UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Is home where the hearth is? : evidence for an early non-domestic structure on the Dundas Islands of… Ruggles, Angela Joy 2007

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

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

Item Metadata

Download

Media
831-ubc_2007-0576.pdf [ 6.58MB ]
Metadata
JSON: 831-1.0101071.json
JSON-LD: 831-1.0101071-ld.json
RDF/XML (Pretty): 831-1.0101071-rdf.xml
RDF/JSON: 831-1.0101071-rdf.json
Turtle: 831-1.0101071-turtle.txt
N-Triples: 831-1.0101071-rdf-ntriples.txt
Original Record: 831-1.0101071-source.json
Full Text
831-1.0101071-fulltext.txt
Citation
831-1.0101071.ris

Full Text

IS H O M E W H E R E T H E H E A R T H IS?: E V I D E N C E F O R A N E A R L Y N O N - D O M E S T I C S T R U C T U R E O N T H E DUNDAS ISLANDS OF N O R T H C O A S T A L BRITISH C O L U M B I A by A N G E L A J O Y R U G G L E S B A , Arizona State University, 2005 A THESIS S U B M I T T E D IN P A R T I A L F U L F I L L M E N T OF T H E R E Q U I R E M E N T S F O R T H E D E G R E E OF M A S T E R OF A R T S in T H E F A C U L T Y OF G R A D U A T E STUDIES (Anthropology) T H E U N I V E R S I T Y OF BRITISH C O L U M B I A August 2007 © Angela Joy Ruggles, 2007 Abstract Northwest Coast cultures are commonly designated as "complex hunter-gatherers" (Matson and Coupland 1995:2; Ames and Maschner 1999:13). This means that although food procurement was from foraging, societies had socially hierarchical characteristics generally associated with late-stage sedentary agricultural societies. Tracing the development of social complexity in Northwest Coast prehistory has been a major endeavour in archaeological research but it has often been a progressive theoretical framework, inferring from material culture that the more 'complex' larger societies analogous with the ethnographic cultures on the N W Coast developed out of simpler, small-scale societies through the accumulation of traits like ceremony and feasting. In this thesis, I argue that the earliest periods o f village life on the N W Coast did not lack these traits but were characterized by another type of ceremonial and ritual organization, one where communities shared control over feasts with somewhat equal access, at least as represented by the structures where these activities took place. I use data from the Dundas Island region on the north Coast of British Columbia to show that the recently-discovered early village at site T512-1, which dates to 2025 cal B C ( IS = 2045 B C - 1945 B C ) includes a non-domestic structure (HP 1) not found in later Tsimshian style villages. To test this hypothesis, I compare T512-1 to another recently discovered village, T512-3, to the range of villages in the Tsimshian area and the suite of early houses known from the N W Coast. I have examined data on several scales of analysis from the village layout, the building architecture, building contents, and finally through a study of the central hearths and their contents in both structures. I believe that these data show that 1) H P 1 is non-domestic when compared to H P 20, the specific analogy o f other Tsimshian houses, and general analogies of domestic behaviour, and 2) that the essence of this difference appears to be specific ceremonial and ritual behaviours that, in later times, are associated with the large houses of village chiefs. n Table of Contents Abstract ii Table of Contents iii List of Tables v List of Figures vi Acknowledgements vii INTRODUCTION 1 C H A P T E R 1: N O R T H W E S T C O A S T A R C H A E O L O G Y 4 NORTHWEST COAST HOUSEHOLD HISTORY 5 TSIMSHIAN ARCHAEOLOGY 8 C H A P T E R 2: H O M E A N D H E A R T H 11 HOUSES AND HOUSEHOLDS AND VILLAGES 11 HEARTHS 12 C H A P T E R 3: D U N D A S ISLAND S E T T L E M E N T A N D SITES O F S T U D Y 15 SITES OF STUDY FOR THE RESEARCH 18 Physical Geography 19 Site T512-3 Anthropogenic Landscape 20 Radiocarbon Dates for T512-3 22 Site T512-1 Anthropogenic Landscape 23 Radiocarbon Dates for site T512-1 24 C H A P T E R 4: M E T H O D O L O G Y A N D R E S U L T S 25 EXCAVATION AND SAMPLING METHODS 25 EXCAVATION DATA: SITET512-3 26 Stratigraphy 26 Features 28 EXCAVATION DATA: SITE T512-1 30 Stratigraphy 30 Features 32 SAMPLING AND PROCESSING METHODS FOR BULK SAMPLES 33 Flotation and Analysis of Samples 34 Sorting and Identification of Light Fraction 34 Sorting and Identification of Heavy Fraction 34 Sorting and Identification of Ecofacts and Artifacts 34 RESULTS: MATERIAL CULTURE 35 RESULTS: HEARTH BULK SAMPLE DATA 36 Shell 36 Fire-Cracked Rock (FCR) and Sand 37 Charcoal 37 Fauna and Flora Remains 37 SUMMARY OF RESULTS 38 C H A P T E R 5: DISCUSSION 42 DAILY LIFE OF THE TSIMSHIAN IN ETHNOGRAPHIC SOURCES 42 HP 1 AS A NON-RESIDENTIAL STRUCTURE 44 EARLY SETTLEMENTS IN THEDUNDAS REGION 46 IMPLICATIONS FOR SOCIAL COMPLEXITY 47 C O N C L U S I O N 50 B IBL IOGRAPHY 51 i i i APPENDIX 1: FLOTATION STEPS 54 APPENDIX 2: DATABASE 55 APPENDIX 3: HEARTH SAMPLE RAW DATA 59 APPENDIX 4: HEARTH SAMPLE PROPORTIONS 60 i v List of Tables T A B L E 1. S U M M A R Y OF E V I D E N C E OF E A R L Y H O U S E S O N T H E N O R T H W E S T C O A S T 7 T A B L E 2: E X A M P L E S OF S O C I A L INTERACTION A N D ACTIVITIES A R O U N D T H E H E A R T H 13 T A B L E 3: E X A M P L E S OF H E A R T H C H A R A C T E R I S T I C S A N D FUNCTIONS 14 T A B L E 4 . D U N D A S I S L A N D S P R E L I M I N A R Y A R C H A E O L O G I C A L S E Q U E N C E 17 T A B L E 5: F A U N A L R E M A I N C O M P A R I S O N 3 5 T A B L E 6. A R T I F A C T S 3 6 T A B L E 7: C H I - S Q U A R E T E S T OF H E A R T H S A M P L E E C O F A C T S (IN WEIGHT PER G R A M ) : 3 6 T A B L E 8. H E A R T H B U L K S A M P L E S W E I G H T (G) A N D PROPORTIONS 3 8 T A B L E 9. S U M M A R Y OF A L L D A T A FROM T 5 1 2 - 1 A N D T 5 1 2 - 3 41 T A B L E 10. D A I L Y L I F E OF TSIMSHIAN FROM E T H N O G R A P H I C SOURCES 43 T A B L E 11. N O N - DOMESTIC STRUCTURE M A T E R I A L EXPECTATIONS O N T H E N O R T H W E S T C O A S T 4 4 v List of Figures F I G U R E 1. N W C O A S T C U L T U R E A R E A 4 F I G U R E 2. D U N D A S ISLANDS A R C H A E O L O G I C A L SITES 16 F I G U R E 3. M A P SHOWING SITES IN THEIR L O C A L C O N T E X T 19 F I G U R E 4. SITE M A P OF SITE T 5 1 2 - 3 21 F I G U R E 5. SITE M A P OF T 5 1 2 - 1 23 F I G U R E 6. N O R T H A N D S O U T H P R O F I L E M A P S OF H P 2 0 AT SITE T 5 1 2 - 3 27 F I G U R E 7. W E S T A N D E A S T W A L L PROFILE M A P S OF H P 2 0 A T SITE T 5 1 2 - 3 28 F I G U R E 8. P L A N M A P OF H P 2 0 AT T 5 1 2 - 3 " 29 F I G U R E 9. S O U T H A N D N O R T H W A L L PROFILE M A P OF H P 1 A T S I T E T 5 1 2 - 1 31 F I G U R E 10. E A S T A N D W E S T W A L L P R O F I L E M A P S OF H P 1 A T S I T E T 5 1 2 - 1 32 F I G U R E 11. P L A N M A P OF H P 1 AT SITE T 5 1 2 - 1 33 F I G U R E 12. S I T E M A P O F T 5 2 2 - 1 47 v i Acknowledgments There are many who have offered their guidance and support throughout this process. First, a great many thanks to my supervisor, Dr. Andrew Martindale who took me on as his first student, you have been a fantastic and patient mentor, thank you so much for allowing me the incredible experience of working on the Dundas Island Project. I must also express my gratitude to my committee, Dr. Felice Wyndham for sticking by the project even as its focus changed; I have enjoyed and learned a great deal from our conversations. A n d thank you to Dr. Brian Chisholm who took time out o f his busy schedule as a third reader to edit and give his comments. I wish to acknowledge David Archer, co-researcher in the Dundas Island Project, your expertise on north Coast prehistory has been of great value to this research. M y love and appreciation goes to my parents who have encouraged me in my dreams no matter how far from home they take me. I need to thank Kisha who has been a wonderful sister, friend, roommate, officemate and partner in crime; I cannot imagine how I would have managed without you. I am eternally grateful to my kindred spirits, Lauryn, Nadia, Cathy, and Neele without you all I never would have checked back into reality. Thanks to Marina and Craig, for always being just around corner when I needed to get out for a pint and to R ich for providing the musical release after dinner. I must acknowledge, M i k e Olivotto who helped with data processing, it was fun having someone to make mud pies with, Vie len Dank! I extend my gratitude to the 2006 field school and crew who helped with the excavations and made Dundas unforgettable and also to the 2007 field crew, our adventures and laughter were memorable. I appreciate the opportunity that I have had to work on the traditional territory of the Tsimshian, it has been a great honor to be able to explore the history o f a people who know and respect their ancestors. Lastly, I thank the funding agencies, the Social Sciences and Humanities Research Council of Canada and U B C ' s University Graduate Fellowship for supporting this research. V l l I n t r o d u c t i o n One of the key theoretical issues in the history of Northwest Coast archaeology has been the development o f so-called social complexity. In order to understand social complexity in the region, Matson and Coupland (1995:5) use the concept of "The Developed Northwest Coast Pattern" to structure the known data; however; this concept has progressive implications that assume later, more complex communities form through the gradual accumulation of new traits out of simpler, smaller-scale communities. This argument is not necessarily wrong, but it does not provide a robust definition o f "complexity." In later work, Martindale (1999) and Coupland et al (2001) argue that complexity is the institutionalization of inequality in cultural structures that provide resources o f power to elites creating organizational hierarchies of economic, political, and social relations. However, as Coupland (2006) acknowledges, the institutionalization of inequality is an interpretation out o f material traits. Thus, the tendency is either to read the N W Coast cultural sequence as a progression from simple to complex or to assign similar traits different significance in earlier or later times. For example, Cannon (1998) argues that the same evidence exists for ceremonial activities like feasting for over 5000 years ago on the central coast, but it is only in the later period when archaeologists argue that it becomes controlled by elites. In this thesis, I consider these implications of the evolution of social complexity and present data that supports both Cannon's (1998) assertion that feasting in earlier communities was not absent and the more widely held view that later feasting was controlled by elite households. In short, I argue that the earliest periods o f village life on the N W Coast were characterized by another type of ceremonial and ritual organization, one where communities shared control over feasts with somewhat equal access, as represented by the structures where these activities took place. To support my thesis, I use data from the Dundas Island region on the north Coast of British Columbia to show that the recently-discovered village site o f T512-1 (temporary site number), which dates to 3650 ± 50 B P or 2025 cal B C ( I E = 2045 B C - 1945 B C ) includes a non-domestic structure (HP 1) not found in later Tsimshian style villages. To test this hypothesis, I compare T512-1 to another recently discovered village, T512-3 (temporary site number), to the range o f villages in the Tsimshian area and the suite of early houses known from the N W Coast. M y logic is based on a comparison of 1 villages and houses in Tsimshian territory at these two sites and a focused comparison between what is arguably a later, typical Tsimshian house (HP 20) and the non-domestic structure (HP 1). To do this, I have examined data on several scales of analysis from the village layout, the building architecture, building contents, and finally through a study o f the central hearths and their contents in both structures. I believe that these data show that 1) H P 1 is non-domestic when compared to H P 20, the specific analogy of other Tsimshian houses, and general analogies o f domestic behaviour, and 2) that the essence of this difference appears to be specific ceremonial and ritual behaviours that, in later times, are associated with the large houses o f village chiefs. M y thesis is structured into 5 chapters in order to give the necessary background for my argument as well as present the original data from this research. In Chapter 1,1 review the Northwest Coast household archaeological data with a focus on the first structures and villages found, as the early dates at T512-1 make H P 1 now one of earliest known structures. Fol lowing this, I present a brief discussion of what is known about Tsimshian prehistory and settlements in order to situate the Dundas Island region in a local context. In Chapter 2,1 explain the assumptions o f my logic by discussing how the material record o f daily life is created and can be used as a control for defining domestic space. Next, I define houses and households on the Northwest Coast before focusing the discussion on hearths and their material manifestations o f daily life activities as they are constructed, used, and maintained. Chapter 3 is a description o f the Dundas Island regional information and the sites, T512-1 and T512-3, used in this research. The purpose of this chapter is to set the context and report on site level data, information that demonstrates the unique characteristics of site T512-1. In chapter 4,1 report on the results of the excavations and processed bulk hearth samples to show that the sites are different on almost every level of comparison from the central hearths and their contents, to building contents, building architecture and the village layout. Lastly, in Chapter 5,1 discuss the implications of these data, by first providing further evidence that the H P 20 at T512-3 is indeed a house, and that H P 1 at T512-1 does not match either T512-3 or the general behaviour patterns of a domestic space. Next, I discuss ethnographic examples o f non-domestic 2 structures and conclude that H P 1 at site T512-1 is an example o f a structure not represented in ethnographic record based on its early dates. I then place T512-1 into the larger context of the Dundas region and show that the site is part of a larger landscape of early sites and that it is not unexpected to find such a structure during this time in prehistory. Final ly, I place the data into the larger theoretical endeavour o f unpacking cultural complexity and suggest that T512-1 is evidence that changes in Northwest Coast political organization were internally motivated and not simply adaptations to demographic or environmental stresses. 