Introduction to the Papers on Central Coast Archaeology

Philip M. Hobler

The 2 papers in this volume are the Fraser University has been fortunate in result of the long term work on the being part of the original organization Central Coast of by the of the University. When Roy Carlson Department of Archaeology at Simon began teaching there in 1966 his prior Fraser University. The papers are based recommendations had already resulted in upon graduate theses presented to the the construction of an archaeology lab­ department in 1976 (Chapman) and 1977 oratory and provision for courses. I (Apland) . They have been abridged joined the faculy in 1967 after two somewhat but have not been updated or years of palaeolithic work in the east­ revised. ern Sahara. It could safely be said that I was unspoiled by previous exper­ Since these are the first reports on ience with maritime archaeology. Central Coast archaeology to be publish­ Shortly afterward I began a search for ed in this series it is in order, by way an area of British Columbia that would of introduction to review the history of be suitable for a long-term programme of the department's work in the area. field teaching and research. Compared Geographically the area begins with the with the interior, the more advanced north end of and asso­ cultural development and presence of ciated mainland coast. It extends deep stratified sites made the maritime northward some 400 km across the Queen region seem more attractive. Even at Charlotte Sound to a point south of that time a fair amount of archeological (Fig. 1.1 ). The topog­ work had been done on the southern coast raphy varies dramatically on any east- by local institutions. On the northern west transect from low lying outer coast coast, National Museum of Canada pro­ to the inner coast with its deeply in­ jects were underway near Prince Rupert cised fjord-like inlets. Historically and on the Queen Charlotte Islands. In the area includes the traditional terri­ between, the Central Coast remained a tory of the Bella Coola, parts of the rather large, unstudied area. Further southernmost Haisla, the Hieltsuk, and encouragement to consider the Central much of the southern Kwakiutl area. Coast was provided by a gasoline company road map that deceitfully promised ease Some 10 field seasons, beginning in of access in the form of a major highway 1968, have seen Simon Fraser University leading westward to Bella Coola from sponsored archaeological projects on the Williams Lake. Only later did we dis­ Central Coast. Archaeology at Simon cover the degree of exaggeration

1 2 CENTRAL COAST ARCHAEOLOGY

Fig. 1.1 Map of Central Coast showing excavated sites. INTRODUCTION 3

LlJ o I - < i_ a 4- ct Qd Qd — i ct LU LU o LU UJ LU no * X 4- > LL _l o > > z LU < LU Qd or Qd -J 1- 3 z >- < L— (- h X 1— < 5 Q ft: < o LU o o Od < < < Q 3 3 5 3 —\_/ LU z z z Z LlJ z O o o O *5 LU 1- h- 1- T 1- IS) IS) IS) IS) 3 z x I < _L_ < z <1 < < \~ IS) < < $ $ $ < $ < o i X LU I-* ¥ * o X. * * * Z s 5 o O CD X

YEARS CVJ O _ G) if ) S —. __ o LO CO GO in 3 3 3 3 3 L- l k_ X n X X a> 3 b.p. CO CO CO CO CO co ( 0 co CO CO l~ ( 0 CO I-* CO CO o o O -Q o a> Q) a> a> — -Q o

0 • 1 1 > __ 1__ l_ 1— I » i i n • i * — 1 n LJ +n + i i I 00 1 1 • • i — • n i * i j • • t 1 i * 200 i i i i ■ • i i i i 1 i * i i i i i • i

1000 •

• 1 1 1 1 • f 1 * i 1 • • 1 • 1 1 1 l 2000 i 1 I i * 1 1 1 l i . ' • l i i 1 • i i i i i 1 1 i i i 1 1 1 1 i 3000 i 1 1 1 i i i • 1 i i 1 1 1 1 i 1 * 1 1 i i 1 1 i | 1 1 « 1 i 1 1 • ( i i 4000 i i ■ i i 1 1 l i i i > i i i l ■ ■ i l i 1 i i N i i i i 5000 1 1 1 i : : i i i » i i i i 6000 • i i i i i i i i i ■ 1 i i ■ i 7000 i • i * i i i i | I 8000 1 i 1 1 1 i 1 • l i 9000 f i » i 1 0 0 0 0 II 000

t CENTRAL COAST EXCAVATIONS NOT SPONSORED BY SIMON FRASER UNIVERSITY

Fig. 1.3 Approxima te time range represented by excavated sites 4 CENTRAL COAST ARCHAEOLOGY

