^n ^^

J ;^: l:t il .^ ...

Edward Sapir



Collected Works


Edward Sapir





The Collected Works of Edward Sapir Editorial Board

Philip Sapir Editor-in- Ch ief

William Bright

Regna Darnell

Victor Golla

Eric P. Hamp

Richard Handler

Judith T. Irvine

Pierre Swiggers


Collected Works


Edward Sapir


Northwest California Linguistics

Volume Editors

Victor Golla Sean O'Neill


Mouton de Gruyter

Berlin New York

Mouton de Gruyter (formerly Mouton. The Hague)

is a Division of Walter de Gruyter GmbH & Co. KG. Berlin.

@ Printed on acid-free paper which falls within the guidelines of the ANSI to ensure permanence and durability.

Library of Congress Catalogmg-m-Puhlicatkm-Data

Sapir, Edward, 1884-1939.

Northwest California linguistics / volume editors, Victor Golla. Sean O'Neill.

p. cm. - (The collected works of Edward Sapir : 14)

Includes bibliographical references and index.

ISBN 3-11-016432-9 (cloth : alk. paper)

1. Indians of North America - California, Northern - Lan- guages. I. Golla, Victor. II. O'Neill, Sean, 1969- III. Title. PM501.C2 S26 2001 497'.09794-dc21 2001032705

Deutsche Bihliothek - Cataloging in Publication Data

Sapir. Edward:

[The collected works] The collected works of Edward Sapir / ed. board Philip Sapir ed. -in-chief .... - Berlin ; New York : Mouton de Gruyter

ISBN 3-11-010104-1

ISBN 0-89925-138-2

14. Northwest California linguistics / vol. ed. Victor Golla ; Sean O'Neill. - 2001 ISBN 3-11-016432-9

© Copyright 2001 by Walter de Gruyter GmbH & Co. KG. 10785 Berlin.

All rights reserved, including those of translation into foreign languages. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, includ- ing photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Printing: Werner Hildebrand. Berlin. Binding: Liideritz & Bauer, Berlin. Printed in Germany.

Edward Scipir in Hoopa Valley, Summer 1927

( Courtesy of Sapir Jaiuily )

Edward Sapir (1884-1939) has been referred to as "one of the most brilliant scholars in linguistics and anthropology in our coun- try" (Franz Boas) and as "one of the greatest figures in American humanistic scholarship" (Franklin Edgerton). His classic book, Lan- guage (1921), is still in use, and many of his papers in general hnguis- tics, such as "Sound Patterns in Language" and "The Psychological Reality of Phonemes," stand also as classics. The development of the American descriptive school of structural linguistics, including the adoption of phonemic principles in the study of non-literary lan- guages, was primarily due to him.

The large body of work he carried out on Native American lan- guages has been called "ground-breaking" and "monumental" and includes descriptive, historical, and comparative studies. They are of continuing importance and relevance to today's scholars.

Not to be ignored are his studies in Indo-European, Semitic, and African languages, which have been characterized as "masterpieces of brilHant association" (Zellig Harris). Further, he is recognized as a forefather of ethnolinguistic and sociolinguistic studies.

In anthropology Sapir contributed the classic statement on the the- ory and methodology of the American school of Franz Boas in his monograph, "Time Perspective in Aboriginal American Culture" (1916). His major contribution, however, was as a pioneer and pro- ponent for studies on the interrelation of culture and personality, of society and the individual, providing the theoretical basis for what is known today as symbolic anthropology.

He was, in addition, a poet, and contributed papers on aesthetics, literature, music, and social criticism.


Frontispiece 6

Preface 11

Introduction 13

Hupa Texts, with Notes and Lexicon (Edward Sapir and Victor Golla) 19

Yurok and Chimariko Materials

Yurok Texts (edited by Howard Berman) 1015

Chimariko Linguistic Material (edited by Howard Berman) 1039

Appendix: Reports on Sapir' s Northwest CaHfornia Work

Letters from the field to A. L. Kroeber 1081

Letters to J. P. Harrington regarding work on Chimariko 1090

An Expedition to Ancient America 1094

A Summary Report of Field Work among the Hupa, Summer 1927 ... 1097 Reminiscences about Edward Sapir (Fang-Kuei Li) 1099

References 1101

Index 1 113


Volumes VII-XV of The Collected Works of Edward Sapir are devoted to re- publication of Sapir's works of monograph length grammars, dictionaries, text collections, and ethnographies, including works that were edited and pub- lished posthumously and to the publication for the first time of edited ver- sions of projects that were left in manuscript at the time of Sapir's death in 1939. This volume falls into the latter group and represents the first major ad- dition to the corpus of Sapir's published work in several decades.

Included in this volume are the fruits of Sapir's field trip to Hoopa Valley in Northwest California during the summer of 1927. The most important of these is Sapir's documentation of the language and culture of the Athabaskan Hupas: 77 narrative texts, a substantial lexicon, and numerous ethnographic notes. Sapir also collected some valuable Yurok data, including three texts, as well as some miscellaneous linguistic materials on Chimariko. The Hupa corpus has been edited and provided with thorough annotation by Victor Golla, with assis- tance in some sections from Sean O'Neill. The Yurok texts have been edited and annotated by Howard Berman, who has also contributed a detailed assess- ment of the Chimariko materials. Editorial responsibility for the entire volume rests with Golla and O'Neill jointly.