3 C h a p t e r 1: N o r t h w e s t C o a s t A r c h a e o l o g y The Northwest Coast is a geographically and culturally distinct area extending from northern California to Prince Wi l l i am Sound off the northern end o f the Alaskan panhandle (Figure 1) (Ames and Maschner 1999: 17; Matson and Coupland 1995: 1-2). This region is divided into the south, central, and north coasts. M y research focuses on the northern coast, specifically the Dundas Islands area of Tsimshian territory, an offshore island cluster located between the Queen Charlotte Islands and the Skeena River. Pacific Ocean O to ISLAND X Strait of A \M ) \ Georgia- ^ - ' ^ E . \ J Strait of / (y^m ~" * ^Juan de Fuca, f V Coos Bay / Cape r~\ — Coquiile Rwer Medocino/*-.,^^-, Rogue Rive; f, ;i J 0 100 200 Km 2 0 0 mi ^7 1 0 0 i Figure 1. NW Coast Culture Area (Source Matson 2003: 4) 4 Northwest Coast cultures are commonly designated as "complex hunter-gatherers" (Matson and Coupland 1995:2; Ames and Maschner 1999:13). This means that although food procurement was from foraging, societies had socially hierarchical characteristics generally associated with later-stage sedentary agricultural societies. Characteristics of social complexity include but are not limited to, permanent or semi-permanent settlement, food storage, specialization, household based economies and hierarchical social and political organization (Ames and Maschner 1999:13). In this chapter, I outline the evidence o f earliest houses and villages on the whole Northwest Coast in order to contextualize the significance of settlement and architectural data from the Dundas region, which w i l l be discussed later in Chapter 3. Lastly, I focus my discussion on the northern Coast and examine briefly what is known about Tsimshian settlements and the first evidence of inequality. Northwest Coast Household History The prehistory of the mid- to late Holocene on the northern Northwest Coast and the Tsimshian area is commonly divided into three Periods: Early (5000-3500 BP) , Middle (3500-2000 BP) and Late (2000-200 B P ) (Ames and Maschner 1999:66). Early and Midd le Period household data are limited, as plank houses are present but rare between 1450 and 800 B C (Ames and Maschner 1999: 160) however; this is partly a logistical problem of the excavation of lower components from shell middens as much as the lack of structures in the early and middle periods. It is assumed that accumulating shell middens existed throughout the region as early as 5000 B P ; coupled with evidence of hearth and postholes from the rare excavated components suggests that domesticity had great antiquity even though few definitive occupational surfaces have been identified. The Early Period data show distinct cultures throughout the whole coast and are interpreted as representing small, egalitarian, mobile populations o f people. During this period, pithouses became more common in the Plateau area and it is suggested that the appearance of large shell middens on the coast represents a similar pattern of settlement (Ames and Maschner 1999: 156). The oldest known structures are found at the Hatzic Rock site and the Maurer site, both in the Fraser Val ley. Both sites date to around 4500 B P (Ames and Maschner 1999: 156; Matson and Coupland 1995: 117, 132). The earliest structure on the coast was found at the Hidden Falls site in Southeastern Alaska dating to 3800 B P ; this appears to 5 be a small temporary structure indicated by a hearth and ring of postholes (Ames and Maschner 1999: 156). Middle Period societies are defined as semi-sedentary egalitarian people using a delayed return economy (Matson and Coupland 1995: 197-198). In this period a major change in house form and community organization occurs. Villages are laid out in formal rows commonly facing a shoreline of a beach or river (Ames and Maschner 1999: 158-159). The first example of this settlement pattern is found at the Paul Mason site located in the Prince Rupert Harbour area (Coupland 1988; Matson and Coupland 1995: 184) This village dates to 3200-2700 B P (1450 and 950 B C ) and is comprised of two rows of small rectangular houses with prepared house floors. In addition, there are two structures oriented in a different direction above the back row; these could be a later addition or special function structures. Because the houses are relatively equal in size and excavations revealed little variation in internal features the Paul Mason site is inferred as representing egalitarian political and social organization (Matson and Coupland 1995: 187-188). A similar village arrangement is suspected at the Boardwalk site during this period with two and possibly three rows of houses, although the architectural data from this site were not well recorded. There are only two well-preserved structures at the rear of the second row, both of which are smaller than the Paul Mason structures (Ames and Maschner 1999: 159). In other areas on the coast during the Middle Period, only single structures have been identified. At the Palmrose site on the south coast there is a very large structure measuring 20 m by 6 m. The Crescent Beach site in the G u l f of Georgia has a small pithouse dating to this period (Matson and Coupland 1995: 161). During the Midd le Period, there is little evidence for ascribed status, winter villages, or large multi-family houses (Matson and Coupland 1995: 193). In the Late Period, social complexity is evident throughout the Northwest Coast although regional variability does still exist. The G u l f o f Georgia region offers the earliest examples o f large winter villages at sites associated with the Locarno Beach phase, although these sites do not appear to have the "ranked" village pattern of large houses located in the front-centre o f the village. At Prince Rupert Harbour, evidence for "ranked" villages is found throughout the harbour after 1800 B P (Archer 2001) at sites such 6 Table 1. Summary of Evidence of Early Houses on the Northwest Coast (First structures, "egalitarian" villages and "ranked" villages in bold) Date (BP) Site Location Description Notes 5000-3500 Hidden Falls Southeast Alaska 37 postholes, 3x4 meter depression Small semi-permanent, Earliest known structure in a coastal setting (4600-3200 BP) Boardwalk Prince Rupert Harbour Small postholes average of 15 cm No depressions or structures confirmed Hatzic Rock Lower Fraser Valley 1 structure 40 sq. meters, outline of postholes Form unknown in ethnographic times (4500-3300 BP) Maurer Lower Fraser Valley Large rectangular structure Due to its architecture appears to be more recent than associated tool assemblage (4500-3300 BP) 3500-2500 Paul Mason Kitselas Canyon 2 row linear village, 10 houses, average 10x5 meters plus 2 First Clear evidence of settled life, "egalitarian" village. (3200-2700 BP) Crescent Beach Gulf of Georgia Small "pithouse" Long Harbour and Shoemaker Bay I Gulf of Georgia Postholes Indicative of large structures. 2500-1400 Hidden Falls Southeast Alaska Numerous hearth features and large postmolds No clear house floor Gitaus Kitselas Canyon Postmold Result of dwelling Boardwalk Prince Rupert Harbour 2 House floors, (9 x 5 m a n d 9 X 6 m ) Undated but are associated with this period. McNichol Creek Prince Rupert Harbour Two-row linear village with 15 depressions "Ranked" village. "House O" front-centre chief's house (1800-1500 BP) Grant Anchorage Hecate-Strait Milbanke Sound 2 charred wood boards, 1 badly decayed board Structural remains from a house, size and shape unknown Beach Grove Gulf of Georgia 10 house platforms minimum house size of 10 x 13 m diameter Good evidence for large multifamily households and large villages of planked houses. Garrison and False Narrows Gulf of Georgia Large house platforms Whalen Farm and Dionisio Point Gulf of Georgia House platforms May be associated with later time period. Palmrose South Coast Rectangular-subterranean house, 6 x 12-16 meters Some confusion on dates, ranges from 2500-2000 BP (Sources: Matson and Coupland 1995; Ames and Maschner 1999) 7 as M c N i c h o l Creek (Coupland et al. 2003). These villages are thought to represent a hierarchical organization under village chiefs, in part because they contain large houses, homes and feasting halls of such leaders in the centre-front row. A t Boardwalk, there is no direct evidence for large winter villages; however, there is evidence for warfare and social inequality in the burials, which suggests, a ranked village settlement during this time (Matson and Coupland 1995: 241-242). In summary, Early and Midd le Period household data on the Northwest Coast are only a small sample o f what was probably a more ubiquitous settlement pattern up and down the coast (Table 1). Other evidence such as burials, shell middens, artifacts and ecofacts play a role in understanding the scope of societies from these early times. A s mentioned previously, it is not unexpected to find structures and villages earlier than 4000 B P ; however, it has been difficult because these early occupational surfaces are buried in meters o f shell. The Dundas region pushes back many of these dates for both early structures and early villages and, as I w i l l discuss further in the following chapters, enhances our understanding o f how these earlier "egalitarian" villages were negotiating and organizing power. Tsimshian Archaeology On the northern coast, specifically Tsimshian territory, the local archaeological sequence is divided in three periods; Period III (Early Period 5000-3500 BP) Period II (Middle Period 3500-1500 B P ) and Period I (Late Period 1500 B P - 1834 A D ) (MacDonald 1969; MacDonald and Inglis 1981). The majority of our information comes from a few sites located on the coast in Prince Rupert Harbour and in the interior near the Skeena River. In this section, I briefly summarize the archaeological data from the north Coast Period III and Period II with a general focus on Prince Rupert Harbour in order to situate the Dundas region in a local context. Information about Period III (5000-3500 BP) comes from Boardwalk (GbTo-31), Lachane (GbTo-33) and Dodge Island (GbTo-18). Other sites like Kitandach (GbTo-34) and Ridley Island (GbTn-19) have Period III dates but have very few associated artifacts. A s mentioned previously, there are no house depressions recorded at any of the sites though the Boardwalk site has small postholes and hearths. Woodworking assemblages have small bone and antler wedges but have no adze or hand mauls found in later periods, which would indicate the construction of plank houses. Shell middens in this period are thin 8 and diffuse. In sum, these data are interpreted as-evidence of a small, mobile population around the area of Prince Rupert Harbour (Ames and Maschner 1999: 96-97, Matson and Coupland 1995: 125-141). Period II has an earlier (3500-2500 B P ) and a later (2500-1500 B P ) component. In the earlier part of Period II, the Paul Mason site (GdTc-16) in the interior at Kitselas Canyon is the earliest example of a large rectangular-house village. In addition, there was a more rapid accumulation o f shell middens occurring at a number of sites in the Prince Rupert Harbour area. Evidence for social inequality is not clear in the archaeological record at this time. Burials and village organization at the Paul Mason site are interpreted as representing an egalitarian social system (Coupland 1988). Despite this, the data suggest through the increasing number of sites that the population in the Prince Rupert Harbour area was growing (Ames and Maschner 1999: 97-99, Matson and Coupland 1995: 183-196). Similarly, we know little about the residential patterns at Prince Rupert Harbour during the second half of Period II (2500-1500 BP) . For the most part house features have not been excavated but winter villages have been inferred from the large, deep shell middens and house depressions at some sites. Two house floors at Boardwalk might date to this period and, as mentioned above, these are similar in size to the earlier Paul Mason structures. A t the M c N i c h o l Creek site, a 15-house linear "ranked" village style is apparent. This is because the largest structure in the village, House O, has characteristics that communicate, wealth, rank and power and is interpreted as a chiefs house (Coupland 2006: 86). Period II has some of the earliest evidence for social stratification from burials found at the Boardwalk site. The inclusion of grave goods such as exotic items like copper and personal adornment like labrets are interpreted to mean that the individual buried had a relatively high status. There is also evidence of increased warfare both in the appearance of bone clubs and the skeletal remains with pathology associated with violence like parry fractures. It is inferred that these data represent the beginnings of social stratification as the grave goods may indicate gestures of social differentiation and conflict (Ames and Maschner 1999: 97-99, Coupland et. al 2001, Matson and Coupland 1995: 183-196). In summary, data from the early periods of north Coast prehistory show a switch from "egalitarian" political organization to more institutionalized inequality, beginning late in Period II. This change is manifest most in stratified burials but also in the changing settlement patterns from the 9 "egalitarian" style at Paul Mason to the "ranked" style at M c N i c h o l Creek. Archer (2001) has argued that the transition occurs throughout Prince Rupert Harbour such that all egalitarian villages were abandoned about 1900BP and that all ranked villages appear after 1800 B P . It is uncertain what caused the change in political organization, although Martindale and Marsden (2003) suggest it is the result of a regional war. Whatever the cause, I argue in this paper that the change involved the movement of ceremonial activities and feasting from community-controlled structures into the houses o f chiefs. Thus, the nature of complexity as institutionalized inequality includes the elite control o f ceremonial and feasting activities. Significantly, my data suggest that such activities were not absent from the earlier period; therefore, it was the control of such activities and not their invention that characterizes the later, more complex communities. 