In 1910 C. Hart Merriam, already well known as a naturalist, came to California and began the study of California ethnography which was to occupy him for the rest of his life. Almost every year from then until his death in 1942 Merriam spent about six months in the field, talking to Indians and recording their memories of aboriginal times. All this field work resulted in an immense collection of data on the California Indians, most of which has never been published (see Merriam's bibliography in Merriam, 1955, pp. 227-229).

In 1950 the greater part of Merriam's field notes was deposited at the University of California, with the intention of making them available for study and publication. One volume of papers has already appeared (Merriam, 1955), and the present study is part of a continuing program.

The California Athabascans were selected as the first group for study at the suggestion of A. L. Kroeber, the reason being that the Athabascans have been and still remain one of the least known aboriginal groups in the State. This is not because they were conquered early and their culture dissipated, as is true of the Mission Indians; there were scarcely any whites in the California Athabascan area before the 1850's. Indeed, as late as the 1920's and '30's there were many good Athabascan informants still available. The reason for the hiatus in our knowledge lies in an accident in the history of ethnology rather than in the history of California.

The early work among the California Athabascans was done by Pliny Earle Goddard. Goddard began his studies of the Athabascans in 1897 at the Hoopa Indian Reservation, where he was a lay missionary. He stayed there until 1900, when he went to Berkeley to work for his doctorate in linguistics under Benjamin Ide Wheeler, President of the University of California. Between 1900 and 1909 Goddard was associated with the University as student and professor and during this time he visited the Athabascans periodically, until he had worked with virtually all the groups considered in this paper.

During this same period A. L. Kroeber was engaged in gathering material for his classic Handbook of California Indians. Because of the scarcity of ethnographers in those years Kroeber could not afford the time to work in the Athabascan area and duplicate Goddard's investigations. Kroeber did study the Hupa and the Kato at either end of the Athabascan area but, except for a hurried trip through the region in 1902, he did not work with the other groups, and the responsibility for the ethnographic field work therefore devolved upon Goddard.

Goddard, however, was not primarily an ethnographer but a linguist, and he directed his chief efforts toward linguistic investigations. He has published an impressive body of Athabascan texts and linguistic analyses but, except for his Life and Culture of the Hupa (1903a), almost nothing on the culture of the Athabascans.

The net result is that the California Athabascans are virtually unknown, and Merriam's fresh data provide an opportunity to piece together the available evidence.

The Merriam files, deposited at the Department of Anthropology of the University of California, contain information on each of the tribes of California, some of it being information gathered by Merriam himself, the rest clippings and quotations from various historic and ethnographic sources. The primary and secondary materials are easily distinguished, since Merriam gave scrupulous citations to his sources.

Merriam's own data consist of word lists, ethnogeographical material, and random notes on various aspects of native culture. I have not used his word lists, since their usefulness is primarily linguistic and I am not competent to perform the necessary linguistic analysis, but all the random ethnographic notes which he recorded for the Athabascan groups are here included under the discussion of the appropriate tribes.

Most of the Merriam Athabascan material is geographic, consisting of lists of villages and place names, of descriptions and lengthy discussions of tribal boundaries. Obviously Merriam attempted to gather a complete file of this sort of information, and he was largely successful. His work provides a good basis for establishing boundaries and for locating tribelets and villages.

Another important source of information, serving the same purpose, is the Goddard material. Evidently Goddard very much enjoyed the long horseback trips he made with an informant, who could point out the village sites, landmarks, and other points of interest of his native territory. This information, carefully recorded by Goddard, has proved extremely valuable in the present work, the more so since it represents firsthand observation.

Goddard's ethnogeographic work for three of the California Athabascan groups has already been published (1914a; 1923a; 1924). Besides this, the present writer has been fortunate enough to have access to Goddard's unpublished notes, which contain information on several hundred additional villages in the area. These notes were in the possession of Dr. Elsie Clews Parsons, Goddard's literary executor, and on her death they were sent to the University of California by Dr. Gladys Reichard. They remained in the files of the University of California Museum of Anthropology until their use in the present work.

This unpublished material of Goddard's consists of a group of file cards, on each of which is typed the name, location, and any other pertinent data for a single village. Some of the lists are accompanied by maps, showing precise location of the villages. In the lists for which there are no maps but only verbal descriptions of the sites, the township, range, and quarter section coÖrdinates are given. The township and range coÖrdinates have been changed since Goddard's time, in accordance with the more accurate surveys of the last thirty years, but county maps of the appropriate period provide a perfectly adequate way of locating Goddard's sites within a few hundred yards.

