An Archaeological Examination of House Architecture and Territoriality in the Salish Sea Region Over Five Millennia

Territory, Tenure, and Territoriality Among the Ancestral Coast Salish of SW British Columbia and NW Washington State

by Chris Springer

M.A., Simon Fraser University, 2009 B.A., Simon Fraser University, 2006

Thesis Submitted in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

in the Department of Archaeology Faculty of Environment

Copyright in this work rests with the author. Please ensure that any reproduction or re-use is done in accordance with the relevant national copyright legislation. Approval

Name: Chris Springer Degree: Doctor of Philosophy (Archaeology) Territory, Tenure, and Territoriality Among the Title: Ancestral Coast Salish of SW British Columbia and NW Washington State

Examining Committee: Chair: Jon Driver Professor Dana Lepofsky Senior Supervisor Professor Michael Blake Supervisor Professor Department of Anthropology University of British Columbia Ross Jamieson Supervisor Associate Professor Christina Giovas Internal Examiner Assistant Professor Elizabeth A. Sobel External Examiner Professor Department of Sociology and Anthropology Missouri State University

Date Defended/Approved: September 26, 2018

ii Abstract

Archaeological studies of territory, tenure, and territoriality seek to understand how past claims and access to land and resources were expressed across landscapes and through time. The foci of such studies include the spatial and temporal patterning of settlements, dwellings, conspicuous burials, monumental constructions, rock art, defensive features, and resources. In line with this research, this dissertation integrates ethnohistoric and archaeological data in three case studies that investigate the roles of house forms, the distribution of local and nonlocal obsidian, and the positioning of defensive networks in communicating territorial and tenurial interests among the ancestral Coast Salish of southwestern British Columbia and northwestern Washington state.

To understand how territorial and tenurial claims were expressed among the ancestral Coast Salish, the three studies consider the significance of the ethnohistoric Coast Salish social structure defined by bilateral kinship, group exogamy, and wide- ranging social networks in the communication of group interests. The first study supports the extant hypothesis of a regional move into large multifamily houses circa 2300 cal. BP. I hypothesize that this move was, in part, a consequence of regional population increases and its attendant territoriality and was facilitated by the structured flexibility of Coast Salish society and a pre-existing modular architecture that both reflected and reinforced the social structure. The distributions of local and nonlocal obsidian across the Salish Sea region are used in the second study to investigate the potential directionality and reach of ancestral social networks. I argue that these networks, developed from the practice of group exogamy, enabled the expression of tenurial claims as part of ongoing practices associated with gaining, maintaining, and legitimizing access to distant resources. Finally, the interrelationship of social networks and defensive networks among the ancestral Northern Coast Salish-Tla’amin are examined. I propose that these linked networks maximized defensibility at settlement and allied settlement scales in a form of defensive territoriality that served to communicate territorial and tenurial interests during periods of conflict.

Keywords: Coast Salish; social networks; defensive networks; tenure; territoriality

iii Acknowledgements

This dissertation would not have been possible without the direct and indirect help of many people. First, I thank my mom, Evelyn whose constant support and encouragement has been an endless source of strength throughout my post-secondary career at SFU. I will forever be grateful for your editorial comments on drafts of chapters and getting me to tone down academes in favour of comprehensible language.

I am forever grateful to my senior advisor, Dr. Dana Lepofsky. Apart from her limitless help and support during the field work, analysis, writing, and presentation stages of my research, she kindly involved me in other areas of her work that have broadened my understanding of archaeology in general and B.C. archaeology specifically. I also thank the other two members of my committee, Dr. Michael Blake and Dr. Ross Jamieson. Their comments and suggestions throughout the writing stage of my dissertation helped me consider my work from other perspectives than the one I had focused on and with which I had become comfortable.

The field work for my research was made possible by the generosity of the Tla’amin community, che che hah tan nah pitch. Everyone in the community were such gracious hosts for the three SFU field schools and two SFU/University of Saskatchewan field schools that stayed at Tla’amin during the 2009-2011 and 2012-2013 field seasons, respectively. In particular, che che hach Hegus Clint Williams, Michelle Washington, Murray and Nancy Mitchell, Jason Francis, Betty Wilson, Lisa Wilson, John Louie, and Erik Blaney for all their help and support. An extra thanks to Murray and Jason for being out there in the rain and dirt with me.

A special thanks also goes out to Georgia Combes who gave freely of her time and vast knowledge of the area both on land and water; without her boating skills, I would certainly be lost at sea. She also generously passed on her knowledge of the ancestral Tla’amin landscape over the five years of my field work. My research also benefited greatly from her skill as a photographer. Georgia’s aerial photographs significantly reduced survey time by giving a bird’s eye view of the numerous mass harvesting features found in the intertidal zone throughout Tla’amin territory.

The field school students also deserve special recognition for all the hard work they put in to make the field seasons a success. They are: Craig Barnes, Nyra Chalmer,

iv Fred Foster, Guillermo Garcia, Allison Hill, Simon Lloyd-Price, Vanessa Medland, Aaron Racicot, Anna Stewart, Mike Szepvolgyi, Andrea Unrau, Diana Wasylik, and Rachel White (2009 SFU Filed School); Alisha Gauvreau, Rhory Gillies, Tyrone Hamilton, Katie Hausch, Tanya Hunt, Buffy Johnson, Nikki Lloyd-Gervais, Sean Matthews, Carleen Novak, Camille Reynolds, Nicole Slade, and Amandah Van Merlin (2010 SFU Field School); Jan Anderson, Aleesha Baakelund, Sarah Balabanov, Amelia Barker, Clayton Crawford, Corey Hartley, Lap Kwan-Tang, Jacob Liddy, Katie Lum, Mark Powell, Jacqueline Sio, and Kasia Zimmerman (2011 SFU Field School); Mika Blundell, Allan Downey, Scott Dumonceaux, Martin Hoffman, Ilya Lipin, Taegan McFarlane, Tylor Richards, Elyse Thiessen, and Kasia Zimmerman (2012 SFU-University of Saskatchewan Field School); and Teresa Baker, Carrie Helter, Isabelle Maurice- Hammond, Adrienne Marino, Lindsey Moore, Colin Osmond, Julie Oltmann Plesner, Jennifer Walkus, and Shelly Wright (2013 SFU-University of Saskatchewan Field School). Also, thank you Drs. Dana Lepofsky, John Welch, Keith Carlson, and Bob Muir for supervising the field schools, and Morgan Ritchie and Megan Caldwell for all your help in the field.

The work study students who helped me sort through the seemingly endless bags of shell midden were also a major factor in me finishing just under the wire. I would still be sorting if not for you: Isabelle Maurice, Teresa Matheson, Sara Bucci, Andrea Mullan, Anna Glass, Megan Poland, Cathi Williams, Jan Anderson, Tessa Fryer, and Hilary Pennock.

I also thank Dawn Ainsley, Brett Beaulieu, Sarah Beaulieu, Nigel Boeur, Nikki Bose, Cynthia Boyer, Aileen Brennan, Terence Clark, Robyn Ewing, Jay Herbert, Rich Hutchings, Marina LaSalle, Peter Locher, Natasha Lyons, Luseadra McKerracher, Peter Merchant, Jessica Nelson, Misha Puckett, Heather Robertson, and Anke Weber for volunteering their time to help with field work during various years. Thank you, Shannon Wood and Peter Locher, for giving so much of your time to making sure the field schools ran smoothly and for helping with any and all lab issues that I ran into. I also want to thank Doug Brown and Nicole Oakes of Brown & Oakes Archaeology for allowing me to include the sourcing results for a piece of obsidian from the Pitt River Site (DhRl-21), and Dr. Rudy Reimer for analyzing a number of pieces of obsidian from Tla’amin territory, free of charge. The results were critical to my second case study presented in this dissertation.

v Finally, an extra special thank you goes out to my partner Heather Kendall. Your love, seemingly boundless patience, and your emotional and, at times, financial support have been a source of strength and comfort for me over the last eight years. Quite simply, this project of mine would not have been possible without you. I am also forever grateful for the times you volunteered in the field, freely giving your knowledge of archaeological practice in aid of my research and to the field school students who were present.

My research was funded by a Social Sciences and Humanities Council of Canada Graduate Doctoral Scholarship (Joseph-Armand Bombardier Canada Graduate Scholarship) #767-2010-1944, a Pacific Century Graduate Scholarship, a Provost Prize of Distinction, and a President’s Ph.D. Scholarship.

vi Table of Contents

Approval ...... ii Abstract ...... iii Acknowledgements ...... iv Table of Contents ...... vii List of Tables ...... ix List of Figures...... x

Chapter 1. Introduction ...... 1 1.1. Territory, Tenure, and Territoriality ...... 3 1.2. Practice ...... 4 1.3. Dissertation Outline ...... 6

Chapter 2. An Archaeological Examination of House Architecture and Territoriality in the Salish Sea Region over Five Millennia ...... 9 2.1. Territory, Tenure, Territoriality among the Coast Salish ...... 13 2.2. The Roles of Kinship and Social Relations in Territory and Tenure among the Ethnographic Coast Salish ...... 14 2.3. Ethnographic Review of Coastal and Lower Fraser Coast Salish Houses ...... 15 2.3.1. In-ground Houses ...... 15 2.3.2. Above-ground Houses ...... 29 2.4. Methods ...... 32 2.5. Results ...... 33 2.5.1. Temporal Variation of both House Forms (Ethnographic and Archaeological Data Combined)...... 33 2.5.2. Coastal and Lower Fraser Houses Pre-2300 versus Post-2300 cal. BP...... 34 2.5.3. Coastal and Lower Fraser In-ground Houses through Time ...... 37 2.6. Discussion ...... 39 2.7. Conclusion...... 41

Chapter 3. Obsidian in the Salish Sea: An Archaeological Examination of Ancestral Coast Salish Social Networks in SW British Columbia and NW Washington State ...... 42 3.1. Territory, Tenure, and Territoriality in Ancestral Coast Salish Social Networks .... 46 3.2. Methods ...... 47 3.3. Results ...... 50 3.3.1. Spatial Distribution of Sourced Toolstone within the Salish Sea Region ...... 66 3.3.2. Spatial Distribution of Sourced Toolstone by Cultural Group within the Salish Sea Region ...... 70 3.3.3. Temporal Distribution ...... 74 3.4. Discussion ...... 80 3.5. Conclusion...... 83

vii Chapter 4. Conflict and Territoriality: An Archaeological Study of Ancestral Northern Coast Salish-Tla’amin Defensiveness in the Salish Sea Region of SW British Columbia ...... 84 4.1. Territory, Tenure, and Territoriality Among the Northern Coast Salish ...... 86 4.2. The Tla’amin-Northern Coast Salish and Their Neighbours ...... 88 4.3. Methods ...... 90 4.3.1. Site Selection ...... 90 4.3.2. Open Shoreline Site Complexes ...... 104 Emmonds Beach (Figures 21-22; Table 10, ID 20) and Keays Bay (Figures 21-22; Table 10, ID 52) ...... 104 Bliss Landing (Figures 21-22; Table 10, ID 37) ...... 108 4.3.3. Inlet Site Complexes ...... 108 Okeover (Figures 21-22; Table 10, ID 39) ...... 108 Portage Cove (Figures 21-22; Table 10, ID 19) ...... 108 4.3.4. Cove Site Complexes ...... 109 Laura Cove (Figures 21-22; Table 10, ID 53), Melanie Cove (Figures 21-22; Table 10, ID 54), and Roffey Cove (Figures 21-22; Table 10, ID 48) ...... 109 4.3.5. Viewshed Analyses ...... 110 Defense Index ...... 114 4.4. Results ...... 115 4.4.1. Open Shoreline Site Complexes ...... 118 Emmonds ...... 118 Bliss Landing ...... 120 4.4.2. Inlet Site Complexes ...... 120 Okeover ...... 120 Portage Cove ...... 122 4.4.3. Cove Site Complexes ...... 125 Prideaux Heaven ...... 125 4.5. Discussion ...... 127

Chapter 5. Conclusion ...... 131

References ...... 136

Appendix A...... 159 Tla’amin Place Names ...... 159

Appendix B ...... 163 Statistical Comparisons of House Forms ...... 163

Appendix C ...... 165 Results of Obsidian Elemental Analyses by Northwest Research Obsidian Studies Laboratory (NWROSL) and the Department of Archaeology pXRF Laboratory at SFU 165

viii List of Tables

Table 1: Ethnographic and archaeological data sets of houses from the Coastal and Lower Fraser areas of the Salish Sea region...... 16 Table 2: In-ground houses recorded in the ethnographic and archaeological records for the coastal areas of the Salish Sea region...... 17 Table 3: Temporal and area data for in-ground and above-ground houses identified in the Salish Sea region...... 19 Table 4: Sites and associated obsidian artifacts and toolstone sources (see Tables 5-7 for names of Washington, Oregon, and Idaho toolstone sources). Salish Sea sites: 38-82, 84-113, 118-126, 138, 142-152 (see Figures 10-13 for all site locations...... 51 Table 5: Toolstone sources with numbers of correlated artifacts and sites, and numbers of dated artifacts...... 59 Table 6: Sites with obsidian artifacts from Washington and Idaho toolstone sources. .... 60 Table 7: Obsidian artifacts from Oregon Toolstone sources...... 61 Table 8: Chronometric and inferred dates associated with obsidian artifacts...... 64 Table 9: Descriptions of archaeological site types1 ...... 92 Table 10: Tla’amin habitation sites in ethnohistoric and archaeological records, arranged from largest to smallest by area...... 93 Table 11: Tla’amin lookouts, trench embankments, and redoubts in ethnohistoric and archaeological records, arranged by type from largest to smallest by area...... 99 Table 12: Dated sites from sample used in defense index and viewshed analyses and associated ancillary sites...... 105 Table 13: Summary of defensiveness of site complexes in three sub-areas...... 107 Table 14 Summary of viewing radius (VR) and area calculations for viewshed analyses...... 113 Table 15: Summary of variables and defense index (DI) calculations for sample of sites used in viewshed analyses...... 116