10 C h a p t e r 2: H o m e a n d H e a r t h Dai ly practice in people's lives structures space and can define how domestic activities are conducted and how refuse is deposited (Lightfoot et al 1998). These daily routines can produce a material record o f the aggregate o f repeated behaviour, even though cleaning and maintenance may erase evidence o f individual days (Arnold 2006: 278). A s people structure their daily lives, they are inevitably reproducing underlying expectations and beliefs that hold influence over the way they maintain, use, and organize household space, and thus the social relations that define the nature o f their households (Ortner 1989, Bourdieu 1977). Because of this, we can observe patterns in the archaeological record and infer cultural meaning and social identity, especially with the use of specific ethnographic analogies and more general behavioural analogies. In this chapter, I discuss households and their importance in archaeological research before defining what constitutes a household as a social phenomenon and a house as its spatial manifestation on the Northwest Coast. Next, I focus the discussion on the hearth as a part of residential space and consider the activities and social interaction that could take place around a domestic hearth to argue that hearths are a defining characteristic o f N W Coast houses. Lastly, I mention that the construction and maintenance of a hearth is a result of a reproduction of knowledge and that different types of hearths may be reserved for specific functions. Thus, a hearth found in a residential context is expected to have a different material pattern than a hearth used for another purpose in another context Houses and Households and Villages Organization of households has been a central research issue in archaeology because it links the everyday lives of people to the larger cultural context of a society (Ames 2006: 17). The household was a primary site for observing, learning and creating cultural identity though the reproduction of daily life routines. Houses are a place where archaeologists can witness the material manifestation of daily life and use these patterns to evaluate the social, economic and political character of a society (Martindale 2006: 140). On the Northwest Coast, there is a distinction between houses and households. Houses are the structures that people live within. On the Northwest Coast, there are four types of houses, which include pit houses, plank houses, mat lodges and longhouses (Ames and Maschner 1999: 151). Most relevant to 11 this research are plank houses, which are large post and beam structures, clad and roofed. In the ethnographic record, these were most commonly constructed from Western red cedar (Thuja plicata) (Boas 1916, Gahr, Sobel and Ames 2006: 5). This research focuses on the archaeological remains o f houses, which are generally U-shaped surface depressions found on anthropogenic shell ridges; however, house depressions are not limited to this shape or location. In addition, we cannot limit our understanding of these depressions to classify them only as houses, as structures with the same or different surface contour could have had other purposes outside of being a domestic space. Households on the Northwest Coast are defined as a basic social and economic unit or corporate groups o f multiple families who lived together or in close proximity to one another (Hayden and Cannon 1982; Ames and Maschner 1999: 147). It is difficult to assess the relationship between houses as the structure and households as the social and political unit in the archaeological record (Gahr, Sobel and Ames 2006: 6). Nevertheless, it is certain that houses are usually part of a larger village landscape and it has been inferred that the spatial arrangement of houses within the village, especially a "ranked" village, reflects the social and political relationships of different households. Sometimes a house within a village w i l l have dual functions as a domestic space and a ceremonial and feasting structure. A n example o f this is House O at M c N i c h o l Creek, which was a chiefs domestic space but also served as a feasting hall, due to the large feasting hearth located within the structure (Coupland et al. 2006). On the other hand, there might also be special function structures, perhaps similar to the two outlying structures at Paul Mason, which might not serve a domestic function but rather a ceremonial, storage, or other purpose. Indeed, both o f these examples differ from other houses mainly in size, orientation and location within the village; however, surface appearance presents insufficient data to discover the function of such structures. A s I argue in the thesis, one valuable way of finding out the identity of a structure is through a focus on hearths, which can yield a great deal of information about the function and frequency o f the space around it. Hearths Hearths are an important part of the household and were located in the centre o f a house on the northern Northwest Coast in ethnographic history (Boas 1916: 48). Within the house, hearths were key locations 12 where daily social and economic interactions occurred. Depending on the size and location of a hearth, activities might have revolved around food consumption, feasting, processing and manufacturing of food and technologies among the nuclear family, extended family or other members o f the community (Martindale 2006: 145). This case study focuses on hearths, which are particularly interesting because they are complex spaces of interaction. Hearths are used in both domestic and non-domestic contexts and are built for a variety o f purposes. In the domestic context, hearths are often a way to quantify the number of people l iving within a plank house, as well as determining the social and economic relationships o f those people (Coupland 2006: 80, Grier 2006: 97, Martindale and Jurakic 2004). Thus, the material record found in and around a hearth could represent a number of daily life activities, interactions, knowledge and identity at a variety o f social and gender levels (Table 2). Table 2: Examples of Social Interaction and Activities around the Hearth Activity Social Correlate Material Correlate Primary Location Associated Knowledge Cooking Women (Slaves) Charcoal, plant remains, faunal and shell remains, charcoal, FCR At hearth or outside in steaming pits Nutrition, recipes, cooking techniques Processing Feasting Women Everyone Food remains, processing tools (knives, scrapers etc.) food remains in all forms (specialized for potlatch) Outside hearth, at collection site, in village, beach Around hearth Techniques and technology, preservation methods Social rules Weaving/ Sewing Women Bone point and needles, plant remains. In house, around hearth Patterns, techniques, technologies, processing, phenological indicators Tool Making Everyone Debotage, grinding and polishing stones Around hearth, in house, on location Materials, technologies, techniques, resource location Story Telling Everyone None Around hearth, in house, on location Oral traditions, social rules, identity, ownership, clan history. Trade and Gift Giving Everyone (elite, guests, house owners) Exotic items, foods (specialized and everyday) In house, around hearth, on location Social rules (Sources: Turner 2003, Turner and Lantz 2003, Turner and Clifton 2006) Furthermore, the maintenance and construction of a hearth represents a reproduction o f knowledge as these characteristics are either repeated or changed during daily life activities. Different 13 activities w i l l require a different type o f hearth and these w i l l be built and used accordingly (Table 3). Ultimately this means that the material manifestation of daily life around a hearth would be different than a hearth used infrequently or for special functions. In summary, hearths can tell a variety of stories about the people that gathered around them. Archaeology still has much to learn in terms of translating the material record into the social interactions, transfer of knowledge and identity formation that took place in prehistory. However, archaeology can recognize when patterns change, retain continuity or are different over the long term, giving the discipline a unique position with which to investigate the larger patterns o f human behaviour. Table 3: Examples of Hearth Characteristics and Functions Hearth Function Frequency of Use Primary Location Characteristics Expected Material Culture Cooking Everyday House structure Medium sized cleaned regularly, basin shaped Ash, fauna/floral remains, some charcoal, artifacts from daily life. Warmth Everyday or Occasional House, non-domestic structure, or outdoors May not be cleaned after use, larger fire, mound shape or basin shape High amounts of charcoal, some ash Feasting Occasional Chiefs House, Ceremonial Structure Large, located in front centre of building Specialized Faunal/floral remains, specialized artifacts Processing Occasional Outdoors Very large mound shape feature High amounts of charcoal, FCR and remnants of processed fauna or flora. (Sources: Matson and Coupland 1995, Ames and Maschner 1999, Coupland 2006) 14 C h a p t e r 3: D u n d a s I s l a n d S e t t l e m e n t a n d S i t e s o f S t u d y This research was done as a part of the Dundas Island Project, a 2005-2007 SSHRC-funded project led by Dr. Andrew Martindale and David Archer. The Dundas Island Group is located 30 km northwest o f Prince Rupert Harbor in The Dixon Entrance on the northern Coast o f British Columbia. The general goals o f the Dundas project were 1) to track mid-late Holocene settlement patterns through survey o f the modern shoreline, 2) to explore conjunctions between the archaeological and oral records and, 3) to explore early Holocene landscapes at higher (5 and 10 m) elevation terraces that correspond to ancient strandlines dating to 8000-5000 B P and over 10,000 B P , respectively. The Dundas Islands are important archaeologically in part because of their long history of settlement. Dates from the Far West Point site indicate cultural activity in the area as early as 9690 ± 30 uncalibrated B P ( U C I A M S 28008, charcoal). According to oral traditions, the earliest Tsimshian people came from the outer islands, such as the Dundas Group. Prior to this project, the earliest archaeological site in Tsimshian territory was Boardwalk (GbTo-33) with early components dating to about 4800 years ago. This project has located 4 shell middens and 1 village site dating to before 5000 years ago. More recent Tsimshian history (Martindale and Marsden 2003) describes two influxes of northern peoples. This project has located an earlier, Tsimshian-style village dating from 4500 to after 2000 years ago and a second, non-Tsimshian style village dating to between 3000 and 2000 years ago. Over 120 sites have been recorded in the Dundas Group, the majority in the last 3 years (Figure 2). In general the preliminary chronology of the Dundas region follows with the development o f north Coast culture history (Table 4). Early sites are consistent with Prince Rupert Harbour Period III observations at Boardwalk (Matson and Coupland 1999: 126). Sites in the Dundas region have small shell middens and the settlements consist o f small camps or villages. Early period II has characteristics o f small "egalitarian" villages like Paul Mason (Archer 2001; Matson and Coupland 1999: 187) followed by large "ranked" villages like M c N i c h o l Creek in late Period II and Period I (Archer 2001; Matson and Coupland 1999: 279-80). 15 Table 4. Dundas Islands Preliminary Archaeological Sequence Dates Sea Level Sites Details 2000 BP - Modern present 3000 -2000 BP Before 8000 BP Modern 4000 - Modern 3000 BP Cabins. Camps. Small shell middens. Rock shelters. 2 major "ranked" village types. Large "ranked" village type. 4500 - Modern Small "egalitarian" villages. 4000 BP 6500 - +5 m 4500 BP 8000 - +5 m 6500 BP Small house clusters. Shell middens. Campsites. +10 m Shell middens. Large villages abandoned after 2000 BP and replaced by seasonal use camps, and possibly small hamlets. Cabins build in the early contact era still in use today. Large villages dominate settlement landscape, possibly associated with a suite of smaller shell midden and village sites. The existing T1 style of village is found in association with a new T2 style that has the largest houses at the centre-back of a curved row. Large Type 1 (T1) villages date from about 4000 BP. Style is similar to Period 1 "ranked" villages from Prince Rupert Harbour with largest houses in the centre-front of linear house rows. A suite of smaller shell middens and hamlets likely dates from this period. Similar to small villages found in the Prince Rupert Harbour area, these sites have linear rows of small, similar sized houses. Represented by only two sites, small clusters of 3-4 houses on shell ridges seem to be the earliest form of architecture in the area or on the NW Coast. Early Holocene shell middens and hearth sites are known from strandline beaches above the modern shoreline. The earliest shell middens are found on 10 m raised beach terraces, where the earliest components date to before 8000 BP and possibly as old as 11,000 BP. (Source: Martindale 2007) So-called egalitarian villages are thought to represent more equitable social, political, and economic relations because of their pattern of small numbers (N<10) o f equally sized, small houses. Larger, ranked villages are characterized by large numbers (N>20) of houses in which one or two very large houses are located prominently. The Prince Rupert Harbour pattern has these large houses in the centre-front of linear house rows. Coupland (2006) interprets these as the houses o f village chiefs in a hierarchical organization analogous to that recorded ethnographically. The "ranked" village styles on the Dundas Islands have been classified in two types, Type 1 and Type 2. They are characterized by the spatial organization of the site and the shape and location o f the shell midden. The Type 1 style villages date between 