Preparation of all of the monographic volumes of The Collected Works of Edward Sapir was aided by grants from the National Science Foundation (grant no. BNS-8609411), the Phillips Fund of the American Philosophical Society, and the Wenner-Gren Foundation for Anthropological Research. Funding for the work on the Yurok and Chimariko materials was specifically provided in the NSF grant. Work on the Hupa material has additionally been supported by a fellowship (to Golla) from the American Council of Learned Societies and a research assistantship (to O'Neill) from the Department of Anthropology, Uni- versity of California, Davis.

Special recognition must be given to the Hoopa Valley Tribe, both collec- tively and individually, for their sustained help in the decades-long project of preparing Sapir's Hupa texts for publication.

The editors would like to give specific thanks to several individuals for their special contributions to the preparation of this volume: Philip Sapir, who in ad- dition to serving as Editor-in-Chief of The Collected Works of Edward Sapir contributed immeasurably to understanding the personal context of his father's work; Michael Sapir and Helen (Sapir) Larsen, for their memories of the sum-

12 XIV Northwest California Linguistics

mer of 1927; Catherine Griggs, secretary for the Sapir editorial project at George Washington University, 1986-89; and Howard Herman, of Seattle, Washington, who was unhesitatingly willing, so many years after completing the editorial work, to correct the proofs of the Yurok and Chimariko sections.


Several significant chapters of Edward Sapir's research were devoted to the languages of Northern California and Oregon. His earliest field project, in 1905, was a study of Wishram Chinook on the Columbia River (Volume VII), fol- lowed by studies of Takelma of Southwest Oregon (Volume VIII) and Yana of Northeast California (Volume IX). From 1913 onwards he played an important role, together with Roland B. Dixon and Alfred L. Kroeber, in the establishment of deeper genetic connections among the languages of this area, and between California and Oregon languages and languages in other parts of the continent. This included linking Yurok and Wiyot of Northwest California to the Algon- quian family, expanding Dixon and Kroeber' s Penutian stock to include a num- ber of Oregon languages (Takelma and Chinook among them), and exploring the relationships of several Northern California languages within Dixon and Kroeber' s Hokan stock (in particular Yana and Chimariko). However, it was none of these areal concerns that led him, beginning in 1921, to plan a field trip to Northwest California. The intended subject of this new investigation was quite specifically Hupa, one of the California Athabaskan languages, and Sapir's immediate goal was to obtain a more accurate phonological and grammatical record of that language than Pliny Earle Goddard had presented in his numerous publications (most importantly Goddard 1905 and 1911). The broader goal was a clearer understanding of the Athabaskan languages, and of the deeper genetic relationships which Sapir believed Athabaskan to have both in North America and in Asia.

In 1921, Sapir's only field experience with an Athabaskan language was a brief encounter with Chasta Costa of Oregon during his Takelma work (Sapir 1914c). A deeper interest in these languages was first stimulated during his continent-wide search for deep genetic connections. Although the Athabaskan languages stood apart from the Algic, Penutian and Hokan relationships, by 1914-15 he had accumulated what seemed persuasive evidence for a "Na-Dene" stock that encompassed not only Athabaskan and Tlingit, which had previously been thought to be related by Swanton, but also Haida (Sapir 191 5d). Unlike the Algic, Penutian, and Hokan relationships, however, further linkages of Na- Dene in North America were not forthcoming. Instead, Sapir began to note evi- dence pointing to the startling possibility that the Na-Dene group was a branch of an ancient stock that straddled the Bering Strait and included Sino-Tibetan (see Volume VI, pp. 133- 140). This possibility so fascinated him that by 1920- 21 he was devoting a substantial part of his research time to assembling com-

14 XIV Northwest California Linguistics

parative data in support of both Na-Dene and "Sino-Dene", as well as to recon- structing the details of Proto-Athabaskan. As his work progressed the need for richer and more accurate descriptive data became acutely apparent, in particular from Athabaskan languages. As soon as funding for research became available again after the economies of World War I, Sapir began laying plans for an exten- sive program of field work on Na-Dene languages in Canada and the western United States.

In the early 1920s Hupa, despite its peripheral geographical position in the family, was probably the best known Athabaskan language, and Goddard's pub- lications on it had done much to determine the paradigm for Athabaskan lin- guistics. A reassessment of Hupa was a major item on Sapir' s research agenda, and his first Athabaskan field trip, in the summer of 1921, was to include both a study of Sarcee (Tsut'ina) in Alberta and a quick visit to Northern California to obtain fresh data on Hupa (Golla 1984a: 368-369). This trip had to be can- celled, in large part because of the illness of Sapir's wife Florence. He was able to reschedule the Sarcee work for the summer of 1922, although not the addi- tional trip to California. Sarcee, however, proved quite rewarding in its own right. Sapir collected a substantial number of texts (to appear in Volume XIII) and sufficient phonological and grammatical data to justify a deeply revised model of Athabaskan structure, in particular one that took into account the fact that Sarcee had revealed itself to be a thoroughgoing tone language. Sapir was now convinced that Goddard had ignored or seriously misinterpreted a number of facts about Hupa, not least that, in all likelihood, it too had tonal distinctions (Sapir 1922a, 1925f; for the evolution of Sapir's thinking on tone and other features of Athabaskan phonology and grammar, see Krauss 1986).