ix List of Figures

Figure 1: The location of the Salish Sea region in SW B.C. and NW Washington state. The dotted line represents the approximate boundary of Coast Salish languages (Suttles 1990). (Map generated from the Environmental Systems Research Institute (ESRI) basemap catalogue using ArcMap 10.3)...... 2 Figure 2: Locations of in-ground and above-ground house sites discussed in the paper. Map 2a shows the distribution of Coast sites and Map 2b shows the distribution of Lower Fraser sites. ID numbers provide additional information in Tables 1-3. The dashed lines in the inset and Map 2a are approximate representations of linguistic boundaries of Northern, Central, Southern, and Southwestern Coast Salish peoples (Suttles 1990) (map generated from the ESRI basemap catalogue using ArcMap 10.3)...... 11 Figure 3: 3a Interpretation of the Coast Salish-Musqueam Nation winter home for wealthy families (Barnett 1944:268; ID 16 in Tables 1 and 2); 3b interpretation of the Coast Salish-Klahoose Nation winter homes used in the Toba River Valley (Barnett 1935-1936:6:84; ID 1 in Tables 2 and 3; drawings by Chris Springer)...... 27 Figure 4: 4a Interpretation of plankhouse-subterranean refuge described for the Coast Salish-Klahoose, Squamish, Sechelt, and Tla’amin Nations; 4b interpretation of subterranean refuge described by Barnett (1955:49-50), without associated plankhouse (Barnett 1955:49-50; ID 2-7, 13, and 15 in Tables 2 and 3); 4c interpretation of Coast Salish-Sts’ailes Nation pithouse from the Lower Fraser (adapted from Springer and Lepofsky 2011:35, Figure 6) (drawings by Chris Springer)...... 28 Figure 5: Generic interpretations of most common plankhouse forms in the Salish Sea region: 5a shed-roof; 5b gable-roof (drawings by Chris Springer, adapted from figures and descriptions in Barnett [1955:35-58], Boas [1890:11-12], Elmendorf [1992:154-165], Suttles [1990:7], and Schaepe et al. [2001:40- 42])...... 30 Figure 6: Upper image – Central Coast Salish village in the Coastal area of Salish Sea region (Alden 1858); lower image – Central Coast Salish village in the Lower Fraser area of the Salish Sea region. Note the high visibility of the large above-ground houses in both villages compared with the more restricted visibility of the in-ground houses (center mid-ground) in the Lower Fraser village. (Alden 1858 Image PDP02144 courtesy of the Royal British Columbia Museum and Archives)...... 31 Figure 7: High-low graphs of in-ground and above-ground houses through time identified in Coast (7a) and Lower Fraser (7b) areas. Bars represent calibrated 2σ range or inferred ranges based on diagnostic and/or historic artifacts. ID numbers provide additional information in Tables 2 and 3. These data represent the presence/absence of in-ground and above-ground houses...... 35 Figure 8: Probability distributions of in-ground and above-ground houses through time based on 2σ calibration ranges of the Coast (8a) and Lower Fraser (8b) houses for which complete chronometric data were available. ID numbers provide additional information in Table 3...... 36

x Figure 9: Area (m2) of in-ground and above-ground houses in coastal territories and in the Lower Fraser plotted by time. Represented are only those houses for which both temporal (chronometric or inferred median dates) and area data were available (9a Coastal, N=46 houses/35 sites; 9b Lower Fraser, N=49 houses/20 sites). The outliers, Old Man House (ID 106, 2400 m2) and the Matsqui House (ID 93, 4200 m2), were not included in the Coastal and Lower Fraser datasets, respectively...... 38 Figure 10: General locations of all archaeological sites (circles) and obsidian toolstone sources (stars) referred to in this paper. Arrows show hypothesized routes through which obsidian toolstone was moved into the Salish Sea region via social networks. Major trading hubs referred to in this paper are marked with an “x”. The dotted line represents the approximate boundary of Coast Salish languages (Suttles 1990) (map generated from the ESRI basemap catalogue using ArcMap 10.3)...... 45 Figure 11: General locations of archaeological sites (circles) with obsidian artifacts sourced to Mount Edziza (11a), Ilgachuz (11b), Anahim Peak (11c), and MacKenzie (11d). ID numbers provide additional information in Table 4 (archaeological sites) and Table 5 (toolstone sources). The dotted line represents the approximate boundary of Coast Salish languages (maps generated from ESRI basemap catalogue using ArcMap 10.3)...... 67 Figure 12: General locations of archaeological sites (circles) with obsidian artifacts sourced to Central Coast Unknown B (12a), Kingcome/UTA (12b), and Nch’kay (12c). ID numbers provide additional information in Table 4 (archaeological sites) and Table 5 (toolstone sources). The dotted line represents the approximate boundary of Coast Salish languages (maps generated from ESRI basemap catalogue using ArcMap 10.3)...... 68 Figure 13: General locations of archaeological sites (circles) with obsidian artifacts sourced to toolstone deposits located in Washington State (13a), and Oregon and Idaho States (13b). ID numbers provide additional information in Tables 4, 6, and 7 (archaeological sites) and Table 5 (toolstone sources). The dotted line represents the approximate boundary of Coast Salish languages (maps generated from ESRI basemap catalogue using ArcMap 10...... 69 Figure 14: General locations of archaeological sites with obsidian artifacts sourced to toolstone deposits south of the Salish Sea region (crosses), north of the Salish Sea region (black circles), north and south of the Salish Sea region (triangles), north and south of the Salish Sea region, and Nch’kay (stars), and Nch’kay (open circles). The ID numbers provide additional information in Table 4. The dotted lines are approximate representations of linguistic boundaries of Northern (NCS), Central (CCS), Southern (SCS), and Southwestern (SWCS) Coast Salish peoples (Suttles 1990) (map generated from ESRI basemap catalogue using ArcMap 10.3). .... 72 Figure 15: Percent obsidian abundance by source. Sources organized from north (Left) to south along the x-axes; four Coast Salish areas organized north (Top) to south on the y-axis. The graphs illustrate deviations from distance- decay null model. Numbers on top of bars represent total counts...... 73 Figure 16: General locations of archaeological sites (circles) with obsidian artifacts sourced to represented toolstone sources (stars), dating from 9,000-8,000 years BP (16a) and 10,000-9,000 years BP (16b). ID numbers provide

xi additional information in Tables 4 and 8 (archaeological sites) and Table 5 (toolstone sources). The dotted line represents the approximate boundary of Coast Salish languages (maps generated from ESRI basemap catalogue using ArcMap 10.3)...... 76 Figure 17: General locations of archaeological sites (circles) with obsidian artifacts sourced to represented toolstone sources (stars), dating from 5,000-4,000 years BP (17a), 6,000-5,000 years BP (17b), 7,000-6,000 years BP (17c), and 8,000-7,000 years BP (17d). ID numbers provide additional information in Tables 4 and 8 (archaeological sites) and Table 5 (toolstone sources). The dotted line represents the approximate boundary of Coast Salish languages (maps generated from ESRI basemap catalogue using ArcMap 10.3)...... 77 Figure 18: General locations of archaeological sites (circles) with obsidian artifacts sourced to represented toolstone sources (stars), dating from 1,000-100 years BP (18a), 2,000-1,000 years BP (18b), 3,000-2,000 years BP (8c), and 4,000-3,000 years BP (18d). ID numbers provide additional information in Tables 4 and 8 (archaeological sites) and Table 5 (toolstone sources). The dotted line represents the approximate boundary of Coast Salish languages (maps generated from ESRI basemap catalogue using ArcMap 10.3)...... 78 Figure 19: Probability distribution of radiocarbon dated archaeological sites with obsidian artifacts that have been sourced to Nch’kay, Kingcome, and CCUB. Probability distributions based on calibrated 2σ range. Bars at bottom of figure represent approximate timelines for LIA and sociopolitical unrest that may have affected access to toolstone deposits. ID numbers (y-axis) provide additional information in Tables 4 and 8. Figure generated using Calib 7.1 (Stuiver et al. 2017) and Adobe Illustrator...... 79 Figure 20: General locations of archaeological sites (crosses) in the Salish Sea region with obsidian artifacts sourced to toolstone deposits located in Oregon State. Historic Squamish, Musqueam, Nooksack, Lummi, and Twana marriage patterns derived from Kennedy (2007). Arrows show hypothesized routes through which obsidian toolstone was moved into the Salish Sea region via social networks formed through affinal relationships and kinship. ID numbers provide additional information in Tables 4 and 7 (map generated from ESRI basemap catalogue using ArcMap 10.3). .... 82 Figure 21: General locations of places, landforms, and bodies of water referred to throughout the text. Outer dotted line represents the approximate extent of Tla’amin traditional territory. Map generated from the ESRI basemap catalogue using ArcMap 10.3...... 86 Figure 22: General locations of archaeological sites referred to throughout the text. The ID numbers provide additional information in Tables 10 and 11. Map generated from ESRI basemap catalogue using ArcMap 10.3...... 91 Figure 23: “Deserted Indian Village” by William Alexander (1798). Thought to be of the Roffey Cove settlement in Prideaux Haven. (Used with permssion from the copyright holder: The Newberry Library, Chicago. Call # VAULT oversize Ayer Art Alexander, Drawing no. 17)...... 111 Figure 24: Combined viewshed of all sites. Map on the left generated from ESRI basemap catalogue using ArcMap 10.3. The ID numbers provide

xii additional information in Tables 10-12, 14-15. Viewshed map on right generated from mosaicked digital elevation model (DEM) using viewshed tool in ArcMap 10.3...... 117 Figure 25: Map A shows combined viewsheds from the Emmonds trench embankment (white), settlement (light grey), and lookouts (medium grey). Map B shows viewsheds from canoes at the points that they are visible from the various components of the Emmonds settlement. The ID numbers provide additional information in Tables 10-12, 14-15. The viewshed maps were generated from a mosaicked DEM using the viewshed tool in ArcMap 10.3...... 119 Figure 26: Map A shows the viewshed from the Bliss settlement. Map B shows the viewsheds from canoes at the point that they are visible from the Bliss settlement. The ID numbers provide additional information in Tables 10- 12, 14-15. The viewshed maps were generated from a mosaicked DEM using the viewshed tool in ArcMap 10.3...... 121 Figure 27: Map A shows the combined viewsheds from the Okeover settlement (white) and lookouts (light grey). Map B shows the viewshed from the canoe at the point it is visible from the Okeover settlement and lookouts. The ID numbers provide additional information in Tables 10-12, 14-15. The viewshed maps were generated from a mosaicked DEM using the viewshed tool in ArcMap 10.3...... 123 Figure 28: Map A shows the combined viewsheds from the Portage Cove settlement and associated lookout. Map B shows the viewsheds from the canoes at the point that they are visible from the settlement or lookout. The ID numbers provide additional information in Tables 10-12, 14-15. The viewshed maps were generated from a mosaicked DEM using the viewshed tool in ArcMap 10.3...... 124 Figure 29 Map A shows the combined viewsheds from the Cove settlements (white) and lookouts (light grey). Map B shows the viewsheds from the canoes at the point they are visible from the Cove settlements and lookouts. The ID numbers provide additional information in Supplemental Tables 10-12, 14-16. The viewshed maps were generated from a mosaicked DEM using the viewshed tool in ArcMap 10.3...... 126

xiii Chapter 1.

Introduction

Archaeological studies of territory, tenure, and territoriality seek to understand how past claims and access to land and resources were expressed across space and through time (Zedeño 1997, 2008). Among ancestral non-state societies, studies of the spatial and temporal distributions of built environments provide some of the more fruitful avenues of inquiry into territorial and tenurial interests. The foci of such studies include the placement of settlements, dwellings, and conspicuous burials (Charles 2010; Earle 2000; Fitzhugh 2002; Lepofsky et al. 2009; Mathews 2014), monumental constructions (Grier 2014; Grier et al. 2017; Klokler 2017; Widmer 2004), settings for rock art (David 2002; Earle 2000; Gunn 2011; Sepúlveda et al. 2017), and the positioning of defensive features (Angelbeck 2009; Baker 2003; Field 2005; Schaepe 2006). However, territorial and tenurial claims can also be made in the abstract through socioeconomic relationships that extend beyond the local. Although not archaeologically visible, one way these claims can be inferred is through the presence of nonlocal resources (e.g., Blake 2004; Morin 2012; Springer et al. 2018). Such resources reflect socioeconomic relationships that facilitated their movement and for the claims of access made manifest in the establishment, maintenance, and ongoing negotiation of those relationships.

In this dissertation, I integrate ethnohistoric and archaeological data in three case studies to investigate the relationships between built environments, social structure, artifact distributions, and expressions of territory, tenure, and territoriality among the ancestral Coast Salish of the Salish Sea region of southwestern British Columbia (B.C.) and northwestern Washington state (Figure 1). The first study considers how in-ground

2 and above-ground house forms reflected and influenced ancestral Coast Salish assertions of territory and tenure over the last 5,000 years. Specifically, the general move from in-ground to above-ground architecture circa 2300 years BP is examined as a potential response to increased regional territoriality. The second study uses the spatial and temporal distributions of sourced obsidian artifacts across the Salish Sea region as a means of exploring abstract notions of territory and tenure realized through social networks. The final study employs visibility analyses and quantitative measures ofdefense in an investigation of ancestral Northern Coast Salish territoriality as expressed through defensive networks. The three studies apply a landscape approach emphasizing the concepts of territory, tenure, and territoriality (Earle 2000; Ingold 1986; Zedeño 1997, 2008) from a practice-based perspective (Bourdieu 1977, 1989, 2005; Hillier and Rooksby 2005). The work is aligned with research that recognizes a range of social, economic, and political behaviours within and between non-state societies including the modification and management of their environments as expressions of territory, ownership, and control (Baker 2003; Eerkens 1999; Field 2005; Hodgetts 2010; Turner and Jones 2000; Turner et al. 2005; Zedeño 1997).

1.1. Territory, Tenure, and Territoriality

The interrelated concepts of territory, tenure, and territoriality are used in this dissertation as interpretive tools in aid of understanding social and physical connections to land and how the human-land relationship may have been articulated in the past (Ingold 1986). Physical and abstract expressions of territorial and tenurial claims, and acts of territoriality are examined in the three case studies. How the concepts are employed is explained in the relevant chapters. General definitions are given here.

Territory refers to nested but permeable spaces with which an individual or group identifies through some combination of spiritual, familial, affinal, and economic relationships. Thus, territories represent both physical entities (natural and/or culturally modified places) and abstract entities (e.g., hunting ranges, trade relations, water routes, and sociopolitical connections). These entities shift in extent and significance in concert with changing kinship and other socioeconomic relationships (e.g., Ingold 1986:138).

Tenure defines how these socioeconomic relationships are communicated within a continuously evolving social sphere (Ingold 1986: 130, 136-137). Fundamental

3 characteristics of tenure include the public legitimation and recognition of access and rights to resources at local and extra-local scales, the generational transfer of prerogatives to resources, and the knowledge and social protocols that are embedded in tenurial relations. As part of the public legitimizing process, tenurial claims weave together past and future socioeconomic relations and practices within a negotiated present.