4500-1500 B P and represent a Tsimshian style village equivalent to the Prince Rupert Harbour style excavated at Boardwalk (Ames 2006) and M c N i c h o l 17 Creek (Coupland et all 1993; Matson and Coupland: 278-281). Type 1 villages generally have linear arrangement of structures facing the same direction, with larger structures located in the front centre of the site and the shell midden forms a linear ridge behind the structures. The Type 2 style villages date between 3000-2000 B P and represent a village style not previously recorded in Tsimshian territory. This classification is based on the layout o f the village in semi-circular arrangement with the larger structures located in the centre-back, and a shell ridge located behind the village in an arc. It is hypothesized that these Type 2 villages represent the first influx of northern peoples who ally themselves with the Tsimshian (Martindale and Archer, pers. comm. 2007) as recorded in the oral traditions (Marsden and Martindale 2003). Sites of Study for the Research It is against this backdrop of complex culture history that this project was undertaken. The data comes from two sites, T512-3, a Type 1 "ranked" village dating to between about 3000 and 1500 years ago, and T512-1 a small "egalitarian" village dating to before 4000 years ago. Each are located on separate, but nearby islands in the Dundas Group. Following Archer (2001) and Coupland (2006), I argue that the earlier, egalitarian village represents a less hierarchical social, political, and economic organization than the later ranked village. To explore this relationship, I excavated the central hearth of one structure from each site, which I argue represent a house (from T512-3) and a non-residential, special function ceremonial building (from T512-1). M y thesis is that in the earlier egalitarian village, the ceremonial and ritual functions that later become controlled by chiefs (and located within ch iefs houses), are found in a separate, special-purpose community structure. I present the data for this argument in Chapter 4 and explore its implications in Chapter 5. Differences between the islands and sites exist in both physical and anthropogenic landscapes. Major differences in the physical environments include access to fresh water, the geography of the ocean channels used to approach the sites and the natural topography. Differences between the anthropogenic landscapes at the two sites include site size, type, dates, subsistence economies and defensive constructions, which w i l l be discussed in detail below. 18 Physical Geography Site T512-3 is located on the east side of Baron Island, and site T512-1 is located about 2 kilometers southeast o f T512-3 on the east side o f a small islet known from its elevation listed on hydrological charts as Island 53 (Figure 3). Both sites have access to a variety o f marine habitats including the shallow waters of the inner-outer passes through the island chain and the deeper waters of both Chatham Sound and the Pacific Ocean. L ike all areas in the north Pacific, this region is affected by extreme differences between low and high tides with variation at some times more than 6 meters. This creates an expansive intertidal zone that supports a variety o f marine life. Both islands have characteristics associated with the coastal needleleaf forest biome. This environment, also called the Western Hemlock zone, is characterized as wet mossy temperate rainforest (Pojar and MacKinnon 1994). N o large land mammals live on either of the islands, with the exception o f wolves, which Tsimshian elders think were either brought, or swam to the islands in recent times. Figure 3. Map Showing Sites in their Local Context 19 The major differences in the physical geography o f the two sites are the location of fresh water, the geography of the island channels used to approach the site, and the natural topography of the islands. A t site T512-3, a freshwater stream runs into the intertidal zone adjacent to the site on Baron Island. N o fresh water source was located on island 53 however; there is a stream 270 meters away across the channel on Dunira Island. The island channels that provide access to the sites vary greatly. Large basaltic rocks jut out from the ocean floor on the west side of Island 53, where site T512-1 is located. These rocks are hard to see at high tide and are difficult to navigate at all tidal levels. The foreshore is also very rocky, making it difficult to land a boat on the beach. In contrast, the channel approach to site T512-3 is much more accessible. There are only a few rock obstructions in the natural harbor formed by the inlet that do not restrict entry to the site in the same manner; however, at extreme low tide, entrance to the site is limited by the expansive inter-tidal zone. Current paleoenvironmental research has found that the Dundas islands are a line of the coastal tectonic plate (Mclaren et al 2007). This means that while other areas of the Northwest Coast were affected by extreme changes in sea levels, the Dundas Islands only experienced a drop of 15-6 meters during the early Holocene as sea levels stabilized around 5000 B P . Based on this information, sites earlier than 5000 B P are found at slightly higher elevations rather than the modern shoreline, unlike the other areas o f the coast (McLaren 2007 et ah). The earliest dates at T512-1 come from the basal and upper deposits on the shell terrace, placing the construction of the ridge before 5000 years ago, a time when the lower terrace would have been intertidal. Thus, the site contains two different cultural components, an earlier shell midden that formed the back terrace, and a later village that formed the lower terrace. The structure explored in this project was built on the back ridge, but dates to about 4000 years ago, after the earlier back terrace shell midden was completed. Thus, although the structure is built on a 5000 year old shell midden, it is most l ikely associated with the more recent village on the lower terrace. Site T512-3 Anthropogenic Landscape Site T512-3 is located on the modern shoreline on the east side of Baron Island. The site is on large shell midden, covering a total area of 16800 square meters and the rise in elevation from the front o f the site to 20 the back shell ridge is 5 meters. A fish trap is located near the site in the intertidal zone on the fresh water stream. Site T512-3 is a Type 1 village and has two rows of 24 house depressions facing roughly toward the south beach (Figure 4). The largest house depressions are located in the centre front row. Three smaller depressions are located on the northwest side of the site, facing in a different direction than the rest of the villages houses and are hypothesized as either a later addition to the village or special function structures (Archer, pers.comm. 2007). Because o f the congruency o f T512-3's spatial organization as described above to large Period I winter villages like Boardwalk, Phi l l ip ' s Point, Tremayne Bay, and M c N i c h o l Creek (Matson and Coupland 1995: 278), T512-3 can be considered a Dundas Island version of the Prince Rupert Harbour Style. Figure 4. Site Map of Site T512-3 21 A l l research to date has shown that later villages in the Prince Rupert Harbour style are composed of houses of two sizes (Coupland 2006). Larger houses, like House O at M c N i c h o l Creek, located at the near the centre front o f the village, are thought to have had dual functions as both a domestic and ceremonial special function structure, such as a feasting hall (Coupland 2006: 86). Smaller houses, which make up the majority o f the structures at such villages and comprise the entire back rows, are domestic spaces as they constitute non-chiefly households o f the village. Data for this research does not come from the large houses at T512-3 but rather from a smaller structure, H P 20, located in the second row of site T512-3. Because of its size, which is consistent with the majority o f the other depressions, and its location within village back row; H P 20 is arguably a typical house, similar to Houses E , D , and F at M c N i c h o l Creek (Coupland 3006: 87), and can be used as a baseline for understanding domestic Tsimshian architecture and the material culture associated with a domestic context on the Dundas Islands. Therefore, H P 20 represents a control to understand the unusual characteristics of H P 1 at site T512-1, which w i l l be discussed in further detail below. Radiocarbon Dates for T512-3 Radiocarbon dates taken for both sites in this study are based on shell and charcoal. Charcoal dates are calibrated for the variability of atmospheric C-14 over the past 15,000 years. Shell dates are calibrated for both atmospheric variability and the marine reservoir effect. The latter is a correction to account for the effect of deep ocean currents that sequester ancient carbon for up to 1000 years. The effect is that charcoal dates are usually calibrated older than their assayed radiocarbon years, while shell dates are calibrated younger. Comparing the two requires shell-charcoal pairs, for which there is only one in the Dundas area, from the Early Holocene (McLaren et al 2007). Site T512-3 has three radiocarbon dates that place it in Period II (3500-1500 BP) . A basal shell midden date of 3170 ± 50 or 215 cal B C ( I I = 340 BC-160 B C ) (TO-13290) was collected from a core test (CT 2006-44). The upper shell midden deposit from the same core dates to 2780 ± 50 or 755 cal B C ( I E = 790 BC-715 B C ) (TO-13291); both these dates were on horse or butter clam. Charcoal from the hearth in H P 20 collected during excavation, dates to 1460 ± 50 or 605 cal A D ( IE = 560 A D - 645 A D ) (TO-13310). That the terminal shell midden date is earlier than the hearth date is not unusual since shell 22 ridges can accumulate laterally as well as vertically. Likely, the hearth date is a more accurate representation of the abandonment of the village. SiteT512-l Anthropogenic Landscape Site T512-1 has two components: a small "egalitarian" style village primarily located on a lower, front terrace, and an older shell midden located on a higher back terrace. The lower terrace contains at least one row of 7 small (6 x 10 m) similar-sized house depressions oriented to the southwest beach (Figure 5). There may be a additional rows o f houses in front and behind this, but the depressions are unclear. The earlier component of the site is located on a 13 meter elevation terrace and has one confirmed and two possible structural depressions. These depressions vary in size and the confirmed depression is larger (11 x 9 m) than the depressions on the terrace below. E d g e of s l o p e B a c k T e r r a c e S u b s u r f a c e s h e l l m a r g i n E x c a v a t i o n A r e a P o s s i b l e H o u s e s E l e v a t i o n = 13m a b l Intert idal rock w a l l s Intert idal B e a c h Figure 5. Site Map of T512-1 23 Radiocarbon Dates for site T512-1 Radiocarbon dates from T512-1 were calibrated in the same way as T512-3 discussed above. These dates show a much earlier occupation at site T512-1 than site T512-3. The base of the shell midden on the upper terrace is dated to 6830 ± 70 or 4960 cal B C ( IS = 5050 B C - 4890 B C ) (Beta 215178, shell). The upper deposit of the upper terrace shell ridge dates to 4910 ± 40 or 3310 cal B C ( IS = 3340 B C - 3250 B C ) (Beta 215178, shell). A charcoal sample from the hearth at H P 1 dates to 3650 ± 50 or 2025 cal B C ( I E = 2045 B C - 1945 B C ) (TO-13309, wood charcoal) indicating that this structure was built into the existing midden. Dates for the lower terrace have not yet been processed. The relationship between the back midden, the depressions on its surface, and the houses on the lower terrace, is complex. The dates for the back midden place its occupation prior to 5000 years ago, a time when the lower terrace would have been at mean sea level, and thus uninhabitable. The depression on the back ridge dates from much earlier, although after the occupation of the shell midden on which it is constructed and when the sea level had dropped to its present elevation. Two scenarios are possible to explain the upper depression. It is either a lone residence, constructed between the occupations of the back midden and the village on the lower terrace, or it is contemporaneous with the lower terrace houses. Its relatively early date suggests that it was not built after the lower terrace houses. Data for this research come from the confirmed depression, H P 1, located on the upper terrace of the site. The unusual characteristics of this upper terrace structure make this site unique, as it is not what is expected for small "egalitarian" style villages from this time period (Archer 2001; Matson and Coupland 1999: 187-188). A s I demonstrate in the next chapter, the upper terrace structure does not have the characteristics of a house and thus is not l ikely to be an isolated residence. I argue that H P 1 has the characteristics of a special-function building, one that included gatherings such as feasts, and that it was l ikely built and used for these purposes by the occupants of the lower terrace houses on top of what was, for them, an existing shell ridge. Thus, the functions of this building likely included a variety of community-focus interests. That these were located in a communal space rather than within a ch ie f s house, as in the later pattern, has implications for our understanding of the local culture history and the issue o f developing social complexity. 