Sapir began making plans for a Hupa trip the following year (1923) (Golla 1984a: 402). Again, however, personal difficulties made the journey impracti- cal. It was necessary for Sapir to spend the field season of 1923 at a sanitorium in Pennsylvania convalescing from a broken leg as well as tending to his now seriously ill wife. Providentially, Sapir discovered that two young Athabaskan men from Alaska were working at a nearby summer camp, and he seized the opportunity to collect data on Anvik (Ingalik, Deg Hit'an) and Kutchin (Gwich'in) (to appear in Volume XIII). But interesting as this material was to him Kutchin, if not Anvik, had a tone system clearly cognate with that of Sarcee Hupa still required attention.

It would be three more years, however, before another opportunity to visit Northwest California presented itself. Florence Sapir died early in 1924, leaving Sapir's life in turmoil for many months and prompting a major career change. In 1925 he resigned from his research and curatorial position with the Victoria Museum in Ottawa to take an Associate Professorship at the University of Chi- cago. The only new Athabaskan data he was able to collect during his first two years of teaching were on Navajo, which he obtained from a speaker living in

Introduction 15

Chicago. His discovery that Navajo also possessed the Sarcee-Kutchin tone system underscored the need to set the record straight on Hupa. After the years of delay, plans finally solidified for a Hupa field trip in the summer of 1927.

Sapir was accompanied to California by a graduate student, Fang-Kuei Li, who had begun studying with him the year before. Li, a native speaker of Man- darin, had come to the University of Chicago for advanced study in comparative linguistics, but had quickly been drawn into Sapir's orbit. Believing that Li's acquaintance with tone languages would make him an excellent field researcher on Athabaskan languages, Sapir had him familiarize himself with Athabaskan structure by writing a study of Sarcee verb stems based on Sapir's 1922 notes (published as Li 1931). On the 1927 trip the plan was for Li to work directly with Sapir on Hupa for a few weeks, then to strike off on his own to find speak- ers of other Athabaskan languages in the area (see Appendix, "Reminiscences about Edward Sapir"). Li spent a month in Petrolia, Humboldt Country, work- ing with a speaker of Mattole, and a second month on the Round Valley Reser- vation, Mendocino County, working on Wailaki. His Mattole material, which is largely paradigmatic elicitations, served as the basis for his doctoral dissertation, which was published in 1930. His Wailaki material, which contains many texts, is of equally high quality but has been published only in part (Seaburg 1977a, 1977b).

Sapir and Li, together with Sapir's teen-aged son Michael, arrived in Hoopa Valley in the last week of June, 1927. Li left for his own field work in early July, and shortly afterwards Sapir and his son were joined by Sapir's (second) wife, Jean, and his daughter Helen. Jean Sapir, a psychiatric social worker, collected English versions of several Yurok narratives (J. Sapir 1928). The group was enlarged further in August by the musicologist George Herzog, who had recently come to America from Hungary to study Indian music under Boas. In 1927, supported by Boas and the American Museum of Natural History, Herzog made an extended field trip through California and the Southwest, ar- ranging to visit Hoopa Valley during Sapir's stay. Herzog collected twenty-two cylinder recordings of Hupa songs, now preserved in the Archives of Traditional Music at Indiana University (Seeger and Spear 1987: 54). Sapir had also hoped that Alfred Kroeber would be able to spend some time in Hoopa while he was there, but Kroeber' s schedule did not make this possible (see Appendix, "Letters from the field to A. L. Kroeber"). Jean and Helen Sapir left Hoopa in late August, and Sapir himself, together with Michael, followed on September 7. Li returned to Chicago independently two weeks later.

Sapir considered his Hupa study to have been "eminently successful" (Appendix, "An Expedition to Ancient America"), although he found himself embarrassingly wrong about phonemic pitch. Hupa, unlike Navajo and Sarcee, is not a tone language, as subsequent research has shown to be true of a signifi- cant number of Athabaskan languages (Krauss and Golla 1981: 69-71). But if

16 XIV Northwest California Linguistics

Goddard was right on this score (see Goddard 1928), the new data showed that Goddard's morphological analysis, as Sapir had suspected, was quite inade- quate. Although Hupa was revealed to have a number of unique or unusual features, it differed from Sarcee and the other Athabaskan languages Sapir had investigated much less dramatically than Goddard's description had suggested. A new grammar of Hupa, and a new comparative grammar of Athabaskan, needed to be written.

Sapir also found Hupa traditional culture much richer than he had expected, and the Indian people with whom he worked particularly his chief consultant and interpreter, Sam Brown unusually knowledgeable and articulate. He later described his extensive Hupa texts as "probably the best I ever collected" (Sapir to Kroeber, August 25, 1938, UCB), and recommended to Kroeber that he "get in closer and closer contact with the Hupa Indians and take a good look at their religion," predicting that he would find it "a most exciting and rewarding task" (Sapir to Kroeber, August 2, 1930, UCB).