It is clear, on the basis of internal evidence, that there is or was more Goddard material than is now accessible to the present author. For the Kato, for instance, Goddard says that he recorded more than fifty villages (Goddard, 1909, p. 67); all that remain in his notes are two village cards numbered 51 and 52 respectively. There may also be some data, once recorded but now lost, from the Lassik, Nongatl, and Shelter Cove Sinkyone. I have communicated with the American Museum of Natural History, where Goddard was a member of the staff, and with Indiana University, where some of his manuscripts are deposited, but neither of these institutions has any knowledge of the material in question.

The Merriam and Goddard material, taken together, provides a fair amount of information on the geography of the California Athabascan groups. We are now in the position of knowing a great deal about the location of the tribes, tribelets, and villages of these people, while we know very little about their way of life, except what can be gained by inference from the surrounding groups.

The author's thanks are due to Dr. A. L. Kroeber and Dr. R. F. Heizer, who gave their full coÖperation throughout the preparation of the present paper. Dr. Henry SheffÉ was kind enough to advise on the statistics used in the section on population.

ATHABASCAN CULTURE

The following sketch of Athabascan culture attempts to provide some background for the later discussion of the various groups. In this sketch I have not used the material from the Hupa, since they are virtually identical with the Yurok and not at all typical of the more southern Athabascans.

Subsistence.—For information on Athabascan economy I have relied heavily on Essene's account of the Lassik (1942, p. 84). There was, no doubt, variation among the different groups, but for the most part, they must have followed a similar pattern.

The most difficult time in the annual cycle of food production was winter. There were then few fish and almost no game animals or crops for gathering. From late November to early March people had to rely on food that had been stored the previous year. Essene's informant said that about every four or five years there would be a hard winter, but she could remember only one when people actually starved to death.

In February or March the spring salmon run began, and after that the danger of starvation was past. At about this time the grass began to grow again, and the first clover was eaten ravenously because of the dearth of greens during the winter.

The herb-gathering and salmon-fishing activity lasted until the spring rains ended in April or May, when the people left their villages on the salmon streams and scattered out into the hills for the summer. Usually only a few families would stay together during the summer, while the men hunted deer, squirrels, and other animals and the women gathered clover, seeds, roots, and nuts. Food was most plentiful at this season, and the places visited varied with the abundance of different crops. If a certain crop was good, the Indians would spend more time that summer in the area where the crop grew best. The next year they might go somewhere else. The vegetation of the Athabascan habitat is not well enough mapped to permit a precise delineation of these various summer camping grounds.

In September or October, when the acorns were ripe, the Indians would return to their winter villages and smoke meat for storing and probably store the acorns. Each family built a new house to protect it from the heavy winter rains. After the first rain in the fall the salmon run again in some of the streams of the region and were caught and smoked for winter storage.

It is evident that the crucial factor in the economy was the amount of food stored for winter and that this food supply was a controlling influence on the size of the population, since, in bad years, people starved. At least, this was so for the Lassik, and it was no doubt true among the other groups as well. Salmon, meat, and acorns were doubtless the chief foods stored, and thus population size would have responded quite sensitively to the quantity and condition of the salmon, deer, and oak trees.

Social organization.—For social organization I have had to rely mostly on Nomland's accounts of the Sinkyone and Bear River groups (1935, 1938). The primary social unit among the California Athabascans was the simple family, including a man, his wife, and his children. Although polygyny was known, at least among some groups, it was rare, and the possessor of two wives was reckoned a rich man. Most marriage was by purchase; the levirate and sororate were common. Divorce was also common and might be obtained by a man because of his wife's barrenness, laziness, or infidelity.

The next social group, larger than the family, was the tribelet. Kroeber (1932, p. 258) has defined the tribelet as follows.

Each of these [tribelets] seemed to possess a small territory usually definable in terms of drainage; a principal town or settlement, often with a chief recognized by the whole group; normally, minor settlements which might or might not be occupied permanently; and sometimes a specific name, but more often none other than the designation of the principal town. Each group acted as a homogeneous unit in matters of land ownership, trespass, war, major ceremonies, and the entertainment entailed by them.

This definition, given for the Pomo, fits the Athabascan area very well. Merriam usually refers to these groups as "bands," while Goddard calls them "subtribes." In the body of this paper I use the word "band" when quoting or paraphrasing Merriam, otherwise I call them "tribelets."