24 C h a p t e r 4: M e t h o d o l o g y a n d R e s u l t s A s discussed in the previous chapter, the significance of H P 1 at T512-1 is in its difference from both Tsimshian history as well as general behavioural patterns associated with domestic spaces. Focused testing o f this structure revealed differences on three levels: the comparison to H P 20 as a house in the Dundas region, houses in Tsimshian territory as demonstrated by previous archaeological work, and the expected material pattern of a house in general, which relates to the patterns o f everyday lived life. I argue that H P 1 differs from all three and is therefore another type o f structure. In this chapter, I describe the methodology and results of the excavations and analysis of the occupational surfaces, features associated with these structures and the analysis of bulk samples o f hearth material from the central hearth in each. M y aim is to demonstrate first, that H P 20 represents a Tsimshian house and second, that H P 20 and H P 1 are dissimilar. The typical characteristics of H P 20 are not only based on the congruency of the village style, house location and architecture as discussed in the previous chapter, but also from the construction and contents of the central hearth and occupational floor. A s discussed in Chapter 3, the central hearth is a location o f frequent use, producing patterns of behaviour that are associated with a range o f domestic and quotidian activities. Thus, a hearth is a good focal point for defining domestic use of the space. Whi le it is possible that H P 1 represents a different form of house than seen in the Tsimshian area and at HP20 from T512-3, H P 1 does not have the material characteristics expected in a space used on a daily basis. I argue that these data show that the function of H P 1 at T512-1 is non-domestic. In Chapter 5,1 explore the possible uses o f this structure. Excavation and Sampling Methods At site T512-3, H P 20 was selected because of its location in the second row and the likelihood that, based on its location and surface contour when compared to all other village and house excavations in Tsimshian territory, it would represent a typical Tsimshian style house. The depression was also reasonably clear of roots and other obstructions. A t site T512-1, interest in the upper terrace arose from the early shell midden dates from the 2005 field season. H P 1 was selected for the positive test of charcoal and high soil compaction near the centre of the depression, as these are signs of possible 25 occupational floors and a hearth. In contrast to H P 20, H P 1 was selected because it looked different from other buildings found at Tsimshian, and other N W Coast villages. The placement of the excavation units within the building depressions was determined through subsurface probe testing that aimed to locate possible hearths. At both sites, the perimeter of each depression was located and the geometric centre calculated. The subsurface was tested systematically with Oakfield soil probes by coring in a 1 x 1 meter grid down the centre-line of the length and depth o f each depression. Probing was done in 30 cm increments to a maximum depth o f 120 cm. The purpose of the probes was to locate high concentrations o f charcoal against the shell layers to locate possible hearths and to provide evidence that the subsurface had the same cultural deposition patterns as the surface. Excavation Data: Site T512-3 H P 20 measured 13.7 meters deep by 5 meters wide. The depression was U-shaped and oriented toward the beach similar to the orientation o f the other depressions on site. A systematic program of soil probing mapped the subsurface contour. Shell was encountered first on the perimeter and was deepest in the basin of the depression, where charcoal and ash were also present. A 1 x 6 meter unit was opened across the width of the depression. The unit was divided up into two 1 x 3 meter sections, Labeled Units C and D , for ease o f recording, but it was excavated as a trench. Stratigraphy A n average o f 80 cm of non-cultural humus was removed by shovel shaving (Figure 6 and 7). A 10 cm layer of leach and a layer of black organic soil that varied between 15 and 20 cm was found below the humus and above the shell. Shell was exposed at 80 cm on both the east and west sides of the unit and was the first encounter with cultural material during excavation. This layer was comprised o f loose shell with lots of whole pieces. The layer o f shell followed the surface topography, sloping into the basin of the depression but it became more dense and fragmentary at 110 cm below unit datum ( B U D ) . In the centre of the depression, cultural material was not encountered until 159 cm B U D . A n occupational surface was uncovered at 179 cm B U D and was defined as a compact layer of dark soil with shell fragments. It had associated features, including postholes and a hearth, which w i l l be discussed in further detail below. The basal layer of the unit was loose midden composed of small 26 fragmentary shell. Excavation was completed at 200 cm B U D , as the purpose of the excavation was to only uncover the most recent occupational surface. SOUTH WALL UNIT B O T T O M Figure 6. North and South Profile Maps of HP 20 at site T512-3 27 EAST WALL UNIT BOTTOM Figure 7. West and East Wall Profile Maps of HP 20 at site T512-3 Features There were 10 postholes ranging in diameter from 10 to 30 cm on the occupational floor. Most were oval, with 3 being circular. A l l but 2 of the postholes were exposed near the occupational surface and extended below 200 cm B U D (Figure 8). In Unit D , the larger postholes were located on the outside o f the depression, while smaller ones were located closer to the central hearth. Unit C had three small postholes near the central hearth. A hearth was identified on the occupational surface and was a distinct deposit o f grey ash. It was found in the centre of the depression between Unit D and Unit C . A small amount o f charcoal and shell were in the hearth; however, it was mostly ash. Underneath this layer of ash was a series of moderately sized rocks, which could represent an associated construction. The north wal l profile shows a pocket o f ash at 179 cm B U D and the south wall has a pocket of ash at 190 cm B U D (Figure 6 and 7). These might be part of the same hearth that shifted over time. A s mentioned previously, a radiocarbon sample was taken from the hearth and was dated to 1460 ± 50 or 605 cal A D ( I I = 560 A D - 645 A D ) (charcoal), which places the building and the latter occupation of the village in Period II (3500-1500) B P o f the north coast chronology. 28 U N I T C 1/ LayerD 180 Layer F 177 ^ - - 191 ® top of siope bottom of slope 168 181 bottom of slope 162 162 154 I ' \ 133 \ > top of slope Layer F 127 Layer D UNIT D Plan Map T512-3 HP 20 LEGEND Rock Shell Post hole QQ Ash Bone point Elevation (BUD) ~y\ Slope Scale (1:20) in cms Figure 8. Plan Map of HP 20 at T512-3 29 Excavation Data: Site T512-1 The excavation at T512-1 was conducted in H P 1 located on the back terrace. H P 1 measured 11 meters deep by 9 meters wide. The depression H P 1 was oriented in the opposite direction than the depressions on the lower terrace. Testing showed that the subsurface matched the surface contour. Shell was encountered first on the perimeter and was deepest at the basin of the depression, where charcoal and ash were also present. A 1 x 2 meter unit was opened near the centre of the depression at the heaviest concentration of charcoal, as we were targeting a central hearth. Stratigraphy A n average 80 cm thick layer o f humus was removed from the unit using shovel shaving. Cultural material was first encountered at 88 cm B U D in the southwest corner o f the unit (Figure 9 and 10). This was a combination o f charcoal and white ash, and was determined to be a hearth. A t 82 cm B U D , a layer of red soil was exposed, which was followed by charcoal rich soil at 95 cm B U D . A t 100 cm B U D , a thin layer o f brown ash was exposed at the east end of the unit and was identified as a second hearth. Directly underneath, a charcoal-rich soil was exposed and continued to the occupational surface being as thick as 20 cm in some places. A small pocket of crushed shell was exposed in the southwest corner of the unit at 115 cm B U D . The excavation was completed at 130 cm B U D , where an occupational surface was exposed. The occupational surface was defined as a layer of compact finely ground shell mixed with dark charcoal stained soil. 30 _ _ i —. UNIT BOTTOM UNIT BOTTOM PROFILE MAP T512-1 Hp1 LEGEND [~R~] RooK H wood Dfinse shell ' ' fragments • — crushed shell F Q ^ J White ash Browf-asti -fj Biack organic soil [ 0 ] Brown soil & » — * crushed ah'iil [ l s t e "" ' Scale (1:10) in cms Figure 9. South and North Wall Profile Map of HP 1 at Site T512-1 31 Figure 10. East and West Wall Profile Maps of HP 1 at Site T512-1 Features N o postholes were found at T512-1; however, two hearth features were exposed in the excavation unit. Hearth 1 was the larger of the two features (Figure 11). It was first encountered at 88 cm B U D in the southwest corner o f the unit and continued to the occupational surface at 130 cm B U D . It was a mound feature and was encircled by a rock construction. Large pieces o f charcoal were found throughout the hearth feature. A ground stone pendant was found inside the hearth, which is discussed in further detail below. Hearth 2 was first encountered at 100 cm B U D in the northeast corner o f the unit and continued to 113 cm B U D . A few rocks were near this feature as well ; however, there was no defined rock construction like the one around Hearth 1. The feature also had a mound shape. A s mentioned previously, a radiocarbon sample was taken from Hearth 1 and is dated to 3650 ± 50 or 2025 cal B C ( I E = 2045 B C -1945 B C ) (charcoal), which puts the upper component of T512-1 in Period III (5000-3500 B P ) in the north coast chronology. 32 Figure 11. Plan Map of HP 1 at Site T512-1 Sampling and Processing Methods for Bulk Samples Several types of samples were taken during excavation, which included column samples, house floor samples, off site control samples and bulk samples from the hearth features. This research focuses on the bulk samples from the hearth features because the aim was for an intensive study o f the hearths and contents. At T512-3 the hearth was only evident in one layer thus the entire layer was collected. Because of the size of the hearth at T512-1, bulk samples were collected in arbitrary levels of 10 cm during excavation. 33 Flotation and Analysis of Samples A random soil sample of from T512-1 and T512-3 was floated totaling 6 litres per site. Flotation was conducted by hand in the lab using the bucket method as described by Pearsall (2000: 35). The volume o f each was recorded prior to flotation to assure that the data could be standardized (Pearsall 2000: 108). For a systematic flotation procedure, see Appendix 1. Sorting and Identification of Light Fraction Laboratory work was done to quantify and identify the material found in the samples. After weighing the light and heavy fractions, the samples were sorted systematically. The light fraction material >.2mm was sorted first using a lighted magnifying glass. Categories for the sorting included: suspected seeds, charcoal, charred needles and uncharred plant materials. Once sorted, each category was weighed. The large quantity of light fraction material made it necessary to examine a 1 gram sample of material < .2mm under the microscope. Suspected seeds were placed on a glass dish underlain by graph paper and looked at under higher magnification using a Nikon S M Z 800. To determine i f seeds were identifiable, characteristics of size, shape, and seed coat texture were considered. Comparative collections already in store at University o f Brit ish Columbia in the Laboratory of Archaeology were used for identification. Sorting and Identification of Heavy Fraction Heavy fraction samples were sorted by hand. Categories for the sort included rock, shell, and other faunal remains. Each sample was weighed and recorded. Shell varieties and faunal remains were identified using Martindale's northern comparative collection in the Laboratory of Archaeology at the University of Bri t ish Columbia. Sorting and Identification of Ecofacts and Artifacts This research also used material found onsite during excavation. The majority o f soil was screened on site through XA inch screens. A sample of the material was also screened through 1/8 inch screens. A catalogue was made of excavation material and classifications were divided by site, type, and material. A l l material was weighed and counted. Various artifacts were measured for length, width, and thickness. Faunal material found was identified to the lowest possible taxa ranging from family to species using comparative collections at U B C in the Laboratory o f Archaeology. 34 Results: Material Culture This section w i l l describe the faunal and artifact material recovered during excavation. The majority o f material removed from the wet screen consisted of faunal remains. Quantity and variety of faunal materials differed significantly between the two sites. The bones were identified to the lowest possible taxa ranging from family to class. Faunal material from site T512-3 had a total weight of 23.82 grams while T512-1 only had 0.80 grams. Site T512-3 contained a greater variety of animal type including bird, fish, mammal, while T512-1 only had fish and mammal (Table 5, Appendix 2). Table 5: Faunal Remain Comparison T512-1 Weight (g) NISP T512-3 Weight (g) NISP Fish 0.30 7 10.5 147 Mammal 0.20 1 9.30 16 Bird - - 0.60 2 Unidentified 0.30 9 3.42 20 Total 0.80 17 23.82 185 A low density o f artifacts were found at both sites during excavation. L o w artifact densities are not uncommon in both shell middens and house floors, and the excavation strategy was focused on architectural features and their contents more than artifacts. Two bone points were found at T512-3 and a bone point, a ground stone pendant and a quartz flake were recovered at T512-1 (Table 6, Appendix 2). 