Although the primary motive for Sapir' s Northwest California trip was to document Hupa and related Athabaskan languages, he did not forego the op- portunity to collect smaller amounts of data on other languages of the area. Early along in his stay, he (and Li) set the routine of taking Sundays off from Hupa to work a few hours on Yurok with Mary Marshall, a half-Yurok woman fluent in both Hupa and Yurok. (Mrs. Marshall was also the source of several Hupa texts.) Although Sapir's interest in Yurok dated back to 1913, and his demonstration that Yurok and Wiyot (the "Ritwan" languages) were distantly but firmly related to Algonquian (Sapir 191 3h), there is no indication that he wanted to devote more than a limited amount of time to the language. His pri- mary motive seems to have been to provide others primarily Kroeber with a detailed phonetic record, and to make suggestions for future investigation.

Although a similar opportunity to gather data on Karuk seems not to have been pursued, the chance to hear Chimariko a nearly-extinct Hokan language of considerable importance to understanding that stock (see Sapir 1920d; GoUa 1984a, 316) could not be passed up, even if it entailed a long automobile trip over the back roads of Trinity County. In the end, the speakers Sapir found were poor and the scanty Chimariko data he collected were less useful than he had hoped they would be. Furthermore, on his return from California Sapir was distressed to discover that John P. Harrington had secretly made an extensive study of Chimariko a few years earlier with the last fluent speaker, and had been continuing field work with other speakers. The days that Sapir had "lost" from his Hupa work thus went largely for naught (see Appendix, "Letters to J. P. Harrington regarding fieldwork on Chimariko").

Other than the two brief summaries he wrote immediately after his return (see Appendix) Sapir published only one fragment of his Hupa material, an ethno- graphic essay on Hupa tattooing based on the text on that subject (text 15 in the

Introduction 17

present edition), together with some notes and diagrams, that he had obtained from Sam Brown (Sapir 1936e). A typescript version of the Hupa texts and eth- nographic notes, the organization of the latter apparently the work of Leslie Spier, circulated privately among California anthropologists in the 1940s and 1950s. Another typescript version of English translations of the texts was pre- pared by the senior editor for the Hoopa Tribe in the early 1980s. Full publica- tion of this material, together with the Yurok and Chimariko notes, was assigned high priority by the Editorial Committee of the Collected Works of Edward Sapir at its first meeting in 1984. That it has taken more than a decade and a half to redeem this pledge is due to the usual exigencies of scholarly work, ex- acerbated by the size of the corpus and the conviction of the senior editor that the texts deserved full linguistic and ethnographic annotation.

