CTLF Corpus de textes linguistiques fondamentaux • IMPRIMER • RETOUR ÉCRAN
CTLF - Menu général - Textes

Boas, Franz. Race, Language and Culture – T39


The Social Organization of the Tribes
of the North Pacific Coast 11

The variety of forms of social organization found among the tribes
of the coasts of Alaska and British Columbia has given rise to extended
discussion which relates to fundamental questions regarding the
theory of the growth of social institutions. 22 In the extreme north we find
a purely matrilineal clan organization, while in the extreme south we
find village communities with a loose family organization with bilateral
descent in which, however, preference is given to paternal descent. In
the central regions a mixed type is found in which descent in the female
line is obtained by the transfer of privileges from a man to his son-in-law.370

During the past winter I had an opportunity to study the little-known
coast tribes of the central part of British Columbia. The conditions
found among them throw additional light upon the probable historical
development of the cultural life of this area.

The matrilineal organization of the northern tribes has been described
by Swanton, Barbeau and myself. The Tlingit and Haida are divided
into two exogamic groups each of which embraces a large number of
localized sibs. It seems probable that the number of these exogamic
groups was larger at an earlier time. Among the neighboring Athapascan
tribes a three-group division has been recorded, 13 and among the
Tlingit Swanton furnishes definite information of the existence of a
small group which may intermarry with the two main divisions, 24 — in
other words, of a group which forms a third unit. Swanton suspects the
existence in past times of a similar group among the Haida. 35 One of the
chief differences in the Tsimshian organization, as compared to that of
the Haida and Tlingit, is that we have here four exogamic groups instead
of the apparent dual division among their northern and western neighbors.
As among them, the exogamic division does not form a unit, but
consists of a number of well localized sibs.

The fundamental idea of exogamy of the matrilineal divisions underlies
the organization of all these tribes.

There are, however, evidences that the fundamental concept of sib
relationship is not the same in all these groups. These differences are
expressed in the systems of terms of relationship, in regard to which the
Tsimshian differ very much from the Tlingit and Haida. The terminology
of all three, however, has in common the trait that parallel cousins
are considered as brothers and sisters, while cross-cousins belong to the
group into which one may marry, and are designated by a separate term.
In the paternal generation different terms are used for the individuals
on the father's side as against individuals on the mother's side. In the
generation of grandparents, and that of grandchildren, no distinction is
made according to the divisional affiliations of individuals. The Tsimshian
system is characterized by a prevalence of reciprocal terms in one's
own generation.

South of the Tsimshian live the tribes of Gardiner Inlet and Douglas
Channel, which speak a Kwakiutl dialect closely akin to the Bella Bella.
371My information in regard to these tribes is very fragmentary. It is based
on information obtained from two individuals whom I happened to
meet at Bella Bella. They have five divisions, four of which correspond
to the Tsimshian divisions, namely, Eagle, Raven, Wolf and Killer
Whale. The last of these corresponds to the Gispawaduweda of the
Tsimshian. Besides these they have the Beaver, which among the Tsimshian
and Haida is an important crest in the Eagle group. The existence
and the number of these five divisions were corroborated by Bella Bella
informants. I was told that the five divisions are exogamic and that the
child belongs to the mother's side. One of my informants, however, told
me that her own children, as they were growing up, had been placed in
different divisions by being given names belonging to the sides of one or
the other of the four grandparents. Nevertheless she claimed that even
after changing the position of the child the laws of exogamy continued,
and that the children counted as members of the division in which they
had been placed. I cannot give a definite statement in regard to this

The conditions among the Bella Bella have been described by Farrand
and myself. In a previous statement I said that the Bella Bella have
three divisions, Eagle, Raven, and Killer Whale. Farrand adds to these
the Wolf. My inquiries during the past winter brought out the fact that,
of old, the northern Bella Bella had actually only three division and that
the southern Bella Bella represent the Wolf group. The idea that these
four sides as such are localized appears here very much more clearly
than among the three northern tribes. According to the concept of the
Bella Bella, the ancestor of any local unit descended from the sky or
sprang up from the ground, and the ancestral tradition shows to which
one of the divisions he belonged. If he came down in the form of an
eagle, or had some other association with an eagle, he would belong to
the Eagle clan, and so on. The Wolf clan is definitely associated with
the village Hauyad. A woman who plays a most important part in Bella
Bella mythology married a wolf and her descendants form the Wolf
clan. The eldest of her children assumed the role of transformer and
culture hero. The tradition is important for all the divisions of the Bella
Bella. The song of the wolf children is the marriage song, and the
mourning song of the mother is the funeral song in the ceremonies of all
the divisions of the Bella Bella tribe.