35 Table 6. Artifacts Site Provience Artifact Material Size (mm) Weight (g) Description T512-3 inch Point (DIP screen, first 114) shell layer. Bone 76 x 8.8 x 4.02 4 4 Gradually tapering point, greatest width below midpoint, small notch at base. T512-3 175 cm Point (DIP BUD, UnitC 119) (Figure 8) Bone 64.4 x 11.2 x 4.07 6 Gradually tapering point, greatest width below midpoint, smal polished groove. Breakage from midpoint to artifact base T512-1 Vi inch screen Point (DIP 188e) Bone 23.5 x 5.5 x .1 3 Highly polished with a fine point. Broken but gradually tapering point possible. T512-1 117-127 cm Groundstone Basalt 76.1x13.4x 10.5 BUD, Hearth Pendant 1 6.4 Rounded at both ends, grooved on one end, little use wear (Morin personal communication). Results: Hearth Bulk Sample Data Hearth contents at T512-3 and T512-1 showed different proportions of shell, FCR/sand, charcoal, fauna, and plant remains. As mentioned previously, the hearth contents at each site were sampled and 6 litres of bulk material were processed using flotation (Appendix 3). Based on a Chi-square test the variation between the two hearth samples was significant (Table 7). The content proportions were calculated for each site by dividing the total weight of the each material by the total weight of the sample (Appendix 4). Table 7: Chi-square Test of Hearth Sample Ecofacts (in weight per gram) Shell FCR/Sand 1031 Charcoal 152 Fauna ) Seeds Total Site T512-1 16.4 " Tj o 1199.7 Site T512-3 1921 661 51.8 .4 .1 2634.3 Total 19374 1692 203.8 0.7 j 0.1 3834 [Degrees of freedom: 4 ? Chi-square = 1704.49273205055 ? p is less than or equal to 0.001. ? The distribution is significant.] Shell A much higher proportion of shell was found at site T512-3 with 73 % of the sample consisting o f shell compared to only 1.4% at T512-1 (Table 8). Total weights o f the shell amounted to 1921.2 grams per 6 litres at T512-3 and 16.4 grams per 6 litres at T512-1. Samples from T512-3 not only had more shell by 36 weight but also bigger pieces and more variety. Species represented included cockle, mussel, barnacle, sea urchin, also butter and horse clam. Shell found in the T512-1 samples was too small for identification. The small fragments of the shell in T512-3 is consistent with shell being used as a floor layer in a high-traffic area. Fire-Cracked Rock (FCR) and Sand F C R and sand made up 86% of the weight in T512-1 while only 25% of T512-3 consisted of F C R and sand (Table 8). This shows a much higher proportion in the hearth at T512-1 when compared to T512-3. The total weight of F C R and sand was 1031.8 grams per 6 litres at T512-1 and 661.1 grams per 6 litres at T512-3. There were no large pieces o f F C R in either of the samples but samples from T512-1 contained more beach sand. Charcoal Charcoal proportions were also different between the two sites. Charcoal made up 13% of the total sample at T512-1 and only 2% at T512-3 (Table 8). The charcoal weighed 152.3 grams per 6 litres at T512-1 and 51.8 grams per 6 litres at T512-3. The T512-1 sample had larger pieces of charcoal than the sample from T512-3. A s noted earlier, the hearth at T512-3 contained mostly ash, indicating a high-combustion/high heat feature, l ikely one that was cleaned regularly. The hearths at T512-1 had much higher charcoal concentrations indicating a reduced atmosphere and suggesting less cleaning. Fauna and Flora Remains The proportion o f fauna and flora remains were low at both sites with only .4 grams at T512-3 per 6 litres and .3 grams per 6 litres at T512-1 (Table 8). These amounts are both too low to figure into the proportion calculation. There were a few charred fish vertebrae in the samples from both sites; however, most of the very small fragments were not identifiable. Samples from T512-3 contained 3 seeds with a total weight o f .1 gram per 6 litres. These seeds were charred and had no identifiable characteristics. Seeds and other plant remains were absent from the T512-1 sample. 37 Table 8. Hearth Bulk Samples Weight (g) and Proportions Site T512-3 Site T512-1 Weight (g) Proportion Weight (g) Proportion Shell 1921.2 73% 16.4 1.4% FCR/Sand 661.1 25% 1031.8 86% Charcoal 51.8 2% 152.3 13% Fauna .4 - .3 Seeds .1 - 0 Modern Plant 4.7 - 3 Summary of Results The data from this research show that H P 20 at site T512-3 and H P 1 at site T512-1 are dissimilar on almost every level o f comparison from hearth contents to village type (Table 9). If my logic is correct, and H P 20 is a domestic structure, then these data indicate that H P 1 is not a house but some other type of structure, one that was used less frequently. This argument is based on the premise that the intense activity represented at T512-3 reflects the history of daily life within a domestic space. M y logic is based on a comparison to both other Tsimshian houses excavated at other sites and to more general behaviour patterns found in domestic structures. Differences between the two structures in their hearth contents demonstrate two different functions o f the hearths. Samples from the hearth in H P 20 had more shell, more other fauna, and plant remains than the hearth in H P 1. Although the fauna and plant remains were modest, this difference suggests that cooking was taking place around the hearth at H P 20 more often than at H P 1. Proportions o f charcoal in the two hearths also support this, as higher amounts o f charcoal at T512-1 are indicative of a large low heat fire. Less maintenance and cleaning o f a hearth can also be inferred from the high proportions of charcoal, which represents a hearth that is not used daily. Artifacts and ecofacts recovered during excavation also support a non-domestic function of H P 1. Over al l , H P 20 had a higher quantity and variety o f fauna than H P 1, which shows a more frequent and longer use of the structure. Artifact assemblages were small but the presence of a ground stone pendant in H P 1 may represent a ritual use o f the structure, as objects o f personal adornment are often in association with a ritual context. Differences between the structures in hearth shape and style also seem to represent two different functions. The hearth in H P 20 had a basin shape, typical of a hearth that is cleaned on a regular basis. 38 Pockets of ash observed in the profiles at H P 20 might indicate a longer period of occupation in the structure, as they might be other hearths from a later period in time. Mound hearths like the ones in H P 1 are most commonly found in an outdoor context. These are fires that are built possibly just for warmth and light and are neither used nor cleaned regularly. The existence of two mound hearths located close to each other also suggest infrequency of use since sequential fire pits are not constructed in exactly the same place. Ten postholes were excavated in H P 20 but no postholes were exposed in H P 1. There is a variety o f explanations for this. First, in both depressions we targeted the central hearth and postholes on the perimeter of the structure would not be excavated, especially because the unit in H P 1 was 1 x 2 m. Second, this could represent different types o f architecture. Cooking hearths are often associated with postholes that represent tripods or racks near the fire; feasting hearths are usually free o f such associated structures (Martindale 1999). Occupational surfaces in the two depressions were similar as they both were a composition o f charcoal stained soil mixed with crush shell. However, it was observed that the occupational surface at H P 20 was more compact than H P 1. This again suggests a less frequent use of H P 1. A s discussed in Chapter 3, there are characteristics of the structures and their location in the villages that indicate H P 20 is a house and H P 1 is not. H P 1 was larger than the structures on the terrace below, as well as being oriented in a different direction. Its location on the back terrace, above and behind the house row suggests that H P 1 was different from the other buildings. Contemporaneity between H P 1 and the lower terrace houses has not been established through absolute dating, but the following is clear: 1) H P 1 is not associated with the back terrace shell midden that dates from about 1000 years earlier, 2) H P 1 may represent an isolated building, although this seems unlikely since small villages like T512-1 are known from elsewhere in the Dundas Islands dating to between 5000 and 3500 B P , and 3) even i f H P 1 is not associated with the lower terrace houses, it is not a domestic space and must serve a resident population from somewhere nearby. Vil lage level data are an important inclusion in the data set. A s discussed previously, T512-3 has characteristics associated with the Prince Rupert Harbour style and is a large "ranked" village with 24 39 houses. Site T512-1 is an "egalitarian" village with 7 houses on the lower terrace and has one confirmed structure on the upper terrace. A small "egalitarian" village style with a non-domestic structure located away from the village might mean a different type of political organization. The significance of this for understanding Northwest Coast culture history w i l l be discussed in further detail in the following chapter. Lastly the dates of the two sites and structures are important both for comparison to each other but also to the larger picture o f north Coastal Brit ish Columbia prehistory. Early dates at T512-1 in Period III (5000-3500 B P ) , coupled with the spatial organization of the village possibly indicate an egalitarian political organization. In summary, the data show that 1) H P 20 is a house, 2) that H P 20 and H P 1 are dissimilar, and 3) that H P 1 is not another type of house but rather a non-domestic structure. The possible function o f H P 1 w i l l be discussed in the following chapter as wel l as the inference of this early non-domestic structure in understanding the development of cultural complexity on the Northwest Coast. 40 Table 9. Summary of All Data from T512-1 and T512-3 T512-1 T512-3 Hearth Contents Faunal (Shell) FCR/Sand Charcoal Faunal (Other) Plant Remains 16.4 g, 1.4% of 6 litre sample. 1031.8 grams 86% of 6 litre sample. 152.3 g, 13% of 6 litre sample. Large pieces in sample. (Indicates low heat fire) 0.3 g of fragments, charred fish vertebrae No plant remains 1921.2 g, 73% of 6 litre sample. 661.1 g, 25% of 6 litre sample. 51.8 grams, 2% of 6 litre sample. (Indicates high heat fire) 0.4 g of unidentified fragments, charred fish vertebrae 0.1 g of 3 unidentified seeds Wet Screen Fauna Fish Mammal Bird Unidentified 0.30 g 0.20 g 0.20 g 10.5 g 9.30 g 0.60 g 3.42 g Artifacts Bone Points 1 small point (23.5 x5.50 x 3 mm, 0.10 g) Personal Adornment Ground stone pendant 2 medium sized points (64.4 x 11.2 x 6 mm, 4.07 g and 76 x 8.8 x 4.4 mm, 4.02 g) None Features Structure Village Dates Hearth shape Hearth construction Hearth location in Structure Postholes Occupational Surfaces Size Location in Village Number of Structures Type Location in Dundas Region Radiocarbon 2 Mound hearths Rock border Near centre of structure but dispersed None Charcoal stained soil with finely crushed shell 1 1 x 9 meters Upper terrace, away from main village 7 on modern shoreline, 1-3 on upper terrace Small "egalitarian" style Island 53 3650 ± 5 0 or 2025 cal B C ( 1S = 2045 BC- 1945 BC) (charcoal) Archaeological Period Period III (5000-3500 BP) 1 Basin hearth No construction Centre of structure 10 postholes Dark soil with shell fragments 13.7 x 5 meters Second row of village 24 on modern shoreline Large "ranked" style (Type 1) Baron Island 1460 ± 5 0 or 605 cal AD(1S = 560 AD- 645 AD) (charcoal) Period 11(3500-1500 BP) 41 C h a p t e r 5: D i s c u s s i o n In this thesis, I argue that the village site o f T512-1 includes a non-domestic structure (HP 1) of a type not found in later villages such as at T512-3. M y logic is based on a comparison o f villages and houses in Tsimshian territory and a focused comparison between what is arguably a later, typical Tsimshian house (HP 20) and the non-domestic structure. I have examined data on several scales of analysis from the village layout, to the building architecture, to building contents, and finally through a study of the central hearths and their contents in both structures. I believe that these data show that 1) H P 1 is non-domestic when compared to H P 20, the specific analogy o f other Tsimshian houses, and general analogies of domestic behaviour, and 2) that the essence o f this difference appears to be specific ceremonial and ritual behaviours that, in later times, are associated with the large houses of village chiefs. I develop this argument below and consider the implications to the concept o f the evolution of social complexity of locating an earlier community in which ritual and ceremonial activities that become part of a hierarchical social and political organization are not absent, but disassociated from individual households. In this chapter, I discuss the importance o f the central hearth as a location for daily activities, first using ethnographic data to demonstrate that daily life was structured by patterns of activities and then by relating this to the data from both H P 20 and H P 1 to show that H P 20 is more indicative of a residence than H P 1. Fol lowing the discussion on hearths, I evaluate the non-residential structures known on the Northwest Coast in order to explore possible non-domestic functions of H P 1. I conclude that H P 1 is different than any known non-residential structures and is potentially a type of community meeting house unknown in both the archaeological and ethnographic record. Next, I relate the data back to the regional context to illustrate that T512-1 is not an anomaly but is part of a larger regional history with early dates and settlements. Daily Life of the Tsimshian in Ethnographic Sources Ethnography gives some clues into the importance of the structuring of daily life around the hearth in prehistory and provides evidence that these spaces are a central location where a good portion of domestic activities took place. Through ethnographic observation and the study of oral traditions, Boas (1916) 42 inferred that daily activities in Tsimshian life prior to contact with Europeans include cooking, gathering of firewood, the serving o f food and use of dishes and utensils (Table 10). In the absence of direct, ethnographic observation, it is unknown how daily life was structured by rules governing these behaviours. However, these data provide some evidence that the repetition of daily activities produced material patterns observable in the archaeological record that help identify domestic spaces. Table 10. Daily Life of Tsimshian from Ethnographic Sources Type of Activity Tradition (as inferred from Boas as mentioned in oral history) Social relationships involved Expect Material Attributes Cooking by Boiling (Fresh Foods) Done in square box or root basket, hot rocks dropped into container with water and food substance, covered with mat Women (slaves or wife of house owner) Fish bones, plant remains, FCR, charcoal Cooking by Steaming Steaming was done in an underground oven. A large fire is built and flat stones are heated, these are thrown into a pit that is lined with skunk cabbage. Women (slaves or house owners wife) Pit, Large FCR, charcoal, disturbed, plant remains, fish bones. Preservation of Food (drying and smoking) Meat (fish and land and sea mammal) and berries are provisioned by drying and smoking for winter and stored Women Concentrated area of berries or fish bones (processing centre, most likely outdoor hearths) Fire Management Fire wood collection of pitch wood, driftwood and spruce. Rotten cedar considered poor kind of fuel. Fire started with pitch wood and tinder. Men Charcoal (specific varieties), hearth construction, ash Daily Feasting Two course meals. 