Hupa Texts, with Notes and Lexicon Edward Sapir and Victor Golla


Preface 25

Key to the Orthography 31

Alphabetical Order 33

I. Ceremonies

1. The White Deerskin Dance {Sam Brown) 35

2. The Jump Dance (Sam Brown) 55

3. The Origin of the Jump Dance (Mary Marshall) 72

4. The Origin of the Misq'id Jump Dance {Jake Hostler) 76

5. The Acorn Feast {Sam Brown) 84

6. The First Salmon Ceremony {John Shoemaker) 99

7. The Origin of the First Salmon Ceremony

{John Shoemaker) 103

8. Bathing the Rain Rock at Sugar Bowl {John Shoemaker) 109

9. Formula for Bathing the Rain Rock {John Shoemaker) 113

10. The Flower Dance {Sam Brown) 1 18

1 1. The Origin of the Flower Dance {Sam Brown) 135

12. The Brush Dance {Sam Brown) 147

II. Traditional Life

13. How to Treat Babies {Emma Frank) 159

14. Traditional Standards of Beauty {Emma Frank) 167

15. Tattooing {Sam Brown) 170

16. Rules of Etiquette {Emma Frank) 175

17. How to Ensure a Long Life {Sam Brown) 179

18. Insults and Bad Behavior {Sam Brown) 181

19. A Jump Dance Sermon {Sam Brown) 185

20. When there is a Death {Sam Brown) 188

21. How to Gather and Prepare Acorns {Sam Brown) 194

22. How Basketry Roots are Gathered

and Baskets Woven {Sam Brown) 208

23. How to Handle Firewood

in the Old-Fashioned Way {Emma Frank) 217

III. Doctoring and the Spirit World

24. How I Became a Doctor {Emma Frank) 223

25. Emma Frank's Method of Doctoring {Emma Frank) 230

26. How Silis Became a Doctor {Sam Brown) 238

27. Silis' Vision of Hell {Sam Brown) 251

22 XIV Northwest California Linguistics

28. Silis Loses and Finds her Pipe {Sam Brown) 257

29. The Kick Dance {Sam Brown) 260

30. How I Became Fond of Kick Dancing {Sam Brown) 268

31. k>i}we" Sickness is Taken Out {Sam Brown) 274

32. Hunting Magic and Prayers for Deer {Sam Brown) 280

33. A Vision of Sam Brown's {Sam Brown) 283

34. Power Over the Grizzly Bear {Sam Brown) 292

35. Training for Striking Fear into the Grizzly Bear

{Sam Brown) 300

36. The Afterworld {Sam Brown) 303

IV. Medicine Formulas

Introductory Note 305

37. To Cure Vomit and Passing of Blood {Emma Frank) 307

38. For Sickness {Jake Hostler) 312

39. For a Menstruating Woman to Shorten her Period

{Emma Frank) 315

40. Pronounced over a Child to make him

Wealthy and Brave {Emma Frank) 320

41 . To Purify one who has Handled a Corpse

{John Shoemaker) 327

42. To Induce Forgiveness {Mary Marshall) 342

43. For War {Sam Brown) 344

44. The Ugly Dog, a Medicine Formula for Love

{John Shoemaker) 349

45. A Woman's Love Medicine for Getting a Man

{Emma Frank) 352

46. A Man's Love Medicine {Emma Frank) 356

47. For a Man who Desires Wealth and Success

in Love {John Shoemaker) 362

48. For Wealth {John Shoemaker) 366

49. The Dawn Maiden {Sam Brown) 371

50. Big Dentalia Comes to the Redwood People

{Mary Marshall) 374

V. Prayers

Introductory Note 377

51. Prayer for Good Luck {Emma Frank) 378

52. Prayer to the Sun {Emma Frank) 379

53. Prayer in Going Over the Trails in the Mountains

{Emma Frank) 380

54. Prayer to the Acorn Feast Ground

of Hostler Ranch {Sam Brown) 382

Hupa Texts: Contents 23

55. Prayer to Yimantiw'^winyay's Resting Place

(Sam Brown) 384

56. A Night Prayer (John Shoemaker) 386

VI. Myths and Tales

57. Salmon's Grandmother (Sam Brown) 387

58. Salmon's Grandmother and Timber Robin (Sam Brown) 394

59. Water Dog and Acorn Worm (Sam Brown) 397

60. Iris and Panther (Oscar Brown) 405

61. Coyote and Frog (Emma Frank) 408

62. Three Tales of Little Woodpecker (Emma Frank) 412

63. Salmon-Trout is Doctored by Bluejay

and Hummingbird (Jake Hostler) 430

64. A Story of the South Wind (Sam Brown) 433

65. The Lake Whale (Sam Brown) 437

66. The Rival Wives (Mary Marshall) 438

67. The Two Brothers (John Shoemaker) 446

68. The One who Established a Medicine at Miyimidaq'id,

and his Grandmother (Emma Frank) 450

69. The Hated Suitor (Jake Hostler) 468

VII. Legends and Traditional History

70. How Ear Aches Started at Me'^dildir) (Sam Brown) 479

71. The Acorn Provider Saves Weitchpec (Sam Brown) 486

72. Two Young Men Do Indian Deviling (Jake Hostler) 490

73. The Village that Blasphemed (Sam Brown) 497

74. How a Man was Destroyed at New River (Sam Brown) 499

75. How a Tattooed-Face Stole a Woman (John Shoemaker) 505

76. The Chimariko Attack a Hupa Village (Sam Brown) 509

77. A War Between the Hupa and Yurok (Sam Brown) 5 1 5

Linguistic NOTES TO THE TEXTS (with Sean O'Neill) 531

Analytic Lexicon

Introductory Note 725

Element List (with Sean O'Neill) 727

Morphological Outline 813

Ethnographic lexicon

Introductory Note 873

General Glossary 875

Kinship Terminology 999


This edition of the Hupa texts and other materials that Edward Sapir collected in the field during the summer of 1927 is the product of 38 years of intermittent labor by Victor Golla, materially aided since 1996 by Sean O'Neill. Over this span of time Golla' s own work on Hupa has become so intricately intertwined with Sapir' s that co-authorship rather than editorship is the more accurate attri- bution, similar to the role Morris Swadesh took with Nootka Texts (1939) or Harry Hoijer with The Phonology and Morphology of the Navaho Language (1967). The texts themselves are presented in as diplomatic an edition of Sapir' s originals as the circumstances permit, with editorial departures from the content of the notebooks explicitly marked. The free translations, however, are entirely Golla' s, as are many of the textual notes. Sapir' s ethnographic notes are partly incorporated into the textual notes, and partly into the Ethnographic Lexi- con. The linguistic analysis reflected in the annotations and lexicon is Golla's, although based on the model of Athabaskan morphosyntax that is explicit in Sapir' s published work on other Athabaskan languages and implicit in the or- ganization of his Hupa files. Whatever the balance of authorship in any specific part of the work, however, the ultimate responsibility for all errors and infelici- ties rests with Golla.