Territoriality is the operational corollary of expressing, maintaining, or redefining territorial and tenurial claims. It encompasses the range of behaviours that serve to communicate territorial and tenurial claims in both the physical and the abstract. These behaviours include occupation, marking, naming, monitoring of access, trade, and, in some cases, violent conflict (Earle 2000:43; Ingold 1986:146-147; Keeley 1996:55-58; Taçon 2008:224; Turner et al. 2005:151-154; Zedeño 2008:211). Territoriality is best inferred archaeologically from the act of marking spaces such that they become meaningful places to those who have claimed the space (e.g., David 2002; Taçon 2008; Zedeño and Bowser 2009). Marking would have specific meaning for members of a related group (this place belongs to “x”) and general meaning for people external to the group (this place belongs to/is used by someone other than ourselves).

1.2. Practice

I use practice theory (Bourdieu 1977, 1989, 2005) to examine how acts of territoriality were used to express ancestral Coast Salish territorial and tenurial claims both spatially and temporally through physical (e.g., modified landscapes, houses) and social structures (e.g., kinship, social networks). From a practice perspective, physical (e.g., houses) and social (e.g., social networks) structures are entangled in an iterative relationship such that built environments both reflect and mediate the formal and informal social practices associated with those spaces (Blanton 1994:9-20; Bourdieu 1977:89-91). Furthermore, physical and social structures function as the medium and outcome of social practices and, as a result, are subject to redefinition or complete change as a consequence of generational, political, or socioeconomic shifts (Bourdieu 1977:78; see also Rapoport 1986:159). In the context of this study, changes in the patterning of built environments and their component parts, and in the spatial and temporal distributions of obsidian toolstone are the archaeological proxies of continuity

4 and change in social practices and how those practices were reflected in the physical and social structures of the ancestral Coast Salish.

Social structure as a malleable framework follows from Bourdieu’s concept of habitus—the suite of individual or group character traits learned through social interaction in structured yet changeable social fields (Bourdieu 1977:77-93, 2005:43, 46- 47; Hillier 2005:164; Jones 1999:226). Because it is learned, individual or group habitus has the capacity to transform with unfolding events, but within limits (Bourdieu 2005:47). Change can occur at a slow, incremental pace (e.g., generationally) or in rapidly shifting circumstances (e.g., violent conflict, large-scale natural disaster, or massive population decline due to war or epidemics). In any case, habitus is how knowledge is currently used and expressed; it “is a dynamic field of behavior, of position-taking where individuals inherit the parameters of a given situation and modify them into a new situation” (Leach 2005:298). The nature of habitus as a set of character traits used to improvise on and redefine cultural norms through practice is realized in the linked concepts of field and capital.

Fields represent the cultural contexts within which the traits that structure practice are acquired and expressed. However, fields only exist to the extent that social actors possess the traits, knowledge, and worldview necessary to constitute or alter fields as culturally meaningful spaces (Bourdieu 1990:67-68, 2005:47; Hillier and Rooksby 2005:22-24). Among the Coast Salish, houses (in-ground and above-ground) were important structured fields comprised of objectified cultural norms (i.e., relations of power, gender, and age that framed all activities from the mundane to the extraordinary) that were incorporated as character traits through social interaction within those spaces (e.g., Coupland et al. 2009). These traits were, in turn, expressed as subjective practices that served to both reify and alter the cultural norms that structured and gave meaning to those spaces and the actions that took place within them.

The practices that both arise from and define fields are generally geared toward the expression and/or acquisition of various forms of capital that give substance and meaning to social practices. Bourdieu (1989) delineates capital into three interrelated categories: 1) economic (material wealth and its attendant power); 2) social (the various benefits derived from social networks and connections at local and extra-local scales); and 3) cultural (the accumulation of culturally recognized manners, credentials,

5 knowledge, and skills). He also identifies symbolic capital, which refers to “the form that the various species of capital assume when they are perceived and recognized as legitimate” (Bourdieu 1989:17). The concept of symbolic capital is particularly germane to this research with respect to the formation and expression of territorial and tenurial claims among the ancestral Coast Salish. Legitimacy, as assessed through oratory, performance, witnessing, and gifting at cultural events was fundamental to clarifying rightful ownership and access, social position, and the valid generational transmission of tangible and intangible wealth:

The mere possession of wealth was not sufficient to give a man prestige. In order to acquire that, he was obliged to make a public assertion of every fact or event which contributed to an advance or change in his social position. Such an assertion always had to be made before formally invited guests from outside his own extended family, who listened to his announcements and vouched for his claims, and to whom he, in turn, distributed presents in the form of blankets, skins, planks, food, etc. To assume a family name, which carried with it a cluster of rights and prerogatives, to commemorate a change in status growing out of a life crisis, or to publicize any event having a bearing on social status demanded a public distribution of goods. Such were the immediate reasons for giving away wealth, and every distribution was in effect an assertion or a reassertion of some claim to distinction on behalf of the donor or some member of his family. No one could raise a house or grave post, or be married, or name his child, and expect to be taken seriously if he did not “call the people” as witness. And to be called meant that the invited guest was to receive a gift or at least a portion of food (Barnett 1955:253).

In sum, economic capital and cultural capital inhere in the local and extra-local social networks (social capital) established, maintained, and expressed through culturally recognized roles and practices (i.e., affinal ties, leadership, and protocols of access, obligation, and generational transfer). These roles and practices then convert to symbolic capital in social fields where they are legitimated in witnessed events (Angelbeck 2009:129-136; Barnett 1955:142, 263; Suttles 1987:15-25; Washington 2004:587-590).

1.3. Dissertation Outline

The following three chapters form the main body of the dissertation. Chapter Two presents the case study on ancestral Coast Salish residential architecture. Chapter Three looks at the spatial and temporal distribution of obsidian toolstone across the

6 Salish Sea region. Chapter Four examines Northern Coast Salish defensive strategies. The case studies were first written as coauthored papers for submission to peer reviewed journals. Chapters Two (coauthors Dana Lepofsky, Bill Angelbeck, and Michael Blake) has not yet been submitted but Chapters Three (coauthors Dana Lepofsky and Michael Blake) and Four (coauthor Dana Lepofsky) have been submitted. The former has been published in the Journal of Anthropological Archaeology (Springer et al. 2018) and the latter has been accepted for publication in the Journal of Island and Coastal Archaeology. I have chosen to use the plural pronoun and adjectives “we” and “our,” in the text to reflect the collaborative nature of these works.

Chapter 2 examines the potential role of large in-ground and above-ground houses in the symbolic communication of household territorial and tenurial claims by comparing the changing significance of these two architectural styles as social fields and venues of symbolic communication from 5500 to 100 years BP. The results confirm the extant hypothesis of aggregation into large multifamily dwellings circa 2300 cal. BP. This aggregation into large, highly visible, multi-family dwellings may have been a response to a rise in regional population and territoriality. The results of the study also suggest that the adoption of large houses circa 2300 years BP provided new social fields and forms of economic capital for the conflict between centralizing and decentralizing power interests inherent in Coast Salish social structure.

The second study, Chapter 3, explores the spatial and temporal distributions of obsidian toolstone within the Salish Sea region. These distributions are used as a means of examining the relationship between ancestral Coast Salish social networks and abstract notions of territory and tenure. The results suggest that the presence of nonlocal obsidian is explained by down-the-line trade along social networks that extended access beyond the local. In addition, these networks enabled the expression of territorial and tenurial claims as part of ongoing practices associated with gaining, maintaining, and legitimizing access to distant resources. The results also suggest that the lower Fraser Valley was an important corridor for much of the Oregon toolstone to reach other Central and Northern Coast Salish communities.

In Chapter 4, the final study examines the defensive strategies of the ancestral Tla’amin-Northern Coast Salish as a form of networked territoriality. Coast Salish ethnohistory describes how various locations associated with settlements were used for

7 defense within the Salish Sea region. During times of conflict, these linked places formed defensive networks that functioned to maximize defensibility at both the settlement and allied settlement scales. Using visibility analyses in a Geographic Information System (GIS) platform and an index of site defensiveness (Martindale and Supernant 2009), the spatial and temporal distributions of defensive networks in geographically distinct sub-areas of Tla’amin territory are analyzed. The results indicate that defensive strategies similar to those used by other Coast Salish populations were in place in Tla’amin territory from approximately 900 to 100 years BP. This in turn suggests the longevity of territorial and tenurial claims expressed through defensive territoriality, as noted in the regional ethnohistoric record.

8 Chapter 2. An Archaeological Examination of House Architecture and Territoriality in the Salish Sea Region over Five Millennia

Throughout history people have used household architecture and other dimensions of built environments to articulate and affirm connections to place. This architectural expression of place-based identity has been especially important for societies that rely primarily on oral history, oration, witnessing, public ceremonies, and material symbols to convey information about intra- and inter-group relations. Among Northwest Coast peoples, these types of communication and associated events, would have been given additional significance and meaning by both architectural forms and modified landscapes within which they took place (e.g., Coupland 2006; Coupland et al. 2009; Grier 2006).

The archaeological remains of houses on the Northwest Coast are examples of constructed spaces that provide one of the best links to ancestral social worlds (e.g.,

Coupland 2013; Lepofsky et al. 2009; Sobel et al. 2006). Houses objectify social institutions, groups, or roles of household members, and are an ideal medium for communicating both literal and symbolic connections to place. On the one hand, the physical structure communicates the household’s presence and potential claim to the surrounding area. On the other, it is a manifestation of the household members’ link to their ancestors who once occupied the same location, thus symbolically and physically affirming an ancestral right to place. Ethnographically, houses on the Pacific Northwest

Coast of North America were social fields where household and non-household members obtained, exchanged, and legitimized capital (sensu Bourdieu 1989) to establish and redefine relationships at various social scales (e.g., Barnett 1938:135,

1955:253; Washington 2004:587-590). Choices about house form, size, and location are not simply reflections of these multidimensional relationships: they express, affirm, and negotiate them (Netting et al. 1984).

9 We examine two different Northwest Coast house forms—in-ground and above- ground—and how they may have influenced household territorial and tenurial interests. Differences between in-ground and above-ground architecture have been discussed by others (e.g., Feinman et al. 2000; Grier and Kim 2012). However, emphasis has been on household production and the restructuring of social relations rather than territoriality and tenure. While examinations of the relationships between built environments and territoriality and tenure has been a focus of monumentality studies (e.g., Grier 2014; Klokler 2017; Urban and Schortman 2013), less often have the roles of household architecture been used to explore this relationship. Here, we bring houses into this wider discussion of relationships between monumental architecture and territoriality by exploring the roles of houses in expressing territorial and tenurial claims among ancestral Coast Salish populations on the Pacific Northwest Coast of southwestern British Columbia (B.C.) and northwestern Washington state, a region collectively known as the Salish Sea (Figure 2).

The link between houses and status, and houses and household production are widely recognized in the literature for the Pacific Northwest Coast generally and the Salish Sea region specifically (Ames 1996; Coupland 1985, 1996a, 1996b, 2006; Samuels 2006; Sobel 2006). We agree that houses played significant roles in household status and production; however, we suggest broadening the discussion to include an examination of territory and tenure given the monumental nature of ancestral Coast Salish architecture (Grier 2006; Suttles 1991). It is not our position that territorial and

Figure 2: Locations of in-ground and above-ground house sites discussed in the paper. Map 2a shows the distribution of Coast sites and Map 2b shows the distribution of Lower Fraser sites. ID numbers provide additional information in Tables 1-3. The dashed lines in the inset and Map 2a are approximate representations of linguistic boundaries of Northern, Central, Southern, and Southwestern Coast Salish peoples (Suttles 1990) (map generated from the ESRI basemap catalogue using ArcMap 10.3).

11 tenurial interests were the causal reason in the adoption of large houses, rather, we suggest that they were part of a suite of concerns, both intentional and unintentional, that resulted in large houses. With this in mind, we analyze ethnographic and archaeological data on house form and size in two parts of the Salish Sea region: the Coastal and Lower Fraser areas.

While joined in their historical roots and on-going interactions, these two areas differ markedly in their environmental settings. As it relates to house form, the Lower Fraser has much colder winters than the Coast. In-ground houses, which can offer greater thermal insulation, likely had a desirable functionality among Lower Fraser households. However, the large size of many Lower Fraser in-ground houses belies a strictly environmental argument for the use of this house form among the riverine populations. Controlling the ambient temperature inside an in-ground decreases as house size increases (MacDonald 2000:243).

Our analyses of Coastal and Lower Fraser houses confirm the extant hypothesis that people were moving into large multifamily dwellings in the region circa 2300 cal. BP (e.g., Matson and Coupland 1995). It is our position that large houses became widespread at this time in association with regional increases in population (e.g., Ritchie et al. 2016); however, different residency strategies were pursued. Coastal populations aggregated as multifamily households in large, above-ground houses. Lower Fraser communities achieved the same through the use of both large in-ground and above- ground house forms.

We suggest that these large structures increased both the individual families’ and the aggregated household’s visibility thereby symbolically communicating and reifying territorial and tenurial claims that were increasingly under stress due to regional population increases. We further suggest that the move into large multifamily houses was facilitated by a pre-existing modular house architecture and a social structure defined by bilateral kinship, group exogamy, and wide-ranging social networks. Finally, we contend that the widespread adoption of large houses provided new venues for the ongoing conflict between centralizing and decentralizing power interests inherent in Coast Salish social structure (e.g., Angelbeck 2009; Angelbeck and Grier 2012).

12 2.1. Territory, Tenure, Territoriality among the Coast Salish

We use the interrelated concepts of territory, tenure, and territoriality to investigate the possible links that connect house architecture to household interests in a given landscape and the socioeconomic practices used to express those interests. Territory refers to nested but permeable spaces with which an individual or group identifies through some combination of spiritual, familial, affinal, and economic relationships. Thus, territories represent both physical entities (natural and/or culturally modified places) and conceptual entities (e.g., hunting ranges, trade relations, water routes, and sociopolitical connections), which shift in extent and significance in concert with changing kinship and other socioeconomic relationships (e.g., Ingold 1986:138).

The operational corollary of expressing, maintaining, or redefining tenurial claims is territoriality. Territoriality refers to the practices that serve to communicate territorial and tenurial interests. It is best inferred archaeologically from the act of marking spaces such that they become meaningful places to those who have claimed the space (e.g., David 2002; Taçon 2008; Zedeño and Bowser 2009). Marking an area can take on many forms in the Salish Sea region: the creation of rock art; conspicuous burial styles; construction of large-scale intertidal mass harvesting features; defensive features; trail systems; permanent settlements; and large houses.

13 2.2. The Roles of Kinship and Social Relations in Territory and Tenure among the Ethnographic Coast Salish

Territory and tenure are intrinsic to Coast Salish social structure. Bilateral kinship and group exogamy provide avenues for residency and access to resources at local and extra-local scales (Collins 1979; Elmendorf 1971; Suttles 1960). Traditionally, individuals were linked with a particular place, community, and family through kinship and membership in a descent group that was legitimized through a combination of kin connections, the acquisition of an ancestral name, residency, and the investment of labour (Kennedy 2007:23-25, 28-29). Yet, the practice of group exogamy within a bilateral kinship structure allowed for membership in far-reaching socioeconomic networks (Barnett 1955:182; Kennedy 2007:6); however, any extra-local territorial or tenurial claims were contingent on legitimate ties of kinship.