There is no rule of exogamy connected with the fourfold division of
the tribe. We find intermarriages between individuals of the same divisions
372not only at the present time, but also in descriptions of occurrences
of an earlier period. Some elderly Bella Bella expressed themselves very
clearly in regard to their concept. They said: “The northern tribes
make a great mistake. Who has ever seen a wolf mating with an eagle?
It is right that an eagle should mate with an eagle.” Although they are
perfectly familiar with the customs of the northern tribes, the idea of
exogamy is entirely foreign to them. They favor local endogamy among
the nobility in about the same way as is done by the Bella Coola, and as
is also found exceptionally among the Kwakiutl. In speaking about the
relations of these divisions they merely say that all members of one particular
division visiting a distant village will be welcomed by their
“friends,” that is to say, by members of the division that bears the same
name, no matter whether these are Bella Bella or Tsimshian or other
northern tribes. I did not hear that they were aware of the absence of
the fourfold division among the northern tribes.

Since these divisions do not form exogamic units, and since, furthermore,
endogamy is only favored, not by any means enforced, the divisions
are fairly evenly distributed over the whole territory. Nevertheless,
the opinion is general that the Wolves belong to the southern Bella
Bella tribe.

The primary position of an individual is definitely with his mother's
division. However, position is not by any means permanent, but in the
same way as among the Kwakiutl a person may take his father's or his
grandfather's position. For this reason the affiliations of an individual
may change as he rises in rank.

It is interesting to note that the terminology of relationship which
underlies the social system of the Bella Bella and of the more northern
Kwakiutl tribes is the same as that of the Kwakiutl proper. In this system
no distinction is made between collateral relatives in maternal and
paternal lines. Father's and mother's brothers and father's and mother's
sisters are designated by the same terms, and the same is true in regard
to brother's and sister's children. While in general the terms of relationship
are the same among all the tribes of Kwakiutl lineage, two terms
show considerable variation according to dialect. These are the terms
for uncle, (both maternal and paternal,) and for brother-in-law, (both
wife's brother and sister's husband).

It seems that farther to the south the system of matrilineal descent
with a small number of divisions never exceeding five, disappears completely.
The tribe of Rivers Inlet speaks the Bella Bella dialect, but so
373far as I have been able to discover from indirect reports, there is no trace
of matrilineal clan organization found among them. From Rivers Inlet
southward we find throughout tribes composed of small units, and derived
through descent from a single ancestor and from other individuals
who at an early period associated themselves with him. The number of
these units in each tribe is quite large. Preference is given to paternal

The general condition on the North Pacific Coast may be described
as follows: In the north we have a group of tribes in which maternal
and paternal lines are clearly distinguished and where we find a small
number of divisions, from two to five, with definite functions regulating
marriage, the matrilineal clans being exogamic. The most southern
group of these tribes, the Tsimshian, have four clans, the Eagle, Raven,
Wolf, and Bear (Killer Whale). Further to the south, the Bella Bella
have the same clans that are found among the Tsimshian, but they lack
entirely the function of regulating marriage, and the idea that intermarriage
between two members of the same clan is incestuous is entirely
foreign to the thoughts of the people. The clans have a political function
determining friendship or enmity between groups. We find also
that maternal descent prevails, although it is not rigidly adhered to in so
far as there is great freedom in assigning to an individual in later life a
position in any one of the clans to which his ancestors belonged. The
contrast between the terminology of the systems of relationship and the
clan organization with preference of the maternal line is quite striking.

When we direct our attention primarily to the village communities
of the Bella Bella, the organization is decidedly similar to that of the
Kwakiutl. The whole tribe is found to consist of a great many local
units, each of which claims certain privileges on account of its descent
from an ancestor who came down from the sky or appeared in some
other supernatural way. The same is true of the northern matrilineal
tribes, except that the character of the traditions of the local units stresses
the encounter of the ancestor with a supernatural being.

I have pointed out that the organization of the Kwakiutl is identical
with that of the northern coast Salish tribes. However, the idea that
local units have certain privileges, is much less developed among the
coast Salish tribes.