History of the Work

Sapir obtained the primary materials on which this work is based during an eleven-week stay on the Hoopa Valley Reservation, California, from late June to early September, 1927 (see Appendix). He mainly worked with six fluent speakers of Hupa: Sam Brown, Emma Frank, John Shoemaker, Mary Marshall, Jake Hostler, and Oscar Brown. He also interviewed several others, including Berman Lack and Ada Masten, for shorter periods of time and without obtaining texts. In all, Sapir transcribed from dictation 77 narrative texts. He collected these in eleven top-bound field notebooks, each with 62 double pages, and en- tered associated lexical and grammatical data on approximately 5,000 4" x 6" slips (Sapir ms. 1927a). Copious ethnographic notes are also found in the note- books, usually on the pages opposite (above) the texts. Many of the notes on file slips are keyed to specific words and phrases in the texts, and consist of pro- nominal and aspectual paradigms and other elicited data expHcating the textual form. All of the texts are in interlinear format, a Hupa line in broad phonetic

26 XIV Northwest California Linguistics

transcription alternating with an English line that translates the Hupa line word by word, although occasionally glosses are omitted or abbreviated when pre- dictable from context. The procedure that Sapir apparently followed was to transcribe the Hupa text from dictation without pausing for translation, and later to go over the text with Sam Brown, who provided careful English interpreta- tions and supplementary linguistic data. Most of the ethnographic data was also supplied by Sam Brown.

Between 1927 and Sapir's death in 1939 some preliminary steps were taken to organize these materials for publication. An assistant typed all of the Hupa texts, with two carbon copies, penciling in the English glosses on the original copy, and a similar typescript was made of all ethnographic and textual notes contained in the notebooks (Sapir ms. 1927b). Sapir, however, created no sec- ondary grammatical files, and so far as is known he prepared no analysis of the language beyond what is implied in his original notes. He outlined his prelimi- nary findings in two short publications, and in letters to colleagues written at the time of his work (see Appendix). He alluded to some particulars of Hupa structure in two publications (Sapir 1931b and 1936f), and he briefly discussed Hupa phonology and grammar in a course at Yale on Athabaskan linguistics (Haas ms. 1936, Newman ms. 1936). Sapir's only publication of the narrative texts was an English translation and ethnographic commentary on text 15, "Tattooing", written for Alfred Kroeber's festschrift (Sapir 1936e).

After Sapir's death his Hupa notebooks and files, together with the type- scripts, were placed in the care of Harry Hoijer, together with all of Sapir's other Athabaskan materials. Leslie Spier appears to have had at least some of the Hupa materials in his possession later in the 1940s. He planned to write an eth- nographic report on Hupa based on Sapir's data, similar to the ethnographies of Wishram and Yana that he had edited from Sapir's notes (Spier and Sapir 1930; Sapir and Spier 1943). A preliminary typescript of the ethnographic notes, or- ganized into broad categories, was prepared, but no further work was done.i A typescript of the texts, under the title "Hupa Myths, Formulas, and Ethnological Narratives in Text and Translation", appears to have circulated among Califor- nia anthropologists, and was available to William J. Wallace and Edith Taylor during their ethnographic work on Hupa in the late 1940s.

GoUa began working on Hupa in 1962, conducting independent fieldwork for three summers (1962-64) as well as the autumn of 1963. In 1963 Hoijer turned Sapir's notebooks and files over to Golla along with a copy of the typescript of the texts. With Hoijer's encouragement, Golla incorporated Sapir's grammatical and lexical materials into the data set that formed the basis of his dissertation, a study of Hupa grammar. In reciprocaion, Hoijer requested Golla to edit Sapir's texts for publication.

Golla finished his dissertation in 1970 and (keeping a photocopy of the note- books) returned Sapir's original materials to Hoijer, who shortly thereafter de-

Hupa Texts: Preface 27

posited them in the Library of the American Philosophical Society. In 1975 Golla received a fellowship from the American Council of Learned Societies that allowed him to prepare a preliminary phonemic version of the texts, with free translations and ethnographic annotations. Except for an annotated version of text 61, "Coyote and Frog" (Golla 1977b), this version remained unpublished. In the early 1980s Golla worked under contract to the Hoopa Tribe to prepare a typescript edition of his free English translations, with basic ethnographic notes. This compilation has circulated in photocopied form among Hoopa Tribal mem- bers for cultural education and language restoration purposes.

The present publication began to take shape in 1996 when, during a year at the University of California, Davis, Golla began working with Sean O'Neill, a graduate student in Linguistic Anthropology. Under Golla's direction, O'Neill prepared a full computer file of Golla's 1975 preliminary edition, restoring from the (photocopies of the) original notebooks various details that had been regu- larized. Golla and O'Neill then drafted linguistic annotations to the texts (for the final version of which Golla must assume responsibility), and Golla reedited the textual notes and compiled an ethnographic glossary. O'Neill drafted an element list based on earlier materials of Golla's, which Golla revised and linked to a morphological outline.

The Consultants

Of the speakers of Hupa whom Sapir interviewed during his stay, the follow- ing six were the sources of the narrative texts and ethnographic data in his note- books:

Sam Brown (the source of thirty-nine texts, all those not noted below) was Sapir's chief consultant and interpreter. He had earlier served as Goddard's in- terpreter (1904: 93), and in later years would work with other anthropologists. A batchelor, he often wore a (woman's) basket cap and excelled in such tradi- tional female occupations as basket weaving, acorn processing, and cooking. By middle age, when Sapir knew him, he had become one of the chief authori- ties on World Renewal matters, as well as for a number of other rituals. Al- though his father was White, his mother came from a branch of the family that owned the xontah nikya-w in toPJiyimH-dirj. He died in 1959 at the age of 80.