The tribelet was the largest corporate group in the area. A larger group, which I call the tribe, has been identified by most ethnographers. This latter group ordinarily had no corporate functions, unless it happened to be coterminous with, and therefore indistinguishable from, the tribelet. The tribe, as the term is used here, was a group of two or more tribelets—or occasionally one single group—with a single speech dialect, different from that of their neighbors. The tribe was also culturally uniform, but not necessarily distinct from its neighbors in this respect. The similarity between people of a single tribe evidently gave them a feeling of community but had no further effect on their social or political organization.

The following tribes have been identified in the Athabascan area, each including several tribelets, except for the Bear River tribe, which consists of one single tribelet.

Kato: The Kato probably included at least 2 tribelets, but we have no information on this point.

Eel River Wailaki: 9 tribelets.

North Fork Wailaki: 6 tribelets.

Pitch Wailaki: 4 tribelets.

Lassik: Probably several tribelets, but there is no information.

Nongatl: There is evidence of 6 subgroups of the Nongatl. Some of these may be dialect divisions, that is, tribes. The information is not sufficient to permit definition and they have therefore been grouped under Nongatl. The extent of Nongatl territory indicates that there must have been several tribelets.

Lolangkok Sinkyone: There were at least 2, and possibly more, tribelets.

Shelter Cove Sinkyone: There were at least 4 tribelets.

Mattole: 2 tribelets.

Bear River: The Bear River tribe consists of a single tribelet.

Whilkut: The 4 subdivisions of the Whilkut—Chilula Whilkut, Kloki Whilkut, Mad River Whilkut, and North Fork Whilkut—all appear to be tribelets. It is possible that the Mad River Whilkut spoke a different dialect than the other groups and, if so, they should be given tribal status. The evidence is not clear on this point and I have therefore included them simply as a Whilkut tribelet.

Hupa: 2 tribelets are to be distinguished for the Hupa proper. In addition, Merriam distinguishes the South Fork Hupa as a distinct dialect division. The linguistic separation is not supported by Goddard or Kroeber and I have therefore included the South Fork Hupa under the Hupa proper, but as a separate tribelet. This gives a total of 3 tribelets for the Hupa.

In general, it may be stated that the California Athabascans did not have the strong local organization characteristic of Central California. Emphasis on wealth, although present, was less strongly developed than among the Yurok and therefore did not lead to the fragmented villages and tight family organization of that group. This statement, of course, does not apply to the Hupa, and probably not to the Whilkut, both of which were more like the Yurok.

Religion and the supernatural.—The clearest account of the religious practices of the Athabascans is given by Nomland (1938, pp. 93-98), who obtained her information from the Bear River woman, Nora Coonskin, herself a shaman. The account, however, may not be representative of the Athabascans as a whole.

The Athabascans thought that each person had a spirit which, leaving him when he died, might come back to earth as a small creature about two feet high. This returned spirit could communicate with shamans. When a person had a fainting spell, the spirit departed from the body and a shaman had to be called in order to get the patient's spirit back. If the shaman failed, the patient died. Shamans' spirits went to a special afterworld and were accompanied only by the spirits of other shamans.

Shamans were important among the Bear River people and probably among the other Athabascans as well. They might be either men or women; most often they were women, men being thought less powerful. The first signs of a shaman's power came in childhood, the visible signs being, for example, excessive drooling in sleep. If the childhood omens were proper, the training began about the age of twelve, under the direction of an older shaman, the main ceremony being a series of dances performed on five successive nights. Other ceremonies followed; then the girl was a full-fledged shaman. She was not supposed to use her power for a period of two to five years or it would harm her. The fee for training the initiate was large, 200 to 300 dollars in Indian money (perhaps a 6-8 ft. string of dentalia shells).

There were two types of shamans—curing shamans and sucking shamans. The curing shaman sang and danced for two nights while her spirit searched for the spirit of the patient. A shaman's fee was from five to ten dollars per night; if the patient died within two months, the fee had to be returned.

The sucking shamans could suck out pains which were causing illness. These shamans were paid more because they were more powerful; having greater power, they were in greater danger and had a shorter life expectancy.

Connections with other groups.—The foregoing account of economy, social organization, and religious practices does not by any means make up a complete picture of Athabascan life, but it illustrates certain salient factors. In particular, the connections with Northwestern California are clear. So far as influence from Northwestern California is concerned the Athabascans may be divided into three groups: the Hupa and Whilkut on the north are an integral part of the northwestern culture center; the Wailaki and Kato on the south are essentially Central Californian; and the groups in between are transitional, but more northern than southern in their outlook.