Legitimacy, as proclaimed and acknowledged at cultural events (e.g., potlatches), was fundamental to clarifying ownership and access, social position, and the generational transmission of capital (Barnett 1938:135, 1955:253; Washington 2004:587-590). Houses were the most common venues where these cultural events took place and were an explicit modification to the landscape that symbolically communicated both presence and socioeconomic interest in the surrounding area (Barnett 1955:29, 39; Elmendorf 1992:169-170, 269). This role of houses is exemplified by the Central Coast Salish Cowichan and Nanaimo summer villages that were maintained in the Lower Fraser area (Suttles 2000:172, 177).

Both the Cowichan and Nanaimo had large winter villages on the East coast of Vancouver Island and the Gulf Islands in the Coastal area of the Salish Sea region. They came to their summer villages during seasonal moves for fish and other resources that were available along the Lower Fraser. The house frames in both areas were left standing year-round and would have been visible to all passersby, acting as enduring marks on the landscape that communicated both the literal and symbolic presence and socioeconomic interests of the household. Gunther (1927:188) made note of this among the Central Coast Salish Klallam: “A house site belongs to an individual as long as he leaves any evidence of a building on it.” Similarly, Elmendorf highlights the movement of wealthy Southern Coast Salish Skokomish households during winter residency from one house to another to express wider territorial claims: “Thus some 'rich families' at…the

14 main Skokomish winter-house-group site at the forks of the river, sometimes spent part of the winter in single plank houses apart from any house group at sites in the Skokomish delta or on the canal shore at the mouth of the river” (Elmendorf 1992:269).

Large above ground houses, because they are more visible from a distance, would have been more effective than large in-ground houses as a means of expressing these interests. Ethnographically, Coast Salish above-ground houses could be as much as six meters in height (Duff 1952:48; Schaepe et al. 2001:42), whereas large in-ground houses were rarely more than two meters above the ground surface (e.g., Schaepe et al. 2001:46).

2.3. Ethnographic Review of Coastal and Lower Fraser Coast Salish Houses

A number of different structures were used for shelter by the Coast Salish depending on time of year and activity (Barnett 1938, 1955; Elmendorf 1992). Here, we focus on in-ground houses (both pithouse and in-ground plankhouse styles) and above- ground, post-and-beam plankhouses (Tables 1, 2, and 3). We use in-ground to refer to any subterranean or semi-subterranean house for which the surrounding earth formed all or part of the house walls. Above-ground houses are those that were independent of the ground except for the purposes of providing a level surface to function as a floor or foundation and for anchoring the house frame.

2.3.1. In-ground Houses

The ethnographic record for the use of in-ground houses among the Coastal groups emphasizes the importance of in-ground architecture for defensive rather than domestic purposes (e.g., Barnett 1944:266-267, 1955:49-52). Functionally speaking, the ethnographic descriptions of coastal in-ground houses can be grouped into three categories—from least to most common these are: residences (Figure 3a; Table 2); residences with a defensive component (Figure 3b; Table 2); and underground bunker- like structures that were purely defensive in nature (Figure 4a and b; Table 2). Given the fully subterranean design of most of the ethnographically described Coastal in-ground houses, they would not have functioned very effectively as venues of symbolic communication. By contrast, ethnographic descriptions of Lower Fraser in-ground

15 dwellings (e.g., Duff 1952; Schaepe et al. 2001), or pithouses (Figure 4c), suggest they functioned in a similar fashion to above-ground plankhouses (e.g., Suttles 1991).

Table 1: Ethnographic and archaeological data sets of houses from the Coastal and Lower Fraser areas of the Salish Sea region. Data set Region ID#* Houses Sites (n=71) (n=120) 1-7,13,15 Coast 16,20,25,32,101- 30 24 Ethnographic 107,111-120 Lower 92-93,108 3 3 Fraser 8-12,14,17-19,21- Coast 24,26-31,33- 33 25 Archaeological 44,109-110 Lower 45-91,94-100 54 19 Fraser Note: *Identification numbers as per Figure 2, and Tables 2 and 3.

16 Table 2: In-ground houses recorded in the ethnographic and archaeological records for the coastal areas of the Salish Sea region. ID Group Form1 Description and context Use Place name2, Source3 location, site number 1 Klahoose R Ramp entrance, escape Residence nat θuwom, Toba (E) Barnett 1935-1936 unpublished field tunnel, single ridge post, River Valley notes University of British Columbia ridge post supports, flat roof Library Special Collections Division, of poles, brush, bark, earth. Vancouver, notebook 6:84, 1944:266, Settlement 1955:49; Kennedy and Bouchard 1983:159 2 Klahoose R Underground refuges Refuge qʷeqʷti:čɩnəm, (E) Barnett 1955:49; Kennedy and entered by hidden ramp or Brem/Salmon Bay Bouchard 1983:158 tunnel. Beneath plankhouse 3 Tla’amin R Ramp entrance, flat roof of Refuge ƛɛkʷanəm, Scuttle (E) Barnett 1935-1936:6:44, 1944:267, logs, planks, earth, alcoves Bay, 1955:50-51; Stanley 1954:3 cut into walls for DlSd-6 sleeping/storage. Away from settlement or beneath plankhouse 4 Tla’amin R As above Refuge χakʷum, Grief (E) Barnett 1935-1936:6:44, 1944:267, Point, DkSd-1 1955:50-51 5 Tla’amin R As above Refuge čɛn, Parker (E) Barnett 1935-1936:6:44, 1944:267, Harbour, EaSe-4 1955:50-51 6,7 Tla’amin R Rectangular depressions. Residence/ kʊmaχən, Smelt (E) Barnett 1955:50; Kennedy and Adjacent to plankhouses refuge Bay, EaSf-2 Bouchard 1983:70; (A) Angelbeck 2009:188-189, 248-257 8,9, 10 Tla’amin O Oval depressions. Residence maloʔhom, (A) Springer et al. 2013:71-93 Settlement Harwood Island, DlSd-36 11 Tla’amin C Circular/oval depression. Residence qɛqɛgɩš, Cochrane (A) Springer et al. 2013: Appendix B:295, Settlement Bay, EaSe-76 307-312 12 Tla’amin O Circular/oval depression. Residence/ t̓i: t̓i: may, Savary (E) Bloomfield 2005:21; Settlement refuge Island, DlSf-6 (A) Springer et al. 2013:138-148 13 Shíshálh R Entered via secret Refuge Pender Harbour (E) Barnett 1944:268, 1955:51 (Sechelt) passageway in plankhouse floor. Beneath plankhouse 14 Shíshálh R Rectilinear depression Residence Sihalt, Porpoise (A) Bilton 2014:128-130 (Sechelt) Bay, DjRw-1

17 ID Group Form1 Description and context Use Place name2, Source3 location, site number 15 Skxwúmish R Entered via secret Refuge Ats’a’n (E) Barnett 1944:268, 1955:53; Rudy (Squamish) passageway in plankhouse (St’et’e7imin) Reimer 2015, personal communication beneath bed platform. Beneath plankhouse 16 Xwméthkwiyem S Flat roof, single central post, Residence Musqueam (E) Barnett 1938:128-129, 1944:268, (Musqueam) 2 cross beams at right 1955:55 angles, covered with poles, planks, mats and earth, accessed by notched ladder. Settlement 17 Semiahmoo C Shallow, circular pit, two Seasonal C’iwá.q, Crescent (A) Matson et al. 1991 small post holes residence Beach, DgRr-1 (E) Don Welsh 2015, personal communication 18,19 Sq’ela’xen C 2 circular pits Residence; Ferndale, (A) Gillis 2007; Grabert 1983; Hutchings seasonal 45WH34 2004; Nokes 2004 residence 20 Wsáneć R Deep pit, small opening for Refuge Saanich (E) Lugrin 1932 (cited in Angelbeck (Saanich) access, spacious inside. Peninsula 2009:187) Away from settlement 21 Wsáneć O Oval depression. Settlement Seasonal Birch Point, (A) Gaston 1975 residence 45WH48 22,23 Klallam R 2 Sub-rectangular pits Seasonal Sequim, 45CA426 (A) Morgan et al. 1999 residence 24 Swinomish N/A Excavated pit with post holes Seasonal Swinomish, (A) Mattson 1971 residence 45SK51 25 Skagit R Excavated pits covered by Refuge Penn Cove, (E, A) Bryan 1955:192-194, 1963:79-80 planks and brush. Away from 45IS50 settlement Note: 1R-Rectangular, C-Circular, O-Oval, S-Square; 2See Appendix A; 3E-Ethnographic, A-Archaeological.

18 Table 3: Temporal and area data for in-ground and above-ground houses identified in the Salish Sea region. ID Sample# Site Material Conventional Calibrated Inferred Type**/ Source Comments age (BP) 2σ (BP)* date Area (BP) (m2) 1 - Náath’úwem - - - 250-100 IG/48 See Table 2 Date range/area inferred from oral record 2 - Kw’íkw’tichenam - - - 250-100 IG/~108 See Table 2 Date range/area inferred from oral record 3 - DlSd-6 - - - 250-100 IG/~108 See Table 2 Date range/area inferred from oral record 4 - DkSd-1 - - - 250-100 IG/~108 See Table 2 Date range/area inferred from oral record 5 - EaSe-4 - - - 250-100 IG/~108 See Table 2 Date range/area inferred from oral record 6 - EaSf-2 - - - 250-100 IG/250 See Table 2 Date range inferred from oral record; area inferred from archaeological record 7 - EaSf-2 - - - 250-100 IG/156 See Table 2 Date range inferred from oral record; area inferred from archaeological record 8 Beta- DlSd-36 Organic 2330+/-40 2469-2304 - IG/40 See Table 2 299968 sediment 9 Beta- DlSd-36 Organic 2610+/-30 2773-2720 - IG/80 See Table 2 342743 sediment 10 Beta- DlSd-36 Organic 2720+/-30 2866-2760 - IG/40 See Table 2 342745 sediment 11 Beta- EaSe-76 Wood 2410+/-30 2490-2350 - IG/49 See Table 2 314097 charcoal 12 Beta- DlSf-6 Shell 2400+/-30 1693-1380* - IG/36 See Table 2 342736

19 ID Sample# Site Material Conventional Calibrated Inferred Type**/ Source Comments age (BP) 2σ (BP)* date Area (BP) (m2) 13 - Pender Harbour - - - 250-100 IG/~108 See Table 2 Date range/area inferred from oral record 14 Beta- DjRw-1 Wood 2880+/-40 3150-2880 - IG/32 See Table 2 270113 charcoal 15 - Ats’a’n - - - 250-100 IG/~108 See Table 2 Date range/area (St’et’e7imin) inferred from oral record 16 - Musqueam - - - 250-100 IG/20.25 See Table 2 Date range/area inferred from oral record 17 WSU-4246 DgRr-1 Wood 3010+/-85 3378-2966 - IG/12.25 See Table 2 charcoal 18 Beta- 45WH34 Wood 4000+/-40 4578-4403 - IG/36 See Table 2 176490 charcoal 19 Beta- 45WH34 Wood 4370+/-90 5301-4820 - IG/16 See Table 2 176489 charcoal 20 - Saanich Peninsula - - - 250-100 IG/~108 See Table 2 Date range/area inferred from oral record 21 - 45WH48 Diagnostics - - 1400-650 IG/70 See Table 2 Date range inferred from diagnostic artifacts 22 Beta- 45CA426 Wood 2470+/-80 2733-2357 - IG/12.25 See Table 2 111181 charcoal 23 Beta- 45CA426 Wood 2480+/-50 2725-2377 - IG/22.5 See Table 2 107612 charcoal 24 - 45SK51 Diagnostics - - 2000- IG/- See Table 2 Date range 1700 inferred from diagnostic artifacts 25 - 45IS50 - - - 250-100 IG/144 See Table 2 Date range inferred from oral record 26 Beta- DlSd-6 Wood 910+/-90 969-676 - AG/- Jackley 2012 265374 charcoal

20 ID Sample# Site Material Conventional Calibrated Inferred Type**/ Source Comments age (BP) 2σ (BP)* date Area (BP) (m2) 27 Beta- DlSd-6 Wood 2020+/-40 2065-1883 - AG/- Jackley 2012 263324 charcoal 28 Beta- EaSe-76 Wood 990+/-40 970-790 - AG/117 Springer et al. 258565 charcoal 2013 29 Beta- EaSe-76 Wood 1300+/-30 1290-1180 - AG/117 Springer et al. 314096 charcoal 2013 30 BGS-2262 DgRv-2 Wood 990+/-105 1089-691 - AG/104 Matson 2003 charcoal 31 - DgRv-6 - - - 1000-650 AG/400 Derr et al. Structure dated, 2012 specific dates have not yet been published 32 - 45KI51 Diagnostics - - 160-125 AG/243 Chatters 1989 Date range inferred from diagnostic artifacts 33 Beta-2798 45KI59 Wood 1570+/-90 1627-1300 - AG/126 Chatters et al. charcoal 1987 34 - DgRv-3 - 1610+/-60 1622-1366 - AG/200 Grier 2003, 2006 35 GAK-2754 DgRw-4 Wood 1679+/-90 1815-1394 - AG/- Wilmeth 1978 charcoal 36 - DfRu-44 Diagnostics - - 3500- AG/64 Johnstone Date range 2400 1991 inferred from diagnostic artifacts 37 SFU-41 DgRs-1 Wood 1270+/-160 1934-1567 - AG/130 Matson et al. charcoal 1980 38 SFU-42 DgRs-1 Wood 1480+/-80 1934-1567 - AG/121 Matson et al. charcoal 1980 39 TO-543 EaSf-36 Wood 1950+/-50 2003-1776 - AG/- Mathews 2003 charcoal 40 - DgRs-14 - 2110+/-65 2208-1931 - AG/293 Archaeology Branch 2016a 41 - DgRs-14 - 2110+/-65 2208-1931 - AG/369 Archaeology Branch 2016a 42 SFU-597 DeRt-1 Wood 2190+/-60 2339-2042 - AG/26 Carlson 1986 charcoal