We might, therefore, describe the whole situation in the following
way: As we go northward from the State of Washington, the idea of the
unity of the village community becomes more and more intimately associated
374with certain privileges which may be described as crests. When
we reach Bella Bella we find overlying this system a system of a small
number of clans which are identical in name with the clans of the northern
matrilineal tribes. The sameness of the clan names can be due only
to historical connection. The four clans are almost functionless as compared
to the functioning of the village communities. Connected with
the occurrence of the clans, the idea of maternal descent prevails. The
emphasis on matrilineal descent is quite contrary to the linguistic forms
used among these tribes. Still farther to the north the emphasis laid upon
local units or village communities persists, but the communities are
strictly subordinated to the exogamic clan and the position of the individual
is absolutely fixed, both in regard to the local unit and to the
clan to which he belongs. Changes occur rarely, and then only by
formal adoption.

We might consider in the same way the clan system of the north, and
follow its characteristics southward. The characteristic exogamy of the
large tribal divisions dwindles down and disappears as we reach the most
southern tribes. In some regions it gives way to a marked tendency to
endogamy. In the north the local units are definitely assigned to one
or the other of the larger divisions; in the south they form units that
are the more independent the farther south we go.

The general conditions among the Bella Coola fit in well with the
distribution just described. They represent an isolated branch of the
Coast Salish tribes, which are organized in village communities. Among
the Bella Coola we have village communities with privileges quite analogous
to those of the Bella Bella, and with a prevalence of paternal
descent. There is no grouping of these units in larger divisions and, as
among the Bella Bella, a tendency to endogamy among the nobility

From my earlier studies of the distribution of types of social organization,
on the North Pacific Coast, I have concluded that the transmission
of social position to the daughter's son which is found among the Kwakiutl
has developed through the influence of the northern tribes, from
whom the Kwakiutl obtained the concept of crest privileges, and that
the stimulus to this development lay in marriages between the men of
the southern tribes and women of the northern tribes. This conclusion
is strongly corroborated by the conditions found among the Bella Bella,
who show a type intermediate between that of the Vancouver Island
Kwakiutl and the Tsimshian.375

The matrilineal clan is fully developed among the tribes of northern
British Columbia and of the coast of Alaska. As we go southward it
loses more and more in significance until it finally disappears entirely.
On the other hand the village community with bilateral descent but
with an inclination to favor the paternal line is most fully developed in
the southern part of the North Pacific Coast. Although it continues to
be an important element in the north, but with matrilineal descent, it is
subordinated to the matrilineal clan organization.

The following table shows these conditions at a glance.

tableau Coast Salish | Kwakiutl | Bella Bella | Northern Tribe | Kinship terms | Bilateral | Unilateral | Descent | Patrilineal | Patrilineal with transfer of privileges to daughter's son. Privilege of change to other ancestral lines | Matrilineal with privilege of change to other ancestral lines | Marriage | Preference to village exogamy among nobility | Exogamy for obtaining new privileges. Endogamy for retaining highly valued privileges in family | Crests and traditions | Privileges of ocal units | Family tradition f chief's family. n the north soradically eak crests | Matrilineal clans | Absent | Present, from two to five

I may add here notes on some of the characteristic traits of the political
and religious organization of the Bella Bella. The tribe is divided
according to rank into a number of classes. At the head of the local community
are two head-chiefs of equal rank who are considered the descendants
of the mythical ancestor of the local unit. They are called
gyā'laxa, which may be translated as “the first down.” According to the
explanation given to me by several Bella Bella, this is not now interpreted
as the first ancestor to come down from the sky, but as “the first
to receive presents in a potlatch.” The existence of two head chiefs
explains a peculiar institution among the Kwakiutl for which I have
376never before been able to obtain an adequate explanation. There are
in each tribe a small number of individuals who are the first to receive in
a potlatch. They are called kwēku, that is eagle, or g˙ā'laxa, 16 first to receive.
They are not considered chiefs, and the Kwakiutl are unable to
explain the origin of their privilege except by the reference to a myth in
which it is told that the ancestors of certain units received their gifts in
order. These positions may be survivals of an older head chieftaincy
which has been superseded by a class of nouveau riche, who now form
the aristocracy of the tribes and claim the highest rank and heavenly