Fig. 1.2 Map of surveyed sites in the Bella Bella-Bella Coola region. INTRODUCTION 5 concealed in that broad red line. The application of research goals appropriate in other archaeological ar­ Three ethnographic factors define the eas might have been non-productive. For background for our archaeological in­ example, in environmentally marginal vestigations. The first is the areas the diachronic study of subsist­ anomalous position of the Bella Coola in ence systems has been greatly aided by their isolation from other Salish speak­ studies of paleoenvironment. Often in ers both on the coast and in the such cases even small cultural changes interior. The second is the boundary on can be seen to be in phase with environ­ the east between the Bella Coola and the mental shifts. Yet, this kind of interior plateau peoples separating approach when applied to the study of groups speaking unrelated languages and Central Coast shell middens has proven with contrasting cultures and to be of only limited explanatory value environments. Third, is the Bella (Hester and Nelson 1978). Archaeologi­ Coola-Hielt suk boundary on the west a cal studies seem to have progressed case of mutally unintelligible languages through the greatest number of develop­ but a sharing of cultural traits and mental stages in those areas where environment. These circumstances of inquiry has been underway longest. varying culture, language, and environ­ Compared with most New World areas, ment during the ethnographic period set archaeological work of any scale on the the stage for our initial archaeological Northwest Coast has been in motion only investigations by posing a host of ques­ a brief time, 30 years on the southern tions concerning prehistoric cultural coast and half that elsewhere (Carlson traits, their distributions, origins, 1978). Perhaps we can shorten the and relationship to environments. stage-by-stage development of research orientation but there is no wisdom in Work began with a data gathering attempting to skip whole stages. It has orientation. Since so little was known been and will be a difficult step-by­ of the archaeology of the area it was step process. In the end the data will not thought advisable to formulate more be our best guide and will determine complex research goals. Productive re­ what can be learned. search must involve an interdependence between the field data and the questions Archaeological projects on the Cen­ we ask since archaeological data from a tral Coast sponsored by Simon Fraser given area are not capable of providing University are listed in chronological illuminating answers to every question order in Table 1.1. The approximate time that might be asked. This has been range represented at excavated sites is recently stresed by Trigger: given in Figure 1.2. My reconnaissance survey in 1968 laid the groundwork for Archaeologists must learn much of our subsequent field efforts. to live with the realization Sponsored by the S.F.U. President's that their desire to study Research Grants fund, the 1968 survey whole cultural systems cannot provided initial familiarity with the be realized .... Archaeologists Central Coast, its archaeological sites, must learn to ask the kinds of and its rather special logistic require­ questions with which the data ments. Almost all of the sites that are equipped to deal were later the focus of large excava­ (Trigger 1978:151). tions were found in the 1968 survey. Also in 1968 an archaeological survey In short, one must know what the data project in the Bella Bella area west of are like before it is possible to see Fisher Channel conducted by J. Hester the problems and formulate the questions was sponsored by the University of that can guide further work. 6 CENTRAL COAST ARCHAEOLOGY

TABLE 1-1 Central Coast Archaeological Projects sponsored by Simon Fraser University

APPROXIMATE YEAR PROJECT DIRECTED BY DURATION CREW SIZE

1968 Bella Coola-Bella Bella area survey Hobler 6 weeks 3 1969 Excavations at Kwatna River Sites, FaSu 1,2,9 Hobler 8 weeks 18 1970 Excavations at Kwatna River Sites, FaSu 1,2 Hobler, Carlson 8 weeks 16 1971 Excavations at Kimsquit Sites, FeSr 1,4 ,7 Hobler 8 weeks 8 1971 Excavations at Kwatna River Sites, FaSu 1,2 Carlson 8 weeks 8 1971 Excavation at 'Port Hardy Site ... Chapman 6 weeks 7 1972 Excavations at Kimsquit Sites, FeSr 4,5 Hobler 8 weeks 8 1972 Excavations at Kwatna River Site, FaSu 10 Carlson 8 weeks 8 1972 Excavations at MacNaughton Island, ElTb 10 Pomeroy 10 weeks 5