1st fish with oil, water passed around to drink, 2nd berries and oils. No mention of social or gender rules Fish bones, berry remains. Daily Feasting (utensils) Food eaten from carves dishes and wooden spoons, spoons (also carved from mountain goat horn). Placed in dishes, taken out with spoon and eaten No mention of social or gender rules Dishes or spoons Hearth Seating Mats were spread around hearth for seating. Owner of house or Chief at the rear of fire, guests to the sides (mostly right side) Plant remains, compact occupational surface (Source: Boas 1916) In considering the data from this research, the hearth and structure at site T512-3 has the most characteristics associated with repeated daily-life activities. This includes plant and animal remains from foodstuffs and the maintenance and cleaning of the hearth, which would have involved the repeated 43 removal of ash, charcoal and other debris. Repeated use of a space is also supported by the structure's multiple postholes and compact occupational surface. On the contrary, the hearth and structure at T512-1 does not demonstrate these patterns o f daily use. H P 1 lacks the accumulation of food remains and the hearth does not appear to have been maintained. Thus, because of the general behavioural patterns expected in a domestic context, H P 1 is not a different type of domestic structure than H P 20 because it does not have the necessary characteristics o f a space that is used on a daily basis. HP 1 as a Non-residential Structure There are a variety of non-residential structures mentioned in Northwest Coast ethnographic sources (Table 11). Most o f these structures have a ritual function; however, defensive structures and smoke houses have also been mentioned. Non-residential structures are rare in later periods (Period I and II) on the northern Coast and, until this research, completely unknown in the earlier Period III (Matson and Coupland 1995). This means that T512-1 is the earliest known non-residential structure on the northern Coast. Table 11. Non- domestic structure material expectations on the Northwest Coast Type Expected Material Attributes Ethnographic Example and Reference Sweat bath house Large exterior hearth, central pit/basin, distinctive seating arrangements, no internal hearth, high quantities of FCR, small structure Tlingit (deLaguna 1972: 305) Shrine Very few artifacts, very clean occupational surface, unusual diversity of artifacts, human remains Nuu-chan-nulth: (Boas 1930: 266-269 Arima& Dewhirst 1990: 395) Secret Society Meeting House Specialized seating, special hearth, prepared floor, distinctive food remains, ornaments or ceremonial regalia, limited trampling of floor, smaller structures, low artifact densities, rare artifact types Nuxalk (Bella Coola) (Mcllwraith 1992a: 177; Mcllwraith 1992b: 11,17,32) Shaman's house Specialized Seating, special hearth, prepared floor, distinctive food remains, ornaments or ceremonial regalia, limited trampling of floor, smaller structures, low artifact densities, rare artifact types Tlingit (Oberg 1973:19) (Modified from Morin 2006) 44 According to the material characteristics of H P 1 at site T512-1 there are not enough expected material attributes to conclude i f H P 1 was particular type of non-domestic structures recorded ethnographically. Fol lowing the expected material attributes o f these types of structures listed in Table 11, PIP 1 is unlikely to have been a sweat bathhouse because there was also no central pit or basin shaped hearth. Although, there was higher quantity of F C R found in HP1 than in H P 20, this evidence is insufficient to support the function of H P 1 as a sweat bath house. It is also unlikely that H P 1 was a shrine as the structure is large in general terms and much larger than the houses in the village below. The occupational surface was not clean but contained two large hearths. The lack o f an unusual diversity of artifacts and human remains also suggest that H P 1 was not a shrine. It is possible that the infrequent use and larger fires in H P 1 could be interpreted as a smoke house structure; however, I argue that due to its location and the absence of fish elements, this is not a l ikely explanation. Smoke houses are generally located near procurement areas, usually near intertidal fish traps. H P 1 's location on top o f a ridge makes its use as a food processing structure unlikely. Strictly defensive structures have not been recorded in Tsimshian territory, although references to them are found in the oral record (Coupland, Martindale and Marsden 2001); it is a possibility that the T512-1 was used as a refuge. Its location on a ridge is more defensive than on the lower terrace, but there is no other evidence that it was used strictly for defense. The structure is also older than the hypothesized conflict that is described on the Dundas Islands in the oral traditions (Marsden and Martindale 2003). Despite these qualifications, it is possible that the structure served several functions, one of which could have been defensive. Lastly, there were no rare artifacts, ceremonial regalia or specialized food found in the H P 1 excavation and samples, as associated with secret society meeting halls (Mor in 2006). Thus, it is unlikely that the function of this structure was for secret society meetings or that it was a shaman's house. The available data do not indicate that H P 1 was any o f the special function buildings identified ethnographically. This is not entirely surprising since it is almost 4000 years earlier than these forms. A more fruitful route o f analysis is to consider the types o f activities found in H P 1 and consider that it may 45 represent a form of structure for which there is no ethnographic or archaeological counterpart. It appears to have been used regularly, but infrequently, for large gatherings in which the central hearth was not used for cooking. This suggests that the building was used for social gatherings, perhaps feasts. Thus, our best analogy is not a special function building, but the large houses of chiefs that, as recorded ethnographically and archaeologically (Coupland 2006), were used to host village ceremonial and social events. In the later period, as Coupland (2006) argues, chiefs established and maintained their authority in part by hosting such events in their houses. Chiefly households are thus associated with large houses. Perhaps what we find at H P 1 at T512-1 are these ceremonial and social functions removed from the control of a village chief and located in a communal building. Such activities do not appear in H P 20 at T512-3, because they presumably take place at the large houses in the centre-front of the village. Thus, H P 1 represents not only a new kind o f structure, but also a more detailed view of what constitutes an "egalitarian" village. Early Settlements in the Dundas Region Unt i l the Dundas Island Project, no sites from before 5000 years ago were known in Tsimshian territory. This research has located 6 sites that have components dating to before 5000 years ago and 4 with dates from before 6500 years ago. Thus, the early dates at T512-1 are not an anomaly. For example, site T522-1 (temporary site number) is a small hamlet located on a 13 m terrace (Figure 12). Excavations in the 2007 field season confirmed a structure on this terrace. The building has yet to be dated, but the terrace itself dates from 3180 cal B C ( I E = 3320-3070 B C ) (Beta 215183) to 5040 cal B C ( I I = 5130-4960 B C ) (Beta215180). Martindale (pers. comm. 2007) hypothesizes that this structure is contemporaneous with the shell ridge and that there may be 3-4 such buildings in total representing a small village. Thus, T512-1 appears to be part of a long history of village occupation in the Dundas Islands area. While no other non-domestic structures are known from this time, the fact that house construction had been taking place for over 1000 years before H P 1 was built makes it less unusual that non-residential buildings were part of the Period III architectural repertoire. 46 I n t e r t i d a l B e a c h "} / S u b s u r f a c e s h e l l m a r g i n E d g e o f s l o p e B a c k T e r r a c e H o u s e y ( •••••--,„-/ P o s s i b l e H o u s e s E l e v a t i o n = 1 1 m a b l J a g aa • T 1 " Figure 12. Site Map of T522-1 Implications for Social Complexity Northwest Coast theoretical perspectives on culture change arguably include two broad themes: those that reference external stresses such as population pressure and environmental changes to explain culture change and those that use internal factors negotiated between people such as conflict, volition, decision-making or agency (Martindale 1999). This research provides evidence that internal factors are significant in both the construction of and changes to cultural history, although it does not exclude the significance o f external forces. One of the key theoretical issues on the N W coast has been the development o f so-called social complexity. One of the assumptions in complexity studies to date is that later, more complex communities form through the gradual accumulation o f new traits. Thus, ceremonial feasting, which is a 47 characteristic o f later complex hunter gatherers, is thought to be absent from earlier, less complex times. Cannon (1998) is a notable exception to this logic. He argues that feasting was l ikely present from the earliest times on the N W Coast. In a similar vein, Martindale (2001) and Coupland (2006) argue that complexity is less about cultural traits and more about the institutionalization of inequalities o f power in the form of differential access to and control over social, economic, and political power. M y research shows that ceremonial activities were present from some of the earliest times on the N W Coast, much earlier than previously known, and associated with some o f the earliest villages. H P 1 at T512-1 can be interpreted as being from a time when ceremonialism was not tied to hierarchies of power. Instead, it seems as though the social gatherings that appear represented in the large mound hearths at H P 1 were collective expressions of the entire community, an arrangement that may represents a more egalitarian organization than formed later at villages such as T512-3. That such activities became incorporated within individual houses of village chiefs by Period I is an expression of control in which ceremonialism was used to augment and symbolize the political power of village leaders (Coupland 2006). Cannon (1998) argues that feasting is an expression of social cohesion and identity that links people together through tradition, ritual, and exchange. If H P 1 has a ceremonial function, then such purposes are l ikely. It is also possible that the collective expression o f ceremonialism found at H P 1 may have been as much an effort to forestall hierarchy, along the lines of Lee's (1990) leveling mechanisms, as it may have been an expression of an egalitarian ethic. This thesis demonstrates that domesticity and architecture have implications for understanding social and spatial relationships. We see in the changes from T512-1 to T512-3, the movement o f ceremonial functions from outside the house in small "egalitarian" villages to inside the largest houses in "ranked" villages. Fol lowing Cannon (1998), I posit that people were feasting in the early periods and that H P 1 is a representation o f community access to feasting structures. The evidence for this argument is based on a range o f data, including the village spatial organization, the size and location of H P 1 in relation to the other structures, the infrequent use of the structure supported both by the hearth construction and maintenance, and the contents o f the hearth. Thus, I interpret H P 1 as evidence of an early period special 48 function building where feasting and ceremony were taking place outside of the domestic sphere. This suggests an egalitarian control over ritual in Period III (5000-3500 B P ) and that cultural change to Period II (3500-2000 B P ) was motivated by agency-like negotiations over ritual as much as access and control over resources (Archer 2001; Coupland 2006). 49 C o n c l u s i o n This thesis explored the function of H P 1 at T512-1, an early structure on the Northwest coast and concluded that it represents a non-domestic structure, most l ikely a community meeting structure used by the people l iving in the village below as an expression o f social cohesion. Thus, H P 1 represents not only a new kind of structure, but also a more detailed view of what constitutes an "egalitarian" village and in larger political context what constitutes "egalitarianism". A s Lee (1990) argues, equality can be enforced and an egalitarian political structure is not necessarily the absence o f power but rather a forestalling o f a hierarchal control of that power. In this way these early villages on the Northwest Coast comment on issues of political structure and enhance how we understand changes in social complexity. Wi th this in mind, we can see that it is not a slow accumulation of traits over time that makes a culture more complex as Matson and Coupland (1995) suggest but that there is a shift in power in social and political organization from an enforced equal or 'egalitarian' control to institutionalized unequal access to resources, or in the case o f this research the access to ceremony and feasting. It may difficult to determine the triggers for why these shifts in power occur; however, an agency-oriented approach can help us understand the complicated nexus of human history and the behaviour, interaction, relationships and conscience decisions that all play into change. In conclusion, I have demonstrated that early Northwest Coast prehistory is not simply the accumulation of more 'complex' traits but that it is l ikely that shifts in pre-existing power over resources and ceremony also contributed to increasing complexity over time. 50 B i b l i o g r a p h y Ames, Kenneth 2006 Thinking about Household Archaeology on the Northwest Coast. In Household Archaeology on the Northwest Coast. Edited by Elizabeth A . Sobel, A n n Trieu Galir and Kenneth M . Ames. International Monographs in Prehistory, A n n Arbor. Ames, Kenneth M . and Herbert D . G . Maschner 1999 Peoples of the Northwest Coast: Their Archaeology and Prehistory. 