The second class among the Bella Bella are the chiefs (hēf'ɛmas). Next
to these are the nobility (ō'ɛma). The common term for chief which is
used by the Kwakiutl (g˙ī'g̣ămeɛ, stem g˙īg̣) 17 is not known to the
Bella Bella. (Another word g˙īg̣ămeɛ occurs in Bella Bella and in Kwakiutl,
but is derived from the stem g˙ī- to be in a certain position; g˙ī'g̣ămeɛ,
standing in front.) The term ō'ɛma is used by the Kwakiutl for a
chief's wife, who is designated by the Bella Bella as k˙a'nił. The Kwakiutl,
however, use the term ō'ɛmayu to indicate high social position,
greatness in a social sense, so that it would seem that this term also had in
former times a more general meaning among the southern Kwakiutl
tribes. The fourth class among the Bella Bella are the g˙a'lg̣em, and the
lowest group are the xā'mala. In Kwakiutl this word means orphan,
and is used as an opprobrium, for orphans are of low rank because they
are not helped by their parents to rise according to the regular scale of

This organization is, to a certain extent, connected with the organization
of the tribe during the winter ceremonial. Only the two head chiefs
can become cannibal dancers. A number of the more important dances
of the winter ceremonial belong to the chiefs and to the nobility. In the
winter ceremonial the seats of the two head chiefs are in the middle of
each side of the house. The management of the winter ceremonial is not
in the hands of the head chiefs, but is controlled by eight members of the
nobility who form a council, and who arrange the winter ceremonial for
each year.

The enormous complexity and confusion in the arrangement of the
winter ceremonial at Fort Rupert seems to have arisen through a confusion
between the family ceremonials of the Bella Bella and the winter
377ceremonial. Although we do find among the Kwakiutl a distinction
between the family ceremonial and the winter ceremonial, many traits
of the former have been transferred to the winter ceremonial. The
number of ceremonies (or dances) among the Bella Bella is limited.
There are essentially only four groups of the winter ceremonial. These
are from the lowest to the highest: the ō' lala, the q!ō'minoqs, the tā'nis,
and finally the nō'nłtsista. These are mutually exclusive, and the dance
house used by one group is taboo for members of the lower groups. Besides
this, all the families have their family ceremonials, the lewe'laxa.
Initiation into one of these is entirely distinct from initiation into the
winter ceremonial.378

11 American Anthropologist, N.S., vol. 26 (1924), pp. 323-332.

22 I mention here a few of the more important publications containing data and

Leonhard Adam, “Stammesorganisation und Häuptlingstum der Tlinkit Indianer,”
Zeitschrift für vergleichende Rechtswissenschaft, vol. 29 (1912), pp.
86 et seq.
“Stammesorganisation und Häuptlingstum der Haida und Tsimshian,” Ibid.,
vol. 30 (1913), pp. 161-268.
“Stammesorganisation und Häuptlingstum der Wakashstämme,” Ibid., vol.
35 (1918), pp. 105-430.

G. M. Barbeau, “Growth and Federation in the Tsimshian Phratries,” Proceedings of the 19th International Congress of Americanists (Washington, 1917),

Franz Boas, “The Social Organization and the Secret Societies of the Kwakiutl
Indians,” Report of the United States National Museum for 1895 (1897), pp.
“Der Einfluss der sozialen Gliederung der Kwakiutl auf deren Kultur,” Internationaler
Amerikanisten-Kongress, 14 Tagung
(Stuttgart, 1904), pp. 141-48.
“Tsimshian Mythology,” 31st Annual Report of the Bureau of American
(1916), pp. 22-1037.
“The Social Organization of the Kwakiutl,” American Anthropologist, N.S.,
vol. 22 (1920), pp. 111-126, pp. 356-369 of this volume.
“Ethnology of the Kwakiutl,” 35th Annual Report of the Bureau of American
(1921), pp. 43-1481.

Edward Sapir, “The Social Organization of the West Coast Tribes,” Transactions
of the Royal Society of Canada
, Series 3, vol. 9 (1915), pp. 355-374.

John R. Swanton, (a) “Social Conditions, Beliefs and Linguistic Relationship of
the Tlingit Indians,” 26th Annual Report of the Bureau of American Ethnology
(1908), pp. 391-485
(b) “Contributions to the Ethnology of the Haida,” Jesup North Pacific Expedition,
vol. 5 (1905).

31 See discussion and literature in Franz Boas, Tsimshian Mythology, pp. 478-480.

42 See Swanton (a), 309, 409.

53 Swanton (b), 90.

61 , are pronounced somewhat like gy, ky; g is a g pronounced at the soft
palate. See note 1 on p. 232.

7 Voir note 6.