1973 , Survey Carlson 9 Hobler 7 weeks 4 1973 Excavations at Port Hardy site ... Chapman 9 weeks 8 1974 Excavations at MacNaughton Site ElTb 10 Carlson 8 weeks 6

1974 Bella Bella area survey for intertidal Pomeroy 9 lithic sites Apland 8 weeks 2 1977 Excavat ions at Namu ElSx 1 Carlson 8 weeks 8 1977 Studies of intertidal lithic sites Hobler 8 weeks 6 1978 Excavations at Namu, ElSx 1 Carlson 8 weeks 6 1978 Excavations at FaSu 19 Hobler 8 weeks 6 1980 Excavations at FaSu 19 Hobler 8 weeks 12

Colorado. Hester's field work continued with a large polyethylene sheet. through 1970 and included excavations at Students and staff were housed in small Namu, Kisameet, and Roscoe Inlet (Hester tents, cocooned in layers of and Nelson 1978). Associated with the polyethylene. It was the first full initial Colorado survey in 1968, Anthony scale archaeological camp and a proto­ Pomeroy continued survey work in the type for several to come. We learned Bella Bella area in 1 969, 1 970, and from it that, despite warnings to the 1974. This work later became the basis contrary, it is possible to keep a large of his doctoral dissertation at Simon field crew reasonably dry and comfortab­ Fraser University (Pomeroy 1980). le in an area of such high summer Figure 1.3 shows the total archaeologi­ rainfall. The camp served us well for 3 cal survey sample for the Bella Bella field seasons. In the early winter of and Bella Coola areas of the Central 1971 it was destroyed by a massive Coast recorded by the author, Pomeroy, mudslide. Now, beneath hardening mud Hester, and others. the ruins of our camp, like those of older sites await discovery by future In 1969 the first S.F.U. field school archaeologists. on the Central Coast established a camp at the mouth of the Kwatna River about In all, 4 field seasons (1969 - 1972) midway between Bella Coola and Bella saw Simon Fraser University field Bella. The camp consisted of a large schools on the Kwatna River. I was laboratory and dining structure made by initially interested in the area because re-erecting posts and beams from a of the remarkable sites of Nutlitliqot- ruined native house and covering them lenk (FaSu 2) and Axeti (FaSu 1). INTRODUCTION 7

Nutlitliqotlenk is the largest shell well represented cultural deposit on the midden encountered on the 1968 survey. shore adjoining the waterlogged inter­ The survey surface collection indicated tidal midden. Work on this area of the that, in addition to an impressive vol­ site revealed a good sample of non­ ume of clam shells, the site was also perishable artifacts and evidence of likely to contain a sufficient sample of architectural features. artifacts to establish a material cul­ On the Kwatna River about 1 km above ture profile if not a chronology. This Axeti, a third site, Anutcix (FaSu 10), assessment proved correct. In addition, was also the subject of a fullscale Carlson's 1971 excavations turned up excavation project. Carlson's 1972 work evidence of a large uniform surface at Anutcix investigated a rectangular which he interprets as a house (Carlson surface depression of architectural ori­ 1972). These remains are fairly deep gin and probed a midden that proved to within the midden and thus are early in be deeper and more extensive than sur­ the site's history. The name Nutlitli- face evidence had indicated. qotlenk has been an enigma. Translated by Mcllwraith as "place of many boul­ Together, these 3 excavated sites ders" (1948: 21) it is today not a place represented the major settlements on the particularly characterized by boulders. lower Kwatna River. A time span cover­ Our excavations showed that in several ing about the last 2000 years is places the shell midden had initially indicated. The evidence for the early accumulated on an old surface littered portion of this span has come principal­ with boulders. These are likely the ly from Anutcix and portions of the source of the name. The boulder areas lower stratigraphic units of Nutlit- were obscured by deepening midden de­ liqotlenk. Surprisingly, a well posits early in the site's history. represented historic component from the Thus, though inappropriate for many late Eighteenth or Nineteenth Centuries centuries, the name has persisted. has not been found. Changes through time in the proportion of certain arti­ Also on the Kwatna River the site of fact types and in the presence or Axeti (FaSu 1) was found to have an absence of other types have prompted extensive waterlogged cultural deposit Carlson to suggest a division of the with perishable materials preserved in a occupational continuum of the Kwatna black anerobic mud. It was the first River sites into 2 phases, Anutcix being such site to be recorded on the B.C. the earlier and Kwatna the later Coast (Hobler 1970, 1976b). Exposed (Carlson 1973; Hobler and Carlson 1973). only during low tides, the deposit is on the upstream shore of a large island at In 1971 I began 2 seasons of excava­ the mouth of the river. Full scale work tion at Kimsquit on the delta of the took place in 1969 and 1 970. Metal Dean River close to the head of Dean tools proved inadequate for the task of Channel some 90 km by water north of excavating the delicate materials and Bella Coola. , the longest were cautiously replaced by hydraulic fjord on the west coast, is cut deeply techniques employing pump driven fire along the edge of the interior plateau. hoses and gravity fed garden hoses with At Kimsquit a noticeably drier, more fine spray nozzles. The materials re­ interior-like climate prevails. Surface covered from the waterlogged portions of water of the inlet are fresh with the Axeti include basketry, matting, rope, result that only traces of fragmented and cordage of cedar bark. A variety of clam and mussel shell are present in the large and small wooden artifacts found cultural deposits. include splitting wedges, bent fish hooks, weir stakes, and numerous pieces Two major excavations and 2 minor of uncertain function. Axeti also has a ones were undertaken during my 1971 and 8 CENTRAL COAST ARCHAEOLOGY