Emma Frank (texts 13-14, 16, 23-25, 37, 39-40, 45-46, 51-53, 61-62, 68) was in her late 60s at the time of Sapir's visit. The source of three of his Hupa Texts, she is described by Goddard as a "very conservative" woman who "surpasses all other Hupa women in basket-making" (1904: 324). She was one of the last woman to have been fully trained as a traditional doctor, and for that and other reasons she was generally referred to by her Indian name, tahse-nc'e'^. Before marrying Henry Frank her (non-Indian) maiden name was Emma Dusky.

28 XIV Northwest California Linguistics

John Shoemaker (texts 6-9, 41, 44, 47-48, 56, 67, 75), also called "Shoemaker John", was in his 70s at the time of Sapir's visit, and a respected elder from the upper half of Hoopa Valley (the me'^dilx'^e-). He was the younger half-brother of Robinson Shoemaker, the source of three of Goddard's Hupa Texts (1904: 237, 265) and the last person to officiate at the First Salmon Feast (text 6). John Shoemaker died at an advanced age in 1949.

Jake Hostler (texts 4, 38, 63, 69, 72) was a member of a prominent family from ta^Uyimii-dirj, but otherwise little is known of him other than that he was the great-uncle of Jasper Hostler, a contemporary Hupa elder.

Mary Marshall (texts 3, 42, 50, 66) was half Yurok on her mother's side (a woman from the village of Sa'^a opposite Kepel [Map C-19] who was married to a Hupa), and spoke both Hupa and Yurok fluently. She was the wife of James Marshall and the mother of Julius Marshall. All three worked with Goddard, and Mary Marshall was the source of nine of the narratives in Goddard's Hupa Texts (1904: 93, 150; see also Goddard 1907: 4, and plates 1-2). Kroeber, who obtained information from Mary Marshall on the World Renewal ceremonies at Kepel (Kroeber and Waterman 1938), describes her as "intelligent, friendly, and helpful, and patient about dictating slowly" (Kroeber ms. n.d.).

Oscar Brown (text 60), Sam Brown's older brother, is described by Kroeber as "a burly, hearty, extraverted man of about my age [Kroeber was bom in 1 876] who rented me saddle horses and helped steer me around when I was new to Hupa. He left most religious matters to his sensitive younger brother Sam, inclining rather to the practical and skeptical" (Kroeber ms. n.d.). Oscar Brown also served as Goddard's interpreter in 1901 and the source of three of his Hupa Texts {\9^A:91>, 135).

GoUa's independent data on Hupa were largely obtained in 1962-64 from three primary consultants:

Ned Jackson and his wife Louisa Jackson. Ned was Sam Brown's nephew, his mother being Sam's sister. Ned and Louisa were the parents of James "Jimmy" Jackson, who at the present time (2000) is the most knowledgeable Hupa elder.

Minnie Reeves, Louisa Jackson's older sister. Louisa and Minnie were chil- dren of the Hill family, and were bom at noleh-ditj (Map D-5), the last Redwood Creek {x'^iyiqid) village to be inhabited. Although both sisters had lived in Hoopa Valley since 1888, their speech may have retained a few characteristics of the Redwood Creek dialect.

Hupa Texts: Preface 29

General Editing Conventions

In the presentation of the Hupa texts, all Hupa forms are retranscribed in the phonemic orthography described in the accompanying Key to the Orthography, but are otherwise unaltered except for: (1) editorial emendations, which are in- dicated by square brackets; (2) the restoration of consonants elided by the rules of external sandhi (Morphological Outline §71), which are indicated by paren- theses. Similarly, all glosses are printed as they stand in Sapir's notebooks, and missing glosses supplied by the editors are indicated by square brackets.

All Hupa words (including underlying forms) are printed in italics, except for words and phrases that are in the General Glossary, which are printed in unitali- cized boldface. Thus, a boldfaced Hupa word in a free translation or textual note refers the reader to a specific entry in the General Glossary.

Hupa words are alphabetized according to the order in the Key to the Orthog- raphy, which is basically the order of the English alphabet, with special conven- tions.


In addition to notes specifically on the form and content of the texts, which are printed as textual notes immediately following the free translations, the texts have been provided with a substantial amount of additional linguistic and ethno- graphic annotation.

The linguistic annotation takes the form of a separate section of Linguistic Notes to the Texts. These notes for which Golla takes primary responsibility, with acknowledgement of important editorial contributions from Sean O'Neill are intended to elucidate the phonological and morphosyntactic structure of forms whose structure is not immediately obvious to a reader familiar with the basic principles of Hupa word formation. The linguistic notes are also designed to be a pedagogical aid for anyone who might wish to use the texts to learn the grammar of Hupa, and to this end they are extensively cross-referenced. The grammatical terminology used in these notes is explained in the Morphological Outline.