21 ID Sample# Site Material Conventional Calibrated Inferred Type**/ Source Comments age (BP) 2σ (BP)* date Area (BP) (m2) 43 Beta- 45SJ169 Charred 2450+/-90 2741-2347 - AG/23 Walker et al. 170650 material 2003 44 WSU-2202 45CA213 Wood 2770+/-90 3080-2744 - AG/36 Croes 1995 charcoal 45 Beta- DiRi-48 Wood 150+/-40 284-54 - IG/59 Schaepe 2009 210177 charcoal 46 Beta- DiRi-48 Wood 510+/-40 560-500 - IG/34 Schaepe 2009 210178 charcoal 47 Beta- DiRi-15 Wood 140+/-40 282-57 - IG/133 Schaepe 2009 213531 charcoal 48 Beta- DiRi-15 Wood 220+/-60 332-60 - IG/130 Schaepe 2009 196134 charcoal 49 Beta- DiRi-15 Wood 320+/-40 480-302 - IG/179 Schaepe 2009 213529 charcoal 50 Beta- DiRi-15 Wood 640+/-40 668-551 - AG/- Schaepe 2009 213534 charcoal 51 Beta- DjRi-14 Wood 250+/-40 436-267 - IG/94 Schaepe 2009 210175 charcoal 52 Beta- DjRi-14 Wood 260+/-40 459-269 - IG/118 Schaepe 2009 210174 charcoal 53 Beta- DjRi-14 Wood 270+/-40 464-275 - IG/76 Schaepe 2009 210176 charcoal 54 Beta- DhRl-83 Wood 300+/-40 473-288 - IG/113 Schaepe 2009 210181 charcoal 55 Beta- DiRj-1 Charred 320+/-40 480-320 - IG/67 Schaepe 2009 208880 material 56 Beta- DiRj-1 Charred 2130+/-40 2180-1995 - IG/73 Schaepe 2009 208882 material 57 Beta- DiRj-1 Charred 2300+/-40 2363-2297 - IG/74 Schaepe 2009 208881 material 58 I-6191 DiRj-1 Wood 2430+/-90 2742-2332 - IG/94 Schaepe 2009 charcoal 59 Beta- DiRj-1 Charred 2380+/-40 2497-2334 - IG/55 Schaepe 2009 208885 material 60 Beta- DiRj-1 Charred 2470+/-40 2717-2379 - IG/95 Schaepe 2009 208879 material 61 Beta- DhRl-65 Wood 390+/-40 513-316 - IG/12 Springer and 235513 charcoal Lepofsky 2011

22 ID Sample# Site Material Conventional Calibrated Inferred Type**/ Source Comments age (BP) 2σ (BP)* date Area (BP) (m2) 62 Beta- DhRl-65 Wood 170+/-40 295-64 - AG/14 Springer and 235512 charcoal Lepofsky 2011 63 M-1511 DjRi-5 - 570+/-100 704-431 - IG/131 Borden 1983 64 Beta- DhRl-74 - 550+/-40 646-512 - IG/68 Ritchie 2010 208883 65 Beta- DhRl-74 - 1550+/-40 1532-1356 - AG/116 Ritchie 2010 235517 66 Beta- DhRl-74 - 1080+/-40 1063-927 - IG/76 Ritchie 2010 235515 67 Beta- DhRl-74 - 1090+/-40 1074-927 - IG/118 Ritchie 2010 208884 68 Beta- DhRl-74 - 1270+/-40 1288-1172 - IG/118 Ritchie 2010 240398 69 Beta- DhRl-74 - 1280+/-40 1293-1172 - IG/68 Ritchie 2010 240399 70 Beta- DhRl-74 - 1070+/-40 1059-927 - AG/220 Ritchie 2010 269609 71 Beta- DhRl-74 - 1130+/-40 1151-960 - AG/116 Ritchie 2010 235516 72 Beta- DhRl-74 - 1250+/-40 1278-1072 - AG/215 Ritchie 2010 240397 73 Beta- DhRl-74 - 1470+/-40 1415-1295 - AG/407 Ritchie 2010 235518 74 Beta- DgRl-17 Wood 1070+/-40 1059-927 - IG/69 Schaepe 2009 210180 charcoal 75 Beta- DgRl-17 Wood 1170+/-40 1181-977 - IG/62 Schaepe 2009 210179 charcoal 76 GAK-5429 DiRi-38 Wood 1300+/-100 1380-1042 - IG/18 Von Krogh charcoal 1980 77 Beta- DiRj-30 Wood 1490+/-40 1419-1303 - IG/77 Schaepe 2009 210171 charcoal 78 Beta- DiRj-30 Wood 1980+/-40 2004-1860 - IG/61 Schaepe 2009 210170 charcoal 79 Beta- DiRj-30 Wood 2040+/-40 2116-1919 - IG/33 Schaepe 2009 210173 charcoal 80 Beta- DiRj-30 Wood 2110+/-40 2159-1987 - IG/59 Schaepe 2009 217439 charcoal

23 ID Sample# Site Material Conventional Calibrated Inferred Type**/ Source Comments age (BP) 2σ (BP)* date Area (BP) (m2) 81 Beta- DiRj-30 Wood 2200+/-29 2314-2141 - IG/58 Schaepe 2009 210169 charcoal 82 GAK-5432 DiRj-14 Wood 1580+/-80 1622-1310 - IG/28 Von Krogh charcoal 1980 83 Beta- DgRk-10 Charred 2620+/-60 2862-2683 - IG/33 Merchant et al. 128241 material 1999 84 Beta- DgRk-10 Charred 4130+/-40 4821-4566 - IG/21 Merchant et al. 128607 material 1999 85 - DhRo-10 - 3004+/-39 3269-3068 - IG/- Archaeology Branch 2016b 86 UCIAMS- DhRp-52 Wood 3290+/-15 3564-3475 - IG/79 Hoffmann et 67317 charcoal al. 2010 87 UCIAMS- DhRp-52 Wood 4490+/-15 5285-5046 - IG/200 Hoffmann et 67303 charcoal al. 2010 88 XA-3581 DhRp-52 Wood 4496+/-24 5290-5046 - IG/188 Hoffmann et charcoal al. 2010 89 GAK-4919 DhRk-8 - 4220+/-100 4982-4511 - IG/38 Lepofsky et al. 2009 90 SFU-888 DgRn-23 - 4490+/-70 5315-4955 - IG/110 Lepofsky et al. 2009 91 Beta- DgRn-23 - 4840+/-110 5759-5316 - IG/16 Lepofsky et al. 77758 2009 92 - Yale - - - 250-100 AG/98 Lamb 1960 Date range/area inferred from historic record 93 - Matsqui - - - 250-100 AG/4200 Lamb 1960 Date range/area inferred from historic record 94 - DhRo-28 - 316+/-40 480-300 - AG/- Archaeology Branch 2016c 95 Beta- DhRl-15 Wood 720+/-40 728-642 - AG/- Schaepe 2009 217441 charcoal 96 Beta- DhRl-70 - 830+/-40 798-676 - AG/175 Ritchie 2010 235520 97 WSU-5052 DhRl-16 Wood 1850+/-50 1893-1692 - AG/- Lepofsky et al. charcoal 2000a 98 CAMS- DhRl-16 Charred 2250+/-70 2401-2043 - AG/187 Lepofsky et al. 61998 wood 2000a

24 ID Sample# Site Material Conventional Calibrated Inferred Type**/ Source Comments age (BP) 2σ (BP)* date Area (BP) (m2) 99 CAMS- DhRl-16 Charred 2410+/-50 2618-2346 - AG/- Lepofsky et al. 61997 wood 2000a 100 WSU-5051 DhRl-16 Wood 2940+/-180 3511-2743 - AG/- Lepofsky et al. charcoal 2000a 101 - Comox - - - 250-100 AG/378 Barnett 1955 Date range/area inferred from oral record 102 - Squamish - - - 250-100 AG/666 Barnett 1955 Date range/area inferred from oral record 103 - Squamish - - - 250-100 AG/108 Mauger 1978 Date range/area inferred from oral record 104 - Port Townsend - - - 250-100 AG/45 Mauger 1978 Date range/area inferred from oral record 105 - Washington - - - 250-100 AG/915 Mauger 1978 Date range/area Harbour inferred from oral record 106 - 45KP2 - - - 250-100 AG/2440 Mauger 1978 Date range inferred from oral record; area inferred from archaeological/or al records 107 - Tulalip Harbour - - - 250-100 AG/455 Mauger 1978 Date range/area inferred from oral record 108 - Katzie - - - 250-100 AG/270 Mauger 1978 Date range/area inferred from oral record 109 Beta- DgRs-2 Charred 430+/-80 557-304 - AG/- Stryd et al. 34785 wood 1991 110 TAB004 45SJ24 Wood 1460+/-50 1418-1288 - AG/500 Stein et al charcoal 2003 111 - - - - - 250-100 AG/- Barnett 1935- Date range 1936:6:106 inferred from oral record

25 ID Sample# Site Material Conventional Calibrated Inferred Type**/ Source Comments age (BP) 2σ (BP)* date Area (BP) (m2) 112 - EaSf-2 - - - 250-100 AG/- Angelbeck Date range 2009 inferred from oral record 113 - EaSe-11 - - - 250-100 AG/- Barnett 1955 Date range inferred from oral record 114 - DkSd-1 - - - 250-100 AG/- See Table 1 Date range inferred from oral record 115 - EaSe-4 - - - 250-100 AG/- See Table 1 Date range inferred from oral record 116 - Klumklums - - - 250-100 AG/- Barnett 1935- Date range 1936:6:106 inferred from oral record 117 - Nanaimo - - - 250-100 AG/- Barnett 1955 Date range inferred from oral record 118 - EaSd-7 - - - 250-100 AG/- Acheson and Date range Riley 1976 inferred from oral record 119 - DlSd-11 - - - 250-100 AG/- Barnett Date range 1955:29 inferred from oral record 120 - DeRv-2 - - - 250-100 AG/- Acheson et al. Date range 1975 inferred from oral record Note: *A marine correction of 421+/-54 was applied to the shell dates using Calib 7.1 (Stuiver and Reimer 2016). The correction is an average derived from local Delta-R dates from five locations in the Salish Sea: Savary Island (510+/-50); Departure Bay (440+/-50); Burrard Inlet (410+/- 40); Comox (380+/-50); and Nanoose Bay (370+/-50); **IG – In-ground; AG – Above-ground.

Figure 3: 3a Interpretation of the Coast Salish-Musqueam Nation winter home for wealthy families (Barnett 1944:268; ID 16 in Tables 1 and 2); 3b interpretation of the Coast Salish-Klahoose Nation winter homes used in the Toba River Valley (Barnett 1935-1936:6:84; ID 1 in Tables 2 and 3; drawings by Chris Springer).

Figure 4: 4a Interpretation of plankhouse-subterranean refuge described for the Coast Salish-Klahoose, Squamish, Sechelt, and Tla’amin Nations; 4b interpretation of subterranean refuge described by Barnett (1955:49-50), without associated plankhouse (Barnett 1955:49-50; ID 2-7, 13, and 15 in Tables 2 and 3); 4c interpretation of Coast Salish-Sts’ailes Nation pithouse from the Lower Fraser (adapted from Springer and Lepofsky 2011:35, Figure 6) (drawings by Chris Springer).

28 2.3.2. Above-ground Houses

There are numerous detailed descriptions of Coast Salish plankhouse architecture in the ethnographic record (e.g., Barnett 1955; Elmendorf 1992; Hill-Tout 1978a, 1978b, 1978c; Suttles 1991). Historically, the most common were the shed-roof and gable-roof forms (Figures 5 and 6). The latter was primarily associated with Northern Coast Salish (Barnett 1938:127, 1955:35, 45, 47; Suttles 1991:14) and Southern Coast Salish populations (Elmendorf 1992:154-165; Suttles 1991:14) but were occasionally used among the Central Coast Salish in the Lower Fraser area (Duff 1952:48; Figure 6). The former was used throughout the Salish Sea region.

The modularity of the Coast Salish plankhouse enabled families and households to communicate territorial and tenurial interests at important locations used throughout their seasonal round (Figure 5). With independent wall and roof planks cladding a stand- alone, post-and-beam frame (Suttles 1991), the houses easily accommodated moves and changes to household size. Families often owned the planks covering their section of the house frame and would carry these planks between seasonal residences (Suttles 2000:184-185). By carrying planks to clad permanent frames maintained at various locations, families literally and symbolically transported with them both their house and their connections to places.

The positioning of above-ground houses would have also contributed to their potential role as vehicles of symbolic communication. At the local scale, prominent households conveyed status by siting their houses in front and central positions facing the water thereby increasing their visibility to both household and non-household members (Barnett 1955:19, 29, 32). This strategy would have effectively communicated some kind of territorial and tenurial interest in the surrounding landscape and its

Figure 5: Generic interpretations of most common plankhouse forms in the Salish Sea region: 5a shed-roof; 5b gable-roof (drawings by Chris Springer, adapted from figures and descriptions in Barnett [1955:35-58], Boas [1890:11- 12], Elmendorf [1992:154-165], Suttles [1990:7], and Schaepe et al. [2001:40-42]).

Figure 6: Upper image – Central Coast Salish village in the Coastal area of Salish Sea region (Alden 1858); lower image – Central Coast Salish village in the Lower Fraser area of the Salish Sea region. Note the high visibility of the large above-ground houses in both villages compared with the more restricted visibility of the in-ground houses (center mid-ground) in the Lower Fraser village. (Alden 1858 Image PDP02144 courtesy of the Royal British Columbia Museum and Archives).

31 associated water routes. Increasing visibility was particularly effective with above-ground houses as highlighted in the two village scenes in Figure 6.

2.4. Methods

To evaluate how territorial and tenurial claims may have played out in past built environments, we rely on both ethnographic and archaeological data sets. The ethnographic data are used to infer ancestral architectural styles and to provide insights into the potential symbolic nature of houses in the past. It is extremely rare for significant, wooden structural remains to survive in the soils of the Northwest Coast. For this reason, the ethnographic record is a critical resource for understanding the architectural characteristics of the houses built and occupied by ancestral populations.

From the ethnographic sources, we sought information on house size and form. For our temporal analysis, an arbitrary range of 100-250 years BP was given to ethnographically described houses. This range was chosen because the given house type was still in community memory and, at least in the case of above-ground houses, was, in many instances, still standing during the 19th or even early 20th centuries AD and could have been standing at least as long as 150 years.

From the archaeological sources, we only included sites from which chronometric or inferred dates (i.e., based on diagnostic or historic artifacts found in direct association with the house) and area measurements of the potential liveable space (i.e., floor area in m2) of a given house were available. Radiocarbon dates were calibrated using the online program Calib 7.1 (Stuiver et al. 2017) to generate the median dates for each house to compare house sizes through time. The averages of inferred date ranges were used as the median dates for sites with no associated chronometric dates. Calib 7.1 was also used to generate probability distributions of house forms through time based on 2σ calibration ranges. Statistical comparisons of house forms were made using independent samples Mann-Whitney U tests (see Appendix B).