1972 field schools at Kimsquit (Hobler in the interior sponsored by the federal 1972a). Of special interest, site FeSr Opportunities for Youth programme. In 4 consists of 23 large rectangular de­ addition to the field schools on the pressions of architectural origin. This Dean River at Kimsquit and on the Kwatna site, though replete with architectural River a third full-scale Central Coast information, was remarkably devoid of project supported by the Opportunities artifactual materials and may have been for Youth programme was initiated at only briefly occupied. At other Kims­ Port Hardy by Chapman ( 1 972). The re­ quit sites extensive cultural deposits sults of her work are published in this revealed both native and historic manu­ volume. Chapman's work at Port Hardy factured materials but showed no was undertaken because construction evidence of subsurface architectural threatened a large shell midden. Full- remains. A small test in front of a scale excavations took place during the standing post and plank house yielded 1971 and 1973 field seasons. A good quantities of late Nineteenth and early cultural sample and stratigraphic sec­ Twentieth Century trash. Site FeSr 5, tions were obtained. Two cultural another Axeti ("occupied mound") was components were identified. A well tested in 1972. This narrow rocky out­ represented assemblage of ground stone, crop had been topped by a small plank bone, and shell artifacts was associated house in early historic times. Surface with the shell midden. Beneath the finds of pebble tools at its base shell midden in a dark essentially non­ prompted trenching which revealed a much shell matrix Chapman found a poorly older component mixed with the historic represented earlier component character­ materials. In addition to the pebble ized by flaked stone. This pattern of tools excavation produced obsidian an earlier flaked stone component under­ microblades and a leaf-shaped bipoint. lying a later shell midden deposit has This early component was not datable by now been widely recognized on the coast. means of radiocarbon analysis but pro­ Its significance has been the cause of bably predates 5000 B.P. considerable speculation. In the author's opinion, until good diachronic In all, the Kimsquit work presents a studies of shell midden chemistry are picture of 3 successively occupied done it will remain impossible to deter­ sites. Occupation at these appears to mine whether the absence of shell in have begun only immediately prior to these early levels is of cultural- historic times and spans the time from environmental significance or is due the introduction of historic trade goods simply to factors of preservation. to the full integration of cash economy and store-purchased items. Evidence for The 1972 field season also saw the the prehistoric occupations so well first season of an S.F.U. Project at represented at the Kwatna sites is yet McNaughton Island (E1Tb 10) about 30 km to be found. Their absence may relate south of Bella Bella (Pomeroy 1980). to an as yet undated geologic event that This small but deep shell midden site is resulted in a major rerouting of the on one of hundreds of small islands and lower 3 km of the Dean River. Such an rocks on the outermost part of the Cen­ event could have destroyed most sites tral Coast facing the . It older than A.D. 1700 or could have so remains the only excavated site in this disrupted the salmon spawning potential unique Central Coast micro-environment. of the river as to have rendered it A second season at McNaughton was con­ unattractive for human settlement for an ducted by Carlson as part of the 1974 extended period. S.F.U. field school (Carlson 1 976). Pomeroy interprets the McNaughton Island The 1971 field season saw several Site as representing a winter and spring S.F.U. field projects on the coast and village that may at times have been used INTRODUCTION 9 as a regular winter settlement. His A participant in the Kwatna area midden chronology begins with a basal intertidal lithic site surveys, Brian radiocarbon age of 2520 + 90 B.P. from a Apland, also served on the 1973 survey sample some 4 m below the present in the Seymour Inlet-Quatsino Sound area surface. An upper radiocarbon age is (Carlson and Hobler 1976). This project 900 + 80 B.P. (Pomeroy 1980). Carlson's was a reconnaissance survey of the ex­ 1974 field school at McNaughton Island tensive territory around the north end expanded the excavation into an adjacent of Vancouver Island and in the Seymour area of the site revealing deposits that Inlet complex on the mainland opposite seemed to complete the temporal sequence