Sapir collected ethnographic information on the Hupas only incidentally to his linguistic work, although for some subjects with considerable thoroughness. Most of this material has been incorporated into the present work, either as fur- ther notes to the texts or as items in the General Glossary. The only significant pieces of ethnographic data omitted pertain to a few special subjects, such as personal names, which require more focused treatment than can be given here.

Both in the textual notes and in the General Glossary efforts have been made to refer the reader to all relevant published sources of information on traditional

30 XIV Northwest California Linguistics

Hupa culture, some of them postdating Sapir's work. Quotations from these sources are frequently included, with appropriate attribution. In general, all un- attributed ethnographic commentary is Sapir's, usually reflecting information from Sam Brown.


1 . These materials are now filed with the earlier typescript of the Hupa material, described above, in the American Philosophical Society Library (Sapir ms. 1927b).

Key to the Orthography

All Hupa words in this volume are written in a phonemic orthography closely modeled on the one Sapir used in his separate publication of text 15 "Tattooing" (Sapir 1936e). The conventions of this orthography follow general Americanist practice as it was standardized in Haas et al. (1934). Although this orthography differs from the writing system adopted by the Hoopa Tribe and used for cul- tural and educational materials (cf. Golla 1996b) the two systems are intercon- vertable. The only phonological feature marked in the present orthography that is not represented in the Tribe's writing system is the high pitch cadence, marked by a circumflex accent.

In his 1927 notes Sapir transcribed Hupa in a broad phonetic orthography in which he distinguished the principal allophones of the Hupa phonemes as indi- cated below in square brackets:


a [a]; [a] before y

a- [a-]

e [8] (occurs only before a laryngeal)

e- [e] before y; [e] elsewhere

/ [i] before y or a palatal stop (gy, ky, ky); [v] before W; [u] before w; [i]

elsewhere o [o] before laryngeal; [o] before y; [v] elsewhere o- [o]

Sapir usually noted a (light) aspiration of final long vowels: tse- 'stone' [ts'e'], sometimes putting the length mark in parentheses [ts'e()'].


b [b] Unaspirated ("intermediate") bilabial stop. Rare. c [ts'] Voiceless aspirated alveolar affricate. c' [ts'] Glottalized alveolar affricate.

c'" [tc'w] Voiceless alveo-palatal affricate with labialized aspiration. c' [tc'] Glottalized alveo-palatal affricate.

d [D'] or [d'] syllable-finally, [d] elsewhere. Unaspirated ("intermediate") alveolar stop; lightly aspirated finally.

32 XIV Northwest California Linguistics

g [g] Unaspirated (or "intermediate") mid-velar stop; lightly aspirated finally.

g^ [§] Unaspirated (or "intermediate") front- velar stop; lightly aspirated finally.

G {^,] Unaspirated (or "intermediate") back-velar stop; lightly aspirated fi- nally.

h [h] initially, ['] elsewhere. Strong aspiration.

k [k'] Aspirated mid- velar stop.

K [k'] Glottalized mid-velar stop.

k^ [If'] Aspirated front- velar stop.

^- [k'] Glottalized front- velar stop.

/ [1] syllable-finally, [1] elsewere

i [1] Voiceless lateral spirant.

A' [tl'] Glottalized lateral affricate.

m [m] Bilabial nasal.

n [n] finally; [n] elsewhere. Alveolar nasal.

(/ [q] "~ Velar nasal.

Q [q'l or [k' ] Glottalized back-velar stop.

s [s] Voiceless alveolar sibilant.

s [c] Voiceless alveo-palatal sibilant. Rare.

t [f] Voiceless aspirated alveolar stop.

/' [t'] Glottalized alveolar stop.

w [w] syllable-fmally; [w] elsewhere. Voiced bilabial semivowel.

W [W] Voiceless bilabial semivowel.

X [x] Voiceless velar fricative.

^w [^x*] Labialized voiceless velar fricative.

y [i] finally (as the second element in a diphthong); [y] initially; [y] finally where /yi/ is the underlying phonemic sequence. Voiced palatal semi- vowel.

J [ts'] or [Dz'] finally; [dz] elsewhere. Unaspirated (or "intermediate") alveolar affricate; lightly aspirated finally.

J [tc'] or [Dj'] finally; [dj] elsewhere Unaspirated (or "intermediate") alveo-palatal affricate; lightly aspirated finally.

? ['] - Glottal stop.

Prosodic features

Sapir used an acute accent to mark the syllable of a word or phrase on which primary stress falls. Since the position of the stressed syllable is predictable

Hupa Texts: Orthographic Key 33

from the (underlying) phonological structure of the phrase, it is not marked in the phonemic orthography employed here.

Two other prosodic features marked by Sapir are not predictable, and are marked in the present orthography:

A circumflex accent on a vowel indicates a syllable with noticeably higher pitch than preceding syllables. The function of this cadence is not clear, + A plus immediately following a vowel or consonant indicates an unusual prolongation of that phoneme (noted mainly in exclamations or song vocables).

Alphabetical Order

The order of the Hupa phonemic alphabet is as follows: '^, a, a-, b, c, c', c'\ c\ d, e, e\ g>\ g, G, h, i, k>, k, k>\ U, I, i,