32 2.5. Results

2.5.1. Temporal Variation of both House Forms (Ethnographic and Archaeological Data Combined)

When plotted over time, the combined ethnographic and archaeological Coastal and Lower Fraser data sets show that in-ground houses were in use in both areas by approximately 5500 cal. BP (Figure 7). Above-ground houses begin to appear among Coastal and Lower Fraser communities circa 3500 cal. BP. An important difference between the two data sets is reflected in the size of the earlier in-ground houses. Three of the eight early in-ground houses in the Lower Fraser region were very large structures (Figure 7; Table 3, ID 87 [200 m2], ID 88 [188 m2], and ID 90 [110 m2]), similar in size to some more recent above-ground plankhouses (e.g., Grier 2003). This may represent an incipient strategy for using architecture as a means of communicating presence on the landscape—a strategy that appears to have been repeated, in a limited fashion, among both Lower Fraser and Coastal populations with above-ground architecture at 3500 cal. BP.

While the practical attributes of houses could have been adequately fulfilled with small structures, people chose instead to build large houses. We suggest that building large houses was of greater symbolic importance than simply satisfying practical needs. Following this logic, 3500 years BP represents an early example of communicating intracommunity social distinction via house size among Coastal and Lower Fraser households with the new addition of above-ground architecture; a strategy that increased in importance post-2300 cal. BP (Figures 7 and 8).

Our temporal results for the widespread appearance of large houses in the Salish Sea region post-2300 cal. BP mirror the demographic results for the Lower Fraser determined by Ritchie et al. (2012). Using radiocarbon data in combination with dated site types for the Lower Fraser, Ritchie et al. (2012) show a general population rise beginning around 6000 years ago, and a move toward aggregation into large settlements between 2250 and 2000 cal. BP. Given the similarity of our temporal results, we suggest the population rise posited for the Lower Fraser area was part of a regional social phenomenon that, in part, led to the general incorporation of large above-ground houses into the pre-existing residency pattern.

33 2.5.2. Coastal and Lower Fraser Houses Pre-2300 versus Post-2300 cal. BP

Building above ground represents the potential for a substantial increase in internal liveable space. House size information derived from both archaeological and ethnographic sources show that there was an increase in liveable space over time in both Coastal and Lower Fraser above-ground houses post-2300 cal. BP (Figure 9). To further examine this phenomenon, we focused on the size of archaeological above- ground and in-ground houses to eliminate the possible bias resulting from our observation that ethnographic/historic documents privilege larger, more visible above- ground plankhouses, and under-represent in-ground houses. We found that there was a significant increase in liveable space in Coastal houses post-2300 cal. BP (MWU 17.0, p=0.001). By comparison, there was no significant change in Lower Fraser houses (MWU 179.0, p=0.458). Among the Coastal groups, the increase in liveable space seems to have been concomitant with the rise in above-ground architecture post-2300 cal. BP. Although our results show no significant change in liveable space within the Lower Fraser data set, it is possible that the above-ground houses in both areas were used as venues of symbolic communication.

Figure 7: High-low graphs of in-ground and above-ground houses through time identified in Coast (7a) and Lower Fraser (7b) areas. Bars represent calibrated 2σ range or inferred ranges based on diagnostic and/or historic artifacts. ID numbers provide additional information in Tables 2 and 3. These data represent the presence/absence of in- ground and above-ground houses.

Figure 8: Probability distributions of in-ground and above-ground houses through time based on 2σ calibration ranges of the Coast (8a) and Lower Fraser (8b) houses for which complete chronometric data were available. ID numbers provide additional information in Table 3.

36 Given the similar sizes of Coastal and Lower Fraser above-ground houses post- 2300 cal. BP (MWU 45.0, p=0.836), we suggest that these large structures were, in part, used for communicating presence on the landscape in both areas. The continued use of in-ground houses in the Lower Fraser post-2300 suggests the possibility that both house forms played a similar role among the riverine populations with respect to communicating household interests. This suggestion is supported by the large size of some of the archaeologically recorded in-ground houses and by ethnographic descriptions for the use of Lower Fraser in-ground houses as venues for ceremonialism (Duff 1952:47)—events that served, in part, to demonstrate socioeconomic status and to express territorial and tenurial claims.

2.5.3. Coastal and Lower Fraser In-ground Houses through Time

We compared the liveable space of Coastal and Lower Fraser in-ground houses, and the liveable space of Lower Fraser in-ground houses to test the possibility that in- ground houses were used for symbolic communication. We found that there is a difference between the two data sets with respect to liveable space in Coastal and Lower Fraser in-ground houses (MWU 94.5, p=0.001), but no comparable change in the liveable space of Lower Fraser in-ground houses through time (MWU 159, p=0.934). Considering the larger size of Lower Fraser in-ground houses (min/max 15.7/200 m2) compared with Coastal in-ground houses (min/max 12.2/80 m2), we hypothesize that in- ground architecture likely played a similar communicative role as above-ground architecture in the Lower Fraser both pre- and post-2300 cal. BP.

Figure 9: Area (m2) of in-ground and above-ground houses in coastal territories and in the Lower Fraser plotted by time. Represented are only those houses for which both temporal (chronometric or inferred median dates) and area data were available (9a Coastal, N=46 houses/35 sites; 9b Lower Fraser, N=49 houses/20 sites). The outliers, Old Man House (ID 106, 2400 m2) and the Matsqui House (ID 93, 4200 m2), were not included in the Coastal and Lower Fraser datasets, respectively.

38 2.6. Discussion

The use of houses as vehicles of symbolic communication has long been recognized as a significant factor in the expression and reification of cultural norms (e.g., Blanton 1994; Coupland 2013; Grier 2006; Rapoport 1969). For the ethnographic Coast Salish, the house was particularly important as a social field of daily practice and ceremonialism. We have proposed here that as loci of legitimation of various rights and claims, houses were at once internal spaces that facilitated formalized expressions of territoriality and were the external objectification of that territoriality through the size of the house marking the landscape. However, the form of house significantly affected its role as a vehicle of symbolic communication.

Our results show a difference in the use of residential architecture among the ancestral Coastal and Lower Fraser households pre- and post-2300 cal. BP. First, there is the possibility that the Lower Fraser has long been a contested space where both in- ground and above-ground houses were used to communicate territorial interests. Earlier populations were likely fewer in number (e.g., Ritchie et al. 2012) such that large, in- ground houses sufficed as markers on the landscape.

Second, at approximately 2300 cal. BP, ancestral Coastal and Lower Fraser families began negotiating new collective relationships as communal households. We suggest this was done, in part, to strengthen and maintain territorial and tenurial interests. To do this, multifamily households, especially on the coast, began to amplify their expressions of territoriality by using large, above-ground houses to communicate their presence. We hold that this new level of aggregation and territorializing was enabled by modular architecture, socioeconomic and genealogical networks that flowed from bilateral kinship and group exogamy, and cultural mores that encouraged both social autonomy and collective action (Angelbeck 2009).

Finally, the use of in-ground houses in both areas manifested differently. Among the Coastal populations, in-ground houses were all but abandoned as residences in favour of above-ground architecture. In the Lower Fraser, in-ground houses were primary residences and played similar communicative roles as above-ground houses with respect to household status and, by extension, household territorial and tenurial claims. The continued practice of building in-ground architecture in the Lower Fraser

39 was, in part, likely due to the colder winter months—but the use of large in-ground houses questions a purely environmental explanation. Large houses were more difficult to keep warm; however, in large multifamily houses, individual families were better able to express and defend territorial and tenurial claims in an increasingly competitive sociopolitical climate.

In a social system defined by familial autonomy, collectivism, and achieved status, the move to multifamily households would have exacerbated the inherent tension between an overarching egalitarian ethos characterized by contextual leadership and tendencies toward vertical power. Aggregation in large houses added a new venue for the expression of this sociopolitical conflict within and between communities that recognized no central authority but acknowledged status differences concomitant with the acquisition of capital. Large houses became an essential part of this signaling behaviour beginning at 2300 cal. BP; especially above-ground houses.

We suggest that the large above-ground house gained wider use among Coastal and Lower Fraser communities post-2300 cal. BP for a number of reasons. First, it was better suited to communal living. Second, it allowed for some level of family autonomy despite collective residency. Third, it was able to easily accommodate shifts in household size. Fourth, it provided a more suitable architecture for large-scale ceremonialism that, among other things, contributed to the legitimization of territorial and tenurial claims. Finally, it was a superior venue for symbolic communication.

Whether intentional or unintentional, we propose that the symbolic nature of above-ground architecture was an important result of communal living, which households and families began to manipulate in their favour. Historically, above-ground houses were highly visible from the water both closeup and from a distance. The enhanced visibility of above-ground houses emphasizes the significant difference between the use of in-ground and above-ground architecture as a means of signaling territorial and tenurial claims. Furthermore, it can, in part, explain the addition of above- ground architecture to the in-ground residency pattern among the Lower Fraser communities. That is, above-ground houses would have augmented the symbolic communicative potential of settlements in an increasingly competitive sociopolitical climate beginning circa 2300 cal. BP.

40 2.7. Conclusion

As built environments, houses are physical manifestations of household social relations and the interhousehold networks that emerge from them. These social networks are fluid by nature and often ephemeral archaeologically. The archaeological remains of houses provide the proxy material evidence with which to track the ebb and flow of past networks that articulated within and across households. As well as being anchors for these dynamic social networks, houses also afford examination of the various socioeconomic and political interests of the household members that comprised the social networks.

The communicative nature of large dwellings is difficult to appreciate from the often trace evidence of houses in the archaeological record. However, ethnographic and archaeological house studies from around the world have shown that houses are major architectural investments that serve as more than simply residential units (Blanton 1994; Madella et al. 2013; Sobel et al. 2006; Rapoport 1969). They are at once the physical spaces of daily practice and venues of symbolic communication. They are, and were, directly implicated in the production and reproduction of household identity and interests within and beyond the local community. Our study on the shift from in-ground to above- ground houses has shown the importance of combined data sets for examining ancestral social relationships. Together, the archaeological and ethnographic records of houses and households is a powerful means of reconstructing how people used their built spaces to give meaning and productive power to their world view.

41 Chapter 3. Obsidian in the Salish Sea: An Archaeological Examination of Ancestral Coast Salish Social Networks in SW British Columbia and NW Washington State

On the Pacific Northwest Coast, as elsewhere among nonstate societies, social networks were realized primarily through the affinal networks that formed the basis of kinship relationships (Abbott 1972; Kennedy 2007; Suttles 1960). An important consequence of these networks was access to resources from neighbouring and distant communities (Collins 1979; Suttles 1963), giving substance to abstract notions of territorial and tenurial claims beyond the local sphere. In this sense, social networks functioned as a form of territoriality that facilitated the intra and interregional movement of people, resources, and information; while these networks are documented in the ethnographic record (e.g., Collins 1979; Elmendorf 1971, 1992; Sobel 2012; Suttles 1960, 1963, 1990), their presence archaeologically must be inferred from the material remains of trade and exchange.

The materials of trade and exchange often leave no significant archaeological footprint. However, the spatial and temporal distributions of toolstone provide an opportunity to infer the presence of the social networks that facilitated trade and exchange (e.g., Aoyama 2014; Bettinger 1982; Carlson 1994; Morin 2012; Reimer 2012a, 2014; Sobel 2012; Tykot 2002). Although toolstone was likely of limited importance among coastal communities relative to perishable items (e.g., Ames 2005), it is the only archaeological resource that survives in sufficient numbers to allow for an investigation into ancestral networks. The patterning of toolstone also provides the opportunity to explore the link between past social networks and their relationship to territorial and tenurial claims.

In this paper, we use the distribution of obsidian toolstone within the Salish Sea region of southwestern British Columbia (B.C.) and northwestern Washington state to examine territorial and tenurial claims as expressed through social networks (Figure 10). The Salish Sea is coterminous with the Coast Salish world and is linguistically divided into four main language groups: Northern Coast Salish, Central Coast Salish, Southern Coast Salish, and Southwestern Coast Salish. The four groups are culturally similar to

42 one another (Barnett 1955; Collins 1979; Elmendorf 1992; Kennedy 2007; Suttles 1960); bilateral kinship, group exogamy, place-based identity, and intra and intervillage network organization are characteristic of each. Given this cultural cohesion, the spatial distribution of obsidian toolstone can be used as a proxy for examining ancestral Coast Salish expressions of territory, tenure, and territoriality as expressed through social networks. Ethnographically, these networks linked families and households at both intra and interterritorial scales. Toolstone would have been one of the many resources that was moved through these wide-ranging networks (Collins 1979:249).

We hypothesize that ancestral Coast Salish toolstone trade will be expressed spatially in the archaeological record in a range of ways related to proximity to source and the directionality and strength of the social networks that moved the toolstone. In the absence of any mitigating social relations or geographical restrictions, we expect that toolstone would exhibit a simple distance-decay pattern along travel corridors (e.g., Dixon et al. 1968; Hodder 1974; Renfrew 1977), moving away from the raw material source. In addition, in this hypothetically socially and geographically textureless landscape, we would expect high quality materials to move farther from the source than lower quality materials.

By contrast, we propose that if social networks influenced the movement of toolstone, the distribution of obsidian would reflect the directionality and reach of these networks. Furthermore, the strength of social networks would influence both the quantity and quality of toolstone ending up in a given territory regardless of proximity to source. We further predict that as connections to, and knowledge of, landscapes deepened there would be an increasing emphasis on preferred toolstone sources through time, even if those sources were located in distant locales. The corollary of deeper connections is a greater spatial awareness of where, and through whom, preferred but locally unavailable resources of any kind could be acquired.

Our results reveal complex spatial and temporal patterns of obsidian toolstone procurement in the Salish Sea region. Although obsidian from a variety of toolstone sources located in Oregon and Washington states to the south, and B.C. to the north of the Salish Sea, was available to ancestral Coast Salish over the last six millennia, distribution was uneven. Furthermore, there is some indication that the lower Fraser Valley may have acted as a trade corridor for much of the Oregon toolstone to other

43 Central and Northern Coast Salish (see Figure 10 for the location of the lower Fraser Valley). Finally, our results also suggest that some combination of environmental and social mechanisms significantly limited access to obsidian toolstone during the last 600 years of Coast Salish history, with an all but complete drop in the use of obsidian toolstone over the last 250 years.

Figure 10: General locations of all archaeological sites (circles) and obsidian toolstone sources (stars) referred to in this paper. Arrows show hypothesized routes through which obsidian toolstone was moved into the Salish Sea region via social networks. Major trading hubs referred to in this paper are marked with an “x”. The dotted line represents the approximate boundary of Coast Salish languages (Suttles 1990) (map generated from the ESRI basemap catalogue using ArcMap 10.3).

45 3.1. Territory, Tenure, and Territoriality in Ancestral Coast Salish Social Networks

To examine the connection between social networks and the movement of obsidian, we use the interrelated concepts of territory, tenure, and territoriality. These concepts are useful for giving substance to bridging arguments between ethnographic descriptions of Coast Salish social networks and archaeological patterns of obsidian toolstone. Although the concepts of territory and tenure most commonly refer to economic claims to physical places, they also have more abstract components such as the definition and regulation of access to resources.