the north end of Vancouver Island. The through to the appearance of manufac­ several intertidal lithic sites recorded tured historic trade goods (Carlson during this field season were character­ 1976). The McNaughton work is signifi­ ized by an assemblage rather unlike that cant in that it documents the presence found in the Kwatna area intertidal on the outer part of the Central Coast sites. In order to fill out the Bella of cultural components resembling those Bella area sample of intertidal lithic of the Anutcix and Kwatna Phases on the sites Apland and Pomeroy devoted the Kwatna River. Much better represented 1974 field season to a specialized sur­ at McNaughton than on the Kwatna River vey aimed solely at the discovery of are cultural materials dating between such sites. Working during low tides about 1500 B.P. and 2500 B.P. from a base camp on McNaughton Island A flaked stone component at McNaught­ they added measurably to the site sample on Island predating the various shell and to the range of materials in the midden occupations was first observed by collections. This material combined Pomeroy who noticed andesite flakes and with Pomeroy's previously recorded a fragment of a leaf-shaped point in the sites, the Kwatna area materials, and intertidal zone in front of the main the sites from the north end of midden deposits. My 1968 survey had Vancouver Island constituted the subject included a few pieces of flaked andesite matter of Apland's thesis which is in­ from the upper intertidal zone at 3 cluded in this volume. Apland's analy­ sites on the inner portions of the Cen­ sis constitutes the only clear tral Coast but it was not known at that description and comparisons of Central time how atypical of the later to be Coast intertidal lithic assemblages. excavated shell middens such materials Because the sites are surface sites they were. In 1969 and 1970 Hester had shown lack the much needed associations and that a flaked stone component at Namu directly datably organic material. The underlay the shell midden deposits and question of redistribution also remained that it was surprisingly early (Hester as did the enigmatic problem of how the and Nelson 1978). His earliest radio­ sites related to hypothesized sea level carbon age is 9140 + 200 B.P. In 1970 variations. Carlson observed flaked stone within the intertidal zone at Cathedral Point 16 km By 1 974 Central Coast work had from the Kwatna River. By then it had achieved several goals. S.F.U. spon­ become clear that such materials were sored work at shell middens at Kwatna, not typical of shell midden sites. Over Kimsquit, Port Hardy and McNaughton the next 2 field seasons several more Island together with Drucker's early intertidal lithic sites were found in work (Drucker 1 943), Simonsen's Grant the Kwatna Bay and Kwatna Inlet areas. Anchorage excavation (Simonsen 1973), Some of these produced large surface and particularly Hester's Namu and Kisa- collections but in no case was it poss­ meet excavations had provided material ible to trace the materials directly to cultural data covering a period of some a source on the shore above the inter­ 5000 years. A pre-shell midden occupa­ tidal zone. tion at Namu adding another 4000 years 10 CENTRAL COAST ARCHAEOLOGY to the sequence had been demonstrated redistribution of intertidal lithic ar­ although not fully defined by Hester and tifacts were the focus of the 1977 work. his associates. An intertidal lithic Most of us had been thinking of the site phenomenon was being explored and intertidal lithic sites as drowned characterized by Apland. With all of camps, eg. relatively intact cultural this ground work underway a host of new deposits dating from a time of slightly questions could be asked. After a per­ lower sea levels. Thus, if the time of iod of 2 years during which no full the lower sea level could be known, the scale projects were initiated on the approximate age of the sites could be Central Coast a new phase of inquiry estimated. But there was an alternative began. One of the primary goals became explanation. The artifacts could have the understanding of the pre-shell mid­ gotten into the intertidal zone through den human population, their origins, the erosion of shoreward cultural de­ subsistence strategies, and posits and the subsequent downslope technologies. To this end Carlson in movement of the materials on to the 1977 began a 2 year S.F.U. field school beaches. If this were so the age of the excavation at Namu, taking up where artifacts need not necessarily be as­ circumstances of time and funding had cribed to a time of lower sea level. forced