taken i n 1889 and published by Emmons (1930; also reproduced i n Barbeau, 1950: 775). I t i s i n c o r r e c t l y i d e n t i f i e d by Emmons as Northern Kwakitul and by Barbeau as T l i n g i t . The painting i s d i f f i c u l t to see, but shows two bi r d s i n p r o f i l e on eit h e r side facing a c e n t r a l s p l i t b i r d 181 f i g u r e painted over the door. A l l three are i d e n t i f i e d by Emmons as ravens. The owner of the house i s not known. 2. The house of sg.agwe't, I , l a x s k * i l k , g* it*and?•, at Fort Simpson i s shown i n an early photograph (see Plate 78b). Although i n one place Barbeau (1950: 116) s a i d that the house front painting " i n the old s t y l e represented the Eagle," he elsewhere (1950: 42) reports i n f o r -mant Herbert Wallace's d e s c r i p t i o n of i t as the g'i!b3lk (a complex mon-ster) : The Gyaibelk or Supernatural-Fly (narhnarem-gyoek) was painted with wings spread out ( i n various colours) on the house front of Sqagwait, one of the leading Eagle chiefs of the Tsimsyans, and was also used by some of h i s r e l a t i v e s on the Nass and the middle Skeena. On h i s head were shown several human faces. His beak, f i x e d to the p o l e , extended about 60 feet forward and had to be supported by a pole standing part of the way towards the t i p . The 60 foot beak Barbeau refe r s to i s missing i n the photograph, although a hole f o r i t s i n s e r t i o n can be seen. S t i l l elsewhere, Barbeau r e p o r t s , again according to Wallace, that i t had been painted by qa'lksak, g.anha'da, g ' i t s i C s , and was sold "to a purchaser from the United States about 1900" ( i b i d . : 774). The totem pole i n front of the house i s the Standing Beaver with at l e a s t 17 rings above him representing the discs on a woven basketry hat (langmg.Eit). 3. The t h i r d i n s i t u photograph of a Tsimshian house front i s that of manE'sk, l a x s k ' i l k chief of g*itlaxda'mks (Plate 78a). His name i s p r i n t e d i n bold l e t t e r s on the projecting beak. The painting's icono-graphy was not recorded, but mgnE 'sk also claimed the g * i l b d l k as a c r e s t , and there i s a marked s i m i l a r i t y between t h i s painting and the g * i ' . bglk on sg.awe't's house. 182 4. The fourth