We use territory to refer to nested but permeable spaces with which an individual or group identifies through some combination of spiritual, familial, affinal, and economic relationships. In this sense, a territory is an abstract entity (e.g., hunting range, trade relationship, water route, and sociopolitical connection) that shifts in extent and significance in concert with affinal ties, kinship, and other socioeconomic relationships (Ingold 1986:138).

Tenure defines the communication of these relationships within a continuously evolving social sphere (Ingold 1986: 130, 136-137). Fundamental characteristics of tenure include the public legitimation and recognition of access and rights to resources at local and extra-local scales, the generational transfer of prerogatives to resources, and the knowledge and social protocols that are embedded in tenurial relations. Tenurial claims weave together past, present, and future socioeconomic relations and practices that are subject to negotiation and redefinition as part of the public legitimizing process (Barnett 1955:253; Washington 2004:587-590). One of the most significant public events during which tenurial claims were made, legitimized, or challenged among the Coast Salish is commonly referred to as the potlatch (Barnett 1955; Suttles 1960; Trosper 2002; Washington 2004). These large-scale events brought invited households together to witness the declaration, display, and distribution of various forms of capital as an expression of the host household’s past and ongoing status and legitimacy. At such events, the familial/affinal ties that formed the basis of trade relations, as well as toolstone acquired through trade, would have been among the many forms of social and economic capital declared and on display for witnessing guests.

46 The operational corollary of expressing, maintaining, or redefining tenurial claims is territoriality. Territoriality refers to the practices that serve to communicate territorial and tenurial interests, including trade and exchange. Among the Coast Salish, the relationships that framed the local, regional, and, in some cases, interregional movement of resources were linked to marriage practices and a kinship structure that allowed for membership in far-reaching socioeconomic networks (Barnett 1955:182; Kennedy 2007:6; Miller and Boxberger 1994:271-272).

3.2. Methods

We use both ethnographic and archaeological data to evaluate how territorial and tenurial claims may have operated within the social networks of the ancestral Coast Salish, and in particular how they are manifest in the movement of obsidian toolstone. The ethnographic data are used to illustrate the means through which obsidian could have moved intra and inter-regionally and to infer the directionality of social networks between ancestral Coast Salish groups and between ancestral Coast Salish and their neighbours. The archaeological data provide the geographic locations with which to infer ancestral networks, the potential directionality of those networks, and time depth estimates (Figures 10-20; Tables 4-8).

To compare the spatial distribution of sourced obsidian, we generated maps showing the general locations of the archaeological sites from which obsidian artifacts were recovered and the approximate locations of the associated toolstone sources. To address our hypotheses regarding the spatial distribution of sourced material within the Salish Sea region specifically, we organized and compared our data based on the four linguistic groups of the Coast Salish: Northern Coast Salish (NCS); Central Coast Salish (CCS); Southern and Southwestern Coast Salish (SCS); and a sub-group of Central Coast Salish, the lower Fraser Valley Central Coast Salish (FV). We defined the latter as a sub-group because of its socioeconomic importance noted in the ethnographic (Barnett 1955), historic (Maclachlan 1998), and archaeological records (Clark 2010; Lepofsky et al. 2005, 2009; Morin 2012). In particular, the Fraser River was an important corridor that moved resources, people, and information throughout the Salishan world (Lepofsky et al. 2009; Mitchell 1971; Morin 2012; Suttles 2000).

47 Organizing the obsidian data into these four groups was based on the assumption that relatively settled populations with recognized territories have been in place throughout the Salish Sea region for at least the last 5,000 years (e.g., Coupland et al. 2016; Mitchell 1990; Ritchie et al. 2016). In order for people to gain access to resources, including toolstone, which were not locally available, they required social networks to reach beyond the local (Collins 1979). We use the distribution of sourced obsidian as a proxy for this networking strategy and to illuminate both the extent and direction of those networks. In our analysis, we assume that a distance-decay model is the material correlate of down-the-line trade. Obsidian toolstone sources are organized into three groups: northern (B.C. sources north of the Salish Sea region); southern (American sources south of the Salish Sea from Washington, Oregon, and Idaho states); and Nch’kay (also known as Garibaldi, the only known source of obsidian within the Salish Sea region).

The archaeological data for this study were compiled from published literature, and unpublished cultural resource management reports and graduate theses that reported on obsidian artifacts recovered from sites along the coasts of B.C. and Washington state. Unpublished reports for B.C. were collated from the Provincial Archaeology Branch’s Remote Access to Archaeological Data and the Provincial Archaeological Report Library, based in Victoria, B.C. The American reports were gathered from Washington state’s Department of Archaeology and Historic Preservation’s Washington Information System for Architectural and Archaeological Records Data. All radiocarbon dates used for this study were calibrated with the online program Calib 7.1 (Stuiver et al. 2017). Given the relatively small sample of obsidian directly associated with dated contexts, we organized dated sites into a series of maps representing 1,000-year blocks to illustrate the availability of sources over time.

In addition to data gleaned from published and unpublished sources, we present the elemental results of an assemblage of 57 obsidian artifacts (see Appendix C) recovered from various sites in the northern Salish Sea during research conducted by Simon Fraser University’s (SFU) Department of Archaeology. We also include a reanalysis of 37 obsidian artifacts recovered by SFU and University of British Columbia (UBC) researchers from the Scowlitz site (Blake 2004; see Table 4, DhRl-16, ID# 92) located in the lower Fraser Valley.

48 Our analysis is based on a comparison of the number of obsidian artifacts and obsidian toolstone sources represented within each of the four cultural groups within the Salish Sea region. The obsidian in our dataset recovered from sites outside of the Salish Sea region is used only to graphically emphasize the spatial and temporal distributions of the various obsidian sources referred to in this paper. We were not able to meaningfully compare artifact types because in most cases, only some portion of a given assemblage was chemically analyzed. Furthermore, flakes rather than formed tools or cores make up the majority of artifacts in our dataset.

We did not include in this study obsidian artifacts recovered from sites on the coasts of B.C. and Washington state that were not sourced to a particular flow, and artifacts that could only be correlated to currently unknown sources. An exception to the latter are artifacts correlated to the source known as Central Coast Unknown B (CCUB) and a single artifact correlated to Unknown Type A (UTA). Although the location of CCUB is currently unknown, its distribution mirrors that of the Kingcome source suggesting both were formed from volcanic events associated with Mount Silverthrone above Kingcome Inlet. The single artifact correlated to UTA is also likely associated with Mount Silverthrone and was included in our analysis as Kingcome obsidian.

The artifacts that make up our dataset were made from obsidians of varying quality with respect to fracture predictability (Rorabaugh and McNabb 2014:375). Fracture predictability is a function of the impurities or imperfections present in the toolstone. Brittle, homogenous, and isotropic toolstone, such as obsidian, is best for flintknapping because it fractures predictably and reliably allowing for the relatively easy production of flakes that can be worked into formal tools (Andrefsky 2005:24; Cotterell and Kamminga 1987:677; Cotterell et al. 1985:205; Odell 2003:43). We assume that obsidian toolstone with phenocrysts (Nch’kay, Kingcome, CCUB, and Washington toolstone) is of lower quality than the more homogenous and isotropic obsidian toolstone (e.g., Edziza, Anahim, and most of the Oregon toolstone). This is because phenocrysts, or other flaws in the stone, whether macroscopic or microscopic, can hamper stone tool production (e.g., Shackley 1998:1078; Cotterell and Kamminga 1987:679) by reducing the reliability and predictability of the fracture. We recognize that it cannot be definitively shown why a particular obsidian toolstone was chosen by ancestral groups; however, we contend that the quality of the toolstone as defined by its reliable and predictable fracture

49 likely played a role. We further contend that this characteristic of the obsidian toolstone found in the Salish Sea region is reflected in the spatial patterning we discuss below.

3.3. Results

A total of 1,577 sourced obsidian artifacts from 152 coastal, island, and riverine sites located along the coasts of B.C. and Washington state were used for this study (Tables 4-7; Figures 10-13). Thirty-seven obsidian toolstone deposits located in B.C. (n=8), and Washington (n=3), Oregon (n=25), and Idaho (n=1) states are represented in the dataset (Tables 5-7; Figures 11-15). Of the total sample, 676 artifacts were recovered from dated contexts at 42 of the 152 sites (Tables 5 and 8; Figures 16-19). The 57 artifacts included in this study that were recovered by SFU’s Department of Archaeology were analyzed by the Northwest Research Obsidian Studies Laboratory and the Department of Archaeology’s pXRF Laboratory at SFU (see Appendix C). The reanalysis of the 37 obsidian artifacts recovered from the Scowlitz site (now known as Qithyil, DhRl-16) by researchers from SFU and UBC (Blake 2004) was conducted by Northwest Research Obsidian Studies Laboratory (2015).

50 Table 4: Sites and associated obsidian artifacts and toolstone sources (see Tables 5-7 for names of Washington, Oregon, and Idaho toolstone sources). Salish Sea sites: 38-82, 84-113, 118-126, 138, 142-152 (see Figures 10-13 for all site locations.

Toolstone sources

ID Site n Reference

Edziza Ilgachuz Anahim MacKenzie Kingcome CCUB Garibaldi Washington Oregon Idaho 1 GcTo-1, Bencke Point 4 4 Carlson 1994 2 GbTo-34, Kitandach 2 2 Carlson 1994 3 GbTo-18, Dodge Island 1 1 1 3 Carlson 1994 4 FlUa-4, Blue Jacket 3 3 Carlson 1994 Creek 5 FlUa-1, Skoglund’s 1 1 Carlson 1994 Landing 6 FeSr-7, Kimsquit 1 1 Carlson 1994 7 FeSr-5, Kimsquit 4 1 5 Carlson 1994 8 FeSr-4, Kimsquit 1 1 Carlson 1994 9 FcTe-4, Grant 1 11 3 1 16 Carlson 1994 Anchorage 10 FbSx-9, Troupe 1 1 Carlson 1994 Passage 11 FbSu-1, Cathedral 1 1 9 6 1 18 Carlson 1994 Point 12 FaTb-13, Campbell 1 1 Carlson 1994 Island 13 FaSu-2, Kwatna 2 14 5 21 Carlson 1994 14 FaSu-18, Kwatna 1 1 Carlson 1994 15 FaSu-10, Kwatna 9 1 10 Carlson 1994 16 ElTb-101, McNaughton 20 1 21 Carlson 1994 17 ElSx-1, Namu 2 109 6 17 1 194 Carlson 1994 5 18 EkTa-10, Hunter Island 1 1 Carlson 1994 19 EkSx-1, Kwakume 2 2 Carlson 1994 20 EfSq-22, Hopetown 1 1 Carlson 1994 21 EeSu-53, O’Connor 39 1 52 Carlson 1994 3 22 EeSp-6, Baker Island 1 1 Carlson 1994

Toolstone sources

ID Site n Reference

Edziza Ilgachuz Anahim MacKenzie Kingcome CCUB Garibaldi Washington Oregon Idaho 23 EeSp-48, Baker Island 1 1 Carlson 1994 24 EeSp-2, Insect Island 1 1 Carlson 1994 25 EeSp-17, Retreat 1 1 Carlson 1994 Passage 26 EeSp-121, Trivett 1 1 2 Carlson 1994 Island 27 EeSp-100, John Island 1 1 Carlson 1994 28 EeSo-19, Denham 1 1 2 Carlson 1994 Island 29 EeSo-1, Gilford Island 4 7 11 Carlson 1994 30 EdSq-7, Compton 4 4 Carlson 1994 Island 31 EdSp-41, Village Island 1 1 Carlson 1994 32 EdSo-34, Klaotsis 3 3 Carlson 1994 33 EdSo-22, Potts Lagoon 1 1 Carlson 1994 34 EdSn-7, Keecekitum 1 1 Carlson 1994 35 EdSn-38, Chatham 1 1 Carlson 1994 Channel 36 EdSn-35, Farquharson 2 1 3 Carlson 1994 Island 37 EcRq-1, Birkenhead 5 5 Witt et al. 2001 River 38 EbSh-36, Small Inlet 1 1 Toniello et al. 2016 39 EaSe-76, Cochrane 1 14 1 16 Springer et al. 2013 Bay 40 EaSe-67, Scott Point 1 1 Johnson and Lepofsky 2009 41 EaSe-2, Bliss Landing 1 1 1 3 Carlson 1994 42 EaSe-18 17 5 1 23 Springer et al. 2013 43 EaRu-5, Elaho 6 6 Reimer 2014 Rockshelter 44 DlSd-6, Scuttle Bay 2 2 Springer et al. 2013 45 DlSd-17, Willingdon 1 1 Springer et al. 2013 Beach

Toolstone sources

ID Site n Reference

Edziza Ilgachuz Anahim MacKenzie Kingcome CCUB Garibaldi Washington Oregon Idaho 46 DlSd-11, Tla’amin (IR1) 1 1 Springer et al. 2013 47 DlRt-9, Yelhi’xw 14 14 Reimer 2014 Rockshelter 48 DkSf-2, Millard Creek 8 8 Carlson 1994 49 DkSd-1, Grief Point 2 1 3 Springer et al. 2013 50 DkSc-15, Lang Bay 1 2 3 Springer et al. 2013 51 DkRs-6, Stawamus 11 11 Reimer 2014 52 DkRs-19, Stawamus 3 57 57 Ritchie and Sellers 2016 53 DkRr-7 1 1 Reimer 2014 54 DkRr-5 1 1 Reimer 2014 55 DkRr-4, Mamquam 14 14 Reimer 2014 Creek 56 DkRr-3, Lava Creek 1 1 Reimer 2014 57 DkRr-2, Gargoyles 5 5 Reimer 2014 Workshop 58 DkRr-1, Columnar Peak 15 15 Reimer 2014 Gargoyles Workshop 59 DkRn-1, 3 3 Ferguson et al. 2011 Xaxtsa/Douglas Lake (IR6) 60 DkRl-1, Shovel Creek 4 4 Dodd and Zibauer 2011 61 DjSf-14, Tsable River 17 17 Carlson 1994; Bridge Mitchell 1974a 62 DjSf-13, Buckley Bay 11 4 4 19 Carlson 1994; Mason and Hoffmann 1998; Mitchell 1974a, 1974b 63 DjSc-5, Maple Bay 1 1 Springer et al. 2013 64 DjSb-3, Tucker Bay 1 1 Springer et al. 2013 65 DjRt-12, Porteau Cove 1 1 Hall et al. 2007 Rockshelter

Toolstone sources

ID Site n Reference

Edziza Ilgachuz Anahim MacKenzie Kingcome CCUB Garibaldi Washington Oregon Idaho 66 DjRi-3, Milliken 2 1 9 12 Carlson 1994; James et al. 1996 67 DiSh-6, Oyster River 20 20 Mason et al. 1998 68 DiSh-17, Elsie Lake 4 8 8 20 Forgeng et al. 2007 Reservoir 69 DiRu-60, Gambier 3 3 Pratt et al. 1998 Island 70 DiRu-56, Gambier 12 12 Pratt et al. 1998 Island 71 DiRu-4, Gambier Island 1 1 Howe 2006 72 DiRu-19 6 6 Reimer 2014 73 DiRu-15 6 6 Reimer 2014 74 DiRt-11, Halkett Bay 6 6 Reimer 2012a 75 DjRr-2, Indian 1 10 10 Ritchie and Sellers 2016 76 DiRq-5, Coquitlam Lake 5 5 Brown and Oakes 2011 77 DiRq-34, Coquitlam 2 2 Brown and Oakes Lake 2011 78 DiRq-27, Coquitlam 1 1 Brown and Oakes Lake 2011 79 DiRq-24, Coquitlam 1 1 Brown and Oakes Lake 2011 80 DiRq-21, Coquitlam 10 10 Brown and Oakes Lake 2011 81 DiRq-2, Coquitlam Lake 1 1 Brown and Oakes 2011 82 DiRq-14, Coquitlam 4 4 Brown and Oakes Lake 2011 83 DhSe-24, Shoemaker 40 3 3 45 McMillan and St. Bay Claire 1982 84 DhRt-65, Locarno 1 17 18 Carlson 1994; Collard Beach et al. 2006; Reimer 2012a, 2014

Toolstone sources

ID Site n Reference

Edziza Ilgachuz Anahim MacKenzie Kingcome CCUB Garibaldi Washington Oregon Idaho 85 DhRs-1, Marpole 2 13 13 28 Carlson 1994 86 DhRr-8, Cates Park 8 5 13 Carlson 1994; James et al. 1996 87 DhRr-6, Belcarra Park 2 2 Carlson 1994 88 DhRr-20, Berry Point 1 1 Reimer 2014 89 DhRr-18, Cove Cliff 1 1 Reimer 2014 90 DhRq-22, Park Farm 1 1 5 2 9 Kristensen et al. 2009; Spurgeon 1994 91 DhRp-17, Port 3 1 4 Carmichael et al. Hammond 2001 92 DhRl-16, Scowlitz 1 1 34 36 Blake 2004; Northwest Research Obsidian Studies Laboratory 2015 93 DhRk-8, Maurer 9 9 Carlson 1994 94 DgRw-4, Senewélets 1 2 8 5 16 Burley 1989; Carlson 1994 95 DgRv-3, Dionisio Point 1 1 Grier and Angelbeck 2007 96 DgRs-1, Beach Grove 2 2 Brolly et al. 1996 97 DgRr-2, St. Mungo 4 29 33 Carlson 1994; Commisso et al. 2015 98 DgRr-1, Crescent 4 4 Carlson 1994; Beach Conaty and Curtin 1984; Percy 1974 99 DfRu-8, Helen Point 2 105 107 Carlson 1994 100 DfRu-24, Georgeson 12 12 Carlson 1994 Bay6 101 DgRs-147, Whalen 3 3 Carlson 1994 Farm 102 DeRw-18, Somenos 9 9 McLay et al. 2009 Creek 103 DeRt-2, Pender Canal 1 1 Carlson 1994 104 DeRt-1, Pender Canal 9 9 Carlson 1994

Toolstone sources

ID Site n Reference

Edziza Ilgachuz Anahim MacKenzie Kingcome CCUB Garibaldi Washington Oregon Idaho 105 DcRt-13, Bowker Creek 4 4 Carlson 1994 106 DcRt-10, Willows 21 21 Carlson 1994 Beach 107 45SJ1, Cattle Point 8 8 Carlson 1994 108 45WH17, Semiahmoo 1 1 Carlson 1994 Spit 109 45WH34, Ferndale 1 1 Carlson 1994 110 45WH47, Blake Site 1 1 Carlson 1994 111 45KI5, Jokumsen 10 10 Carlson 1994 112 45SK41, Soler 1 1 Carlson 1994 113 45KI435, Mule Spring 2 15 17 Vaughan 2010 Camp 114 45LE415, Beech Creek 75 35 110 Mack et al. 2010; Site Vaughan 2010 115 45PI408, Sunrise Ridge 1 39 40 Vaughan 2010 Site 116 45PI406, Tipsoo Lakes 5 28 33 Vaughan 2010 Site 117 45LE2858, Packwood 1 1 Vaughan 2010 Lake Outlet 118 45LE803 1 1 Rooke et al. 2011 119 45LE116 5 5 Rooke et al. 2011 120 45KI464, Tolt River Site 22 22 Blukis Onat et al. 2000 121 45LE222 Judd peak 3 2 5 Daugherty et al. 1987 Rockshelter (North) 122 45LE425 1 1 Wessen 2000 123 45LE426 1 1 Wessen 2000 124 45LE428 1 1 Wessen 2000 125 45LE439 2 2 Wessen 2000 126 45LE457 1 1 Wessen 2000 127 45SA303, Camp Lithic 2 2 Mack and Baucom Scatter Site 1992

Toolstone sources

ID Site n Reference

Edziza Ilgachuz Anahim MacKenzie Kingcome CCUB Garibaldi Washington Oregon Idaho 128 45SA117, Falls Creek 9 9 Tolfree 1989 Site 129 Wind River Ranger 10 10 Hughes 1990 Station9 130 45PC222, Mill Creek 1 1 Nakonechny 2015 Isolate 131 45PC101, North 1 1 Nakonechny 2015 Nemah River Bridge 132 45PC175, Forks Creek 17 17 Nakonechny 2015 Willapa River Site 133 45PC510, Oysterville- 1 1 Nakonechny 2015 Kemmer Collection 134 45PC195, Jenkins 1 1 Nakonechny 2015 Farm North River Prehistoric Site 135 45PC183, Swiss 7 7 Nakonechny 2015 Scatter Willapa River Site 136 45PC180, Martin Dairy 1 1 Nakonechny 2015 Willapa River Site 137 45PC212, Carruthers 1 1 Nakonechny 2015 Slough Willapa Bay Prehistoric Intertidal Site 138 45GH15, the Minard 1 1 Nakonechny 2015 Site 139 Menlo-Niemcziek 17 17 Nakonechny 2015 Collection11 140 Upper Willapa Valley- 4 4 Nakonechny 2015 Kaech Collection12 141 Sandy Point-Rubey 1 1 Nakonechny 2015 Collection13 142 45KI428, West Point 4 4 Larson and Lewarch Shell Midden Site 1995

Toolstone sources

ID Site n Reference

Edziza Ilgachuz Anahim MacKenzie Kingcome CCUB Garibaldi Washington Oregon Idaho 143 45SJ169, Decatur 2 2 Walker et al. 2003 Head 144 DhRp-96, Fort Langley 2 2 James et al. 1996 145 DgRn-23, Hatzic Rock 6 6 James et al. 1996 146 DiRj-5, Hunter Creek 1 1 James et al. 1996 147 DjRi-5, Esilao 6 6 James et al. 1996 148 DhRl-29, Tapadera 36 36 Reimer 2012b Estates 149 SE Lasqueti Island14, 1 1 Northwest Research Rezansoff Collection Obsidian Studies Laboratory 2016 150 EaSe-71, Hare Point 1 1 Northwest Research Obsidian Studies Laboratory 2016 151 DgRr-6, Glenrose 26 15 41 Commisso et al. 2015 Cannery 152 DhRq-21, Pitt River Site 2 2 Northwest Research Obsidian Studies Laboratory 2017 n 15 12 222 8 16 5 29 89 629 1 1577 8 9 5 7 Note: 1KlTb-10 in Carlson 1994; 2EfSp-2 in Carlson 1994; 3EeSu-1 in Carlson 1994; 4DhSe-7 in Carlson 1994, 5DhR56 in Carlson 1994; 6Tolan’s Beach in Carlson 1994; 7DfRs-3 in Carlson 1994 (portion of site located in USA, later amalgamated with portions on the Canadian side under DgRs-14); 845LE285 source identified but artifact amounts not given, an arbitrary number of one was used here; 9,10,11,12,13,14approximate locations on map based on general provenience of collected artifacts.

58 Table 5: Toolstone sources with numbers of correlated artifacts and sites, and numbers of dated artifacts. No. of dated artifacts in 1,000 (K) year blocks (cal.

BP)

No. of No. of

ID Toolstone sources Location n

artifacts sites 1K

2K 3K 4K 5K 6K 7K 8K 9K 10K

------

100 1K 2K 3K 4K 5K 6K 7K 8K 9K 1 Edziza B.C. 15 9 1 1 2 Ilgachuz B.C. 12 6 1 5 1 7 3 Anahim Peak B.C. 222 21 6 42 10 6 16 45 2 1 1 129 4 MacKenzie B.C. 88 10 1 7 2 5 16 29 2 1 63 5 Kingcome B.C. 169 21 1 27 63 2 2 8 2 8 113 6 Nch’kay (Garibaldi) B.C. 297 44 13 58 39 1 7 71 189 7 Elk Pass WA 79 5 0 8 Indian Rock WA 7 2 0 9 Bickleton Ridge WA 3 2 0 10 Clackamas River* OR 1 1 0 11 Inman Creek OR 8 3 0 12 Obsidian Cliffs OR 185 38 6 10 7 17 9 49 13 Newberry OR 115 30 3 7 3 22 7 3 45 14 Big Obsidian Flow OR 2 2 0 15 Cougar Mountain OR 3 3 1 1 16 Quartz Mountain OR 16 3 1 1 17 Sqaw Ridge OR 6 3 1 1 18 Silder Lake/Sycan Ridge OR 2 2 0 19 Spodue Mountain OR 1 1 0 20 Drews Creek/Butcher Flat OR 2 1 0 21 Glass Buttes 1 OR 127 30 3 10 5 1 19 22 Glass Buttes 3 OR 10 4 0 23 Horse Mountain OR 1 1 0 24 Tank Creek OR 1 1 0 25 Rim Rock Spring OR 1 1 0 26 Burns OR 23 20 20 20 27 Wolf Creek OR 3 3 0 28 Whitewater Ridge OR 111 33 1 1 2 2 8 2 16 29 Tule Spring OR 1 1 0 30 Eldorado OR 1 1 0 31 Indian Creek OR 1 1 0 32 Ebell Creek OR 1 1 1 1

59 No. of dated artifacts in 1,000 (K) year blocks (cal.

BP)

No. of No. of

ID Toolstone sources Location n

artifacts sites 1K

2K 3K 4K 5K 6K 7K 8K 9K 10K

------

100 1K 2K 3K 4K 5K 6K 7K 8K 9K 33 Gregory Creek OR 3 3 1 1 34 Timber Butte ID 1 1 0 35 Paulina Lake OR 4 2 0 N/A Central Coast Unknown B B.C. 55 20 5 15 20 (CCUB)** N/A Unknown Type A (UTA)** B.C. 1 1 0 n 1577 330 34 16 17 58 66 88 4 73 11 1 676 9 2 Note: *Not a specific source, cobbles found along course of Clackamas river; **Location unknown.

Table 6: Sites with obsidian artifacts from Washington and Idaho toolstone sources. Washington toolstone sources Idaho toolstone source ID Code Elk Indian Rock Bickleton Timber Butte n Pass Ridge 66 DjRi-3 1 1 113 45KI435 2 2 114 45LE415 70 5 75 115 45PI408 1 1 116 45PI406 3 2 5 117 45LE285 1 1 121 45LE222 3 3 122 45LE425 1 1 149 N/A 1 1 n 79 7 3 1 90

60 Table 7: Obsidian artifacts from Oregon Toolstone sources.

Toolstone sources

ID n

18 22

13 16

9 20

8 10

3 12

2 25

5 17

11 26

14 15

CR OC IC EC E InC WR GB3 WC TS GC RRS B N BOF SR CMT QMT GB1 TC HMT SLK SMT DC PL 9 1 1 17 1 1 38 1 1 39 1 1 42 1 1 45 1 1 50 1 1 2 63 1 1 66 1 2 3 3 9 67 20 20 83 1 1 2 85 3 1 3 3 3 13 86 2 3 5 90 1 1 2 91 1 1 92 1 4 1 14 4 1 2 1 5 1 34 93 8 1 9 94 1 1 3 5 97 3 1 16 2 1 3 3 29 98 1 2 1 4 99 45 10 37 13 105 100 6 1 1 1 1 2 12 101 2 1 3 102 1 3 1 4 9 103 1 1 104 1 3 1 4 9 105 3 1 4 106 1 2 18 21 107 3 2 3 8 108 1 1 109 1 1 110 1 1 111 2 2 4 2 10 112 1 1 113 14 1 15

Toolstone sources

ID 19 n

18 22

13 16

9 20

8 10

3 12

2 25

5 17

11 26

14 15

CR OC IC EC E InC WR GB3 WC TS GC RRS B N BOF SR CMT QMT GB1 TC HMT SLK SMT DC PL 114 8 11 1 1 14 35 115 17 1 6 14 1 39 116 8 6 14 28 118 1 1 119 3 1 1 5 120 18 1 1 1 1 22 121 2 2 123 1 1 124 1 1 125 1 1 2 126 1 1 127 2 2 128 9 9 129 7 2 1 10 130 1 1 131 1 1 132 6 2 4 5 17 133 1 1 134 1 1 135 4 1 2 7 136 1 1 137 1 1 138 1 1 139 1 3 2 3 1 1 1 3 1 1 17 140 1 1 2 4 141 1 1 142 1 2 1 4 143 1 1 2 144 2 2 145 2 3 1 6 146 1 1 147 5 1 6 148 36 36 150 1 1 151 1 5 4 4 1 15 152 2 2

Toolstone sources

ID 19 n

18 22

13 16

9 20

8 10

3 12

2 25

5 17

11 26

14 15

CR OC IC EC E InC WR GB3 WC TS GC RRS B N BOF SR CMT QMT GB1 TC HMT SLK SMT DC PL n 1 185 8 1 1 1 111 10 3 1 3 1 23 115 2 6 3 16 127 1 1 2 1 2 4 629 Note: 1See Table 4 for site codes and names; 2Clackamas River; 3Obsidian Cliffs (Three Sisters in Blake 2004; Carlson 1994; James et al. 1996); 4Inman Creek; 5Ebell Creek (Baker in Carlson 1994); 6Eldorado; 7Indian Creek; 8Whitewater Ridge (John Day in Carlson 1994); 9Whitewater Ridge/Glass Buttes 3; 10Wolf Creek; 11Tule