Hester to leave off. Hester's With these questions in mind 5 sites main trench was cleared and expanded. were subjected to intensive surface Numerous radiocarbon samples and an scrutiny. The provenience of all inter­ expanded cultural sample were obtained. tidal specimens were recorded in precise 3 dimensions. The results of the sub­ One face of the main trench was pre­ sequent analysis show little statistical pared for the removal of a full strati­ evidence for sorting of artifacts by graphic profile. The profile sections natural redistributive processes. are now re-assembled as part of the Namu Downslope movement was in evidence sta­ exhibit in the S.F.U. Museum of Arch­ tistically only on 1 site with a steeply aeology and Ethnology. In 1978 the sloping beach. The remaining intertidal locus of the Namu excavations moved back lithic distributions appear to be in up the hill some 30 m. Sterile in this place and relatively undisturbed. area was reached at 3.95 m. These deep While recording the intertidal lithic excavations produced a huge sample of distribution at Joashila (FaSu 19) some flaked stone and many radiocarbon sam­ 10 km from the Kwatna River, a small ples. The basal cultural unit now has shell midden was found on the shore an age of 9720 + 140 B.P. From this immediately adjacent to the beach arti­ basal unit upward the Namu excavations facts. This 1977 find encouraged the are now the most thoroughly dated by expectation that we might be able to radiocarbon analyses of any British relate the site’s well represented inter­ Columbia site (Carlson 1979). tidal cultural deposit to an intact deposit on the shore and thus be able to Pursuing a different approach in 1977 explore its age and associations. With I undertook a series of specialized this in mind I devoted the 1978 and 1980 studies of Central Coast intertidal field school projects to excavations at lithic sites. Participants in this work Joashila. The work revealed a large were S.F.U. field school students who lithic deposit partly underlying a were rotated at regular intervals with small Anutcix Phase shell midden. The those learning excavation techniques at early cultural materials had been left Namu. This diversification of field on the surface of a broad talus slope training experience has been a feature that may have served as a raw material of several of our Central Coast field source. A radiocarbon age on the upper schools. The problem of datable assoc- portion of the lithic deposit is ia tions and the question of possible 5340 +_ 100 B.P. The much deeper earlier INTRODUCTION 11 portions of the site are as yet undated. an effort to relate site distributions to known resource distributions (Hobler In retrospect, there has been an 1981). evolution in research orientation on the Graduate theses such as those of Central Coast over the last 14 years. Chapman, Apland, and Pomeroy are signi­ Initial survey work identified sites and ficant contributions to the literature site clusters. The first seasons were as is Hester's 1978 monograph and Simon- explorative and aimed at defining the sen's 1978 paper. Journal articles and outline of local culture history. Visi­ papers delivered at meetings have kept ble sites such as shell middens received professional audiences apprised of cur­ most of our attention. The ethnographic rent work but there remains a backlog of literature along with our local native full descriptive reports. informants immensely enriched and guided the work. At the same time they impart­ Future field work will probably see a ed a distinct bias in the direction of move to less strictly culture-historical later sites. Much recent work is still research. Efforts to find and study in a culture history mode, as probably earlier sites are still needed and may it should be, but emphasis has shifted be specifically directed to unorthodox to the less well understood early localities such as high elevations and period. New directions include special perhaps even the sub-tidal zone. A studies such as the work on downslope positive new development is the interest movement described above. In his and desire on the part of native people doctoral work Pomeroy interwove exten­ to be directly involved in archaeologi­ sive survey data with ethnographic and cal studies. The future may see fully historical information in a reconstruc­ cooperative projects especially at late tion of early historic band territories sites near reserves where the educative of the Bella Bella. I am currently value for young people of direct histor­ analyzing Central Coast survey data in ical interpretation can be realized.