1 st ed. Thames and Hudson, London. Archer, D .J .W. 2001 Vil lage patterns and the emergence of ranked society in the Prince Rupert area. In Perspectives on northern Northwest Coast prehistory, V o l . 160, pp. 203-222. H u l l : Canadian Museum of Civi l izat ion, Arnold , J .E. 2006 Households on the Pacific Coast: The Northwest Coast and California in Comparative Perspective. In Household Archaeology on the Northwest Coast. Edited by Elizabeth A . Sobel, A n n Trieu Gahr and Kenneth M . Ames. International Monographs in Prehistory, A n n Arbor. Boas, F . 1916 Tsimshian Mythology. Johnson Reprint Corporation. New York . Bourdieu, P. 1977 Outline of a theory ofpractice. Cambridge University Press, Cambridge. Cannon, Aubrey 1998 Contingency and agency in the growth of Northwest Coast maritime economies. Arctic Anthropology. 35(1): 57-67. Coupland, Gary 2006 A Chiefs House Speaks: Communicating Power on the Northern Northwest Coast. In Household Archaeology on the Northwest Coast. Edited by Elizabeth A . Sobel, Ann Trieu Gahr and Kenneth M . Ames. International Monographs in Prehistory, A n n Arbor. 1988 Prehistoric cultural change at Kitselas Canyon. Canadian Museum of Civil izat ion, Mercury Series, Archaeological Survey o f Canada Paper No.138, Ottawa. Coupland, Gary, Roger H . Colten and Rebecca Case 2003 Preliminary Analysis of Socioeconomic Organization at the M c N i c h o l Creek Site, Brit ish Columbia. In Emerging from the Mist: Studies in Northwest Coast Culture History. Edited by R . G . Matson, Gary Coupland and Quentin Mackie. U B C Press, Vancouver. 152. Coupland, Gary, Andrew R . C . Martindale and Susan Marsden 2001 Does resource abundance explain local group rank among the coast Tsimshian? Perspectives on northern Northwest Coast Prehistory. 160: 223-248 51 Gahr, D . A n n Trieu, Elizabeth A . Sobel and Kenneth M . Ames 2006 Introduction In Household Archaeology on the Northwest Coast. Edited by Elizabeth A . Sobel, A n n Trieu Gahr and Kenneth M . Ames. International Monographs in Prehistory, A n n Arbor. Grier, C o l i n 2006 Temporality in Northwest Coast Households. In Household Archaeology on the Northwest Coast. Edited by Elizabeth A . Sobel, A n n Trieu Gahr and Kenneth M . Ames. International Monographs in Prehistory, A n n Arbor. Hayden, Br ian and Aubrey Cannon 1982 The corporate group as an archaeological unit. Journal of Anthropological Archaeology 1(2): 132-158. Lee, Richard B . 1990 Primitive Communism and the Origin o f Social Inequality. In The Evolution of Political Systems: Sociopolitics in Small Scale Sedentary Societies. S. Upham, ed. pp.225-246. Cambridge University Press. Lightfoot, K . G . , A . Martinez and A . M . Schiff 1998 Dai ly practice and material culture in pluralistic social settings: A n archaeological study of culture change and persistence from Fort Ross, California. American Antiquity 63(2):199-222. MacDonald , George and Charles Borden et.al 1969 Current Archaeological research on the Northwest Coast. Northwest Anthropological Research Notes. Volume 3. MacDonald , George and Richard Inglis 1981 A n overview of the North Coast Prehistory Project. In BC Studies No. 48, Fragments of the Past: British Columbia Archaeology in the 1970s, ed. B y K . Fladmark. 37-63. Marshall , Yvonne 2006 Houses and Domestication on the Northwest Coast. In Household Archaeology on the Northwest Coast. Edited by Elizabeth A . Sobel, . A n n Trieu Gahr and Kenneth M . Ames. International Monographs in Prehistory, A n n Arbor. Martindale, A . 2006 Tsimshian House and Household through the Contact Period. In Household Archaeology on the Northwest Coast. Edited by Elizabeth A . Sobel, A n n Trieu Gahr and Kenneth M . Ames. International Monographs in Prehistory, A n n Arbor. 1999 The River of Mis t : Cultural Change in the Tsimshian Past. Unpublished PhD. Dissertation. University o f Toronto. Martindale, A . , and I. Jurakic 2004 Northern Tsimshian elderberry use in the late pre-contact to post-contact era. Canadian Journal of Archaeology 28(2):254-80. Martindale, A . and S. Marsden 2003 Defining the Midd le Period: (3500 B P to 150 B P ) in Tsimshian History through a comparison o f archaeological and oral records. BC Studies. No . 138. 52 Matson, R . G . 2003 Introduction: the Northwest Coast Perspective. In Emerging from the Mist: Studies in Northwest Coast Culture History. Edited by R . G . Matson, Gary Coupland and Quentin Mackie . U B C Press, Vancouver. Matson, R . G . and Gary Coupland 1995 The Prehistory of the Northwest Coast. 1st ed. V o l . 0, Academic Press, Inc., San Diego. McLaren , Duncan, Quentin Mackie , Daryl Fedje, Andrew Martindale and David Archer 2007 Relict Shorelines and Archaeological Prospection on the Continental Hinge of North Coastal Bri t ish Columbia. Unpublished Report. University o f Victoria . M o r i n , Jesse 2006 Non-domestic architecture in prehistoric complex hunter-gatherer communities: A n example from Keatley Creek on the Canadian plateau of Bri t ish Columbia. Unpublished M A Thesis. University o f Brit ish Columbia Ortner, S. B . 1984 Theory in Anthropology since the 60s. Comparative Studies in Society and History 26(1): 126-166. Pearsall, D . M . 2000 Paleoethnobotany: A Handbook of Procedures. 2nd ed. V o l . 0, Academic Press, San Diego. Pojar, J im and Andy Mackinnon 1994 Plant of Coastal British Columbia. Lone Pine Publishing. Turner, Nancy 2003 'Passing on the News': Women's Work, Traditional Knowledge and Plant Resource Management in Indigenous Societies o f North-western North America. In P. Howard (ed) Women and Plants: Gender Relations in Biodiversity Management and Conservation. Zed Books, London 133-149. Turner, Nancy and Helen Clif ton 2006 The Forest and the Seaweed: Gitga'at Seaweed, Traditional Ecological Knowledge, and Community Survival. In C . Menzies (ed.) Traditional Ecological Knowledge and Natural Resource Management. University of Nebraska Press, L incoln . Pgs. 65-86 Turner, Nancy and Trevor Lantz 2003 Traditional Phenological Knowledge o f Aboriginal Peoples in Brit ish Columbia. In Journal of Ethnobiology 23(2) 263-286 53 A p p e n d i x 1: F l o t a t i o n S t e p s Step Flotation 1 Collect materials samples randomly for each level 2 1 litre is measured 3 Manual flotation using bucket technique (Pearsall 1980) 4 Bucket is filled with water, siphoned and filled again releasing buoyancy material 5 During flotation, samples are screened into >.2mm and <2mm to ease processing 6 Once all material is removed, both heavy and light fraction are laid out to dry. 7 Dry material is bagged and labeled for processing 54 A p p e n d i x 2: D a t a b a s e DIP Site (T5) Unit Layer Level Type Length (mm) Width (mm) Thickness (mm) Count Weight (9) Notes Ground stone 107 12-1 B 117-127cm Artifact 76.1 13.4 6.4 10.5 artifact, groove at neck (maybe a pendant) Bone 188e 12-1 B artifact 23.5 5.5 3 1 0.1 unidentified fish, salmon 189d 12-1 B Faunal 1 0.1 vertebrae 188b 12-1 B Faunal 1 0.1 Fish 189e 12-1 B Faunal 5 0.1 Fish 188d 12-1 B Faunal 1 0.2 Mammalian tooth 188c 12-1 B Faunal 4 0.1 unidentified 189c 12-1 B Faunal 4 0.1 unidentified 189b 12-1 B Flake 13 0.1 quartz flake, clear glasslike 123 12-3 Beach Artifact 122.6 61.4 60.1 610.5 Cobble Chopper, found on beach 120a 12-3 C E Bone 29.8 7.4 3.5 0.17 Bone, appears unmodified Bone point, one 175mm Bone rough (hollowed out) end, 119 12-3 C F BUD artifact 64.4 11.2 6.0 4.07 point/flattish Narrow Bone 199mm Bone Point, one tip, one 114 12-3 C E/F BUD artifact 76.0 8.9 4.4 4.02 slightly flat end 160c 12-3 C D 9 Faunal - 1 0.5 Bird 170d 12-3 D D 4 Faunal - 1 0.1 Bird 161a 12-3 C D 9 Faunal - 1 0.1 Fish 163b 12-3 D C 3 Faunal - 1 0.1 Fish 170c 12-3 D D 4 Faunal - 1 0.1 Fish 172d 12-3 D E Faunal - 4 0.1 Fish 175b 12-3 D F Faunal - 16 0.1 fish 164b 12-3 D C 4 Faunal - 2 ' 0.1 fish, salmon 166a 12-3 D C 8 Faunal - 1 0.1 fish, salmon 171b 12-3 D E Faunal - 1 0.1 fish, salmon 172a 12-3 D E Faunal - 19 1.9 fish, salmon fish, salmon 162 12-3 D C 2 Faunal - 1 0.3 vertebrae fish, salmon 167 12-3 D D 1 Faunal - 2 0.1 vertebrae fish, salmon 173 12-3 D E Faunal - 3 0.3 vertebrae fish, salmon 120b 12-3 C E Faunal 7.4 6.1 4.9 1 0.10 vertebrae fish, salmon 120c 12-3 C E Faunal 8.2 7.2 5.2 1 0.19 vertebrae 55 120d 12-3 C E Faunal 7.4 6.5 5.6 1 0.16 fish, salmon vertebrae 120e 12-3 C E Faunal 7.7 7.6 4.8 1 .10 fish, salmon vertebrae 120f 12-3 C E Faunal 6.4 4.8 4.6 1 0.06 fish, salmon vertebrae 120g 12-3 C E Faunal 5.6 4.9 3.1 1 0.04 fish, salmon vertebrae 160a 12-3 C D 9 Faunal 8 0.8 fish, salmon vertebrae 161b 12-3 C D 9 Faunal 11 0.1 fish, salmon vertebrae 163a 12-3 D C 3 Faunal 2 0.1 fish, salmon vertebrae 165a 12-3 D C 6 Faunal 3 0.2 fish, salmon vertebrae 168a 12-3 D D 2 Faunal 8 0.2 fish, salmon vertebrae 169a 12-3 D D 2 Faunal 3 0.1 fish, salmon vertebrae 170a 12-3 D D 4 Faunal 9 0.4 fish, salmon vertebrae 171a 12-3 D E Faunal 17 2.8 fish, salmon vertebrae 174a 12-3 D F Faunal 20 1.8 fish, salmon vertebrae 175a 12-3 D F Faunal 8 0.1 fish, salmon vertebrae 160b 12-3 C D 9 Faunal - 1 0.2 human tooth 160d 12-3 C D 9 Faunal - 1 0.1 mammal 161c 12-3 C D 9 Faunal - 0.9 mammal 161d 12-3 C D 9 Faunal - 1 0.3 mammal 164a 12-3 D C 4 Faunal - 1 1.9 mammal 165b 12-3 D C 6 Faunal - 1 0.2 mammal 171c 12-3 D E Faunal - 2.6 mammal 172b 12-3 D E Faunal - 1 1.6 mammal 185 12-3 D E Faunal 1 1.3 mammal 166b 12-3 D C 8 Faunal - 1 0.1 mammal 168b 12-3 D D 2 Faunal - 1 0.3 mammal 170b 12-3 D D 4 Faunal - 1 3.2 shell, kyton 176 12-3 D H Faunal - 1 0.1 unidentified 120h 12-3 C E Faunal 11.8 2.2 2.3 1 0.02 unidentified 120i 12-3 C E Faunal 6.1 5.2 0.1 1 0 unidentified 161e 12-3 C D 9 Faunal - 1 0.1 unidentified 164c 12-3 D C 4 Faunal - 1 0.2 unidentified 166c 12-3 D C 8 Faunal - 1 0.1 unidentified 168c 12-3 D D 2 Faunal - 0.3 unidentified 169b 12-3 D D 2 Faunal - 1 0.5 unidentified 172c 12-3 D E Faunal - 0.6 unidentified 174b 12-3 D F Faunal - 1.4 unidentified 175c 12-3 D F Faunal - 1 0.1 unidentified 146d 12-3 D D 4 Flake 14 0.1 Flake Basalt 186 12-3 D F possible 1.3 "bone point thingy" 56 artifact broken since excavated (unknown if modified) 109 12-3 D possible artifact 844 43.4 13.2 63.52 Ground stone Bifacial, snapped, looks like retouched for further use after snapping, one real nice shaped side, one broken 144f 12-3 D E possible artifact 32 1.3 possible artifact broken, not point of percussion, basalt material 113 12-3 C 131mm BUD possible artifact 45.3 7.1 2.8 0.58 Possible Bone point 142g 12-3 D D 2 possible artifact 19.6 14.7 5.2 2.05 possible ground stone 142i 12-3 D D 2 possible artifact 25 2.57 Potential ground stone 139e 12-3 D D 3 possible artifact 38 9.5 potential ground stone triangular with flat edge on one side 138a 12-3 C D 9 possible artifact 21.0 14.8 5.8 3.71 Potential ground stone, bottom smooth, half cylindrical 142d 12-3 D D 2 possible artifact 36.8 5.2 7.5 2.51 Semi-Regular potential ground stone 142f 12-3 D D 2 possible artifact 24.8 15.0 5.5 2.81 semi-regular potential ground stone 138c 12-3 C D 9 possible artifact 51.5 16.5 6.3 7.44 Semi-regular, point blunted) potential ground stone 144c 12-3 D . E possible artifact 46.7 41.0 6.0 10.8 slate 142m 12-3 D D 2 possible flake 16 0.36 Possible flake 110e 12-3 C E possible flake 9.3 6.9 4.2 0.33 possible flake, small "triangle like" 137e 12-3 C D 9 possible flake 9.1 8.3 1.9 .20 Possible Flake, Triangular Quartzite 110a 12-3 C "E possible flake 35.2 26.3 10.2 9.55 Possible Quartzite Flake, 3 possible edges 139o 12-3 D D 3 possible flake 14 0.58 possible small triangular/irregular flake, Quartzite 150b 12-3 D H Possible flake 42 5.6 potential flake, slate material but 57 has questionable shape 110h 12-3 C E possible flake 25.5 12.2 7.1 2.87 Potential Modification Quartzite 139m 12-3 D D 3 possible flake 19 .30 potential small Quartzite Flake 142s 12-3 D D 2 possible flake 11 0.80 quartzite possible flake 139k 12-3 D D 3 possible flake 19 0.96 Quartzite semi triangular (Irregular) off colouration, potential flake 58 A p p e n d i x 3: H e a r t h S a m p l e R a w D a t a Site Level Bag Material Weight (g) Notes 12 1 107-117 25 Charcoal 68.9 From 2 litres 12 1 117-127 30 Charcoal 10.3 1 litre 12 1 127-137 2 of 4 charcoal 30.5 1 litre 12 1 97-107 9 Charcoal 42.6 From 2 litres 12 1 107-117 25 fauna 0.1 From 2 litres 12 1 117-127 30 fauna 0.1 1 litre 12 1 127-137 2 of 4 fauna 0.1 1 litre 12 1 107-117 25 modem plant 0.2 From 2 litres 12 1 117-127 30 modern plant 0.1 1 litre 12 1 127-137 2 of 4 modern plant 1.8 12 1 97-107 9 modern plant 0.9 From 2 litres 12 1 107-117 25 Rock/sand 286.2 From 2 litres 12 1 117-127 30 Rock/sand 55.7 1 litre 12 1 97-107 9 Rock/sand 579.2 From 2 litres 12 1 127-137 2 of 4 Rock/sand 110.7 1 litre 12 1 97-107 9 shell 0.1 From 2 litres 12 1 117-127 30 shell 0.1 1 litre 12 1 127-137 2 of 4 shell 16.2 1 litre 12 3 G 1 Charcoal 5.1 1 litre 12 3 G 2 charcoal 10 1 litre 12 3 G 3 Charcoal 9.8 1 litre 12 3 G 4 Charcoal 13.2 1 litre 12 3 G 5 Charcoal 13.7 From 2 litres 12 3 G 1 fauna 0.1 1 litre 12 3 G 2 fauna 0.1 1 litre 12 3 G 4 fauna 0.1 1 litre 12 3 G 5 fauna 0.1 From 2 litres 12 3 G 1 Rock/sand 64.2 1 litre 12 3 G 2 Rock/sand 102.9 1 litre 12 3 G 3 Rock/sand 46 1 litre 12 3 G 4 Rock/sand 182.5 1 litre 12 3 G 5 Rock/sand 265.5 From 2 litres 12 3 G 1 modern plant 0.6 1 litre 12 3 G 2 modern plant 3.6 1 litre 12 3 G 3 modern plant 0.2 1 litre 12 3 G 4 modern plant 0.2 1 litre 12 3 G 5 modern plant 0.3 From 2 litres 12 3 G Seeds 0.1 3 unidentified 12 3 G 1 shell 342.4 1 litre 12 3 G 2 shell 369.1 1 litre 12 3 G 3 shell 316.4 1 litre 12 3 G 4 shell 271.5 1 litre 12 3 G 5 shell 621.8 From 2 litres 59 A p p e n d i x 4: H e a r t h S a m p l e P r o p o r t i o n s Site Litres Material Total (g) % 12 3 6 Charcoal 51.8 2% 12 3 6 Fauna 0.4 0% 12 3 6 Rock/sand 661.1 25% 12 3 6 MP 4.7 0% 12 3 6 Seeds 0.1 0% 12 3 6 Shell 1921.2 73% Total (g) 2639.3 100% 12 1 6 Charcoal 152.3 13% 12 1 6 Fauna 0.3 0% 12 1 6 Modern Plant 3 0% 12 1 6 Rock/Sand 1031.8 86% 12 1 6 Shell 16.4 1% Total (g) 1203.8 60 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0101071/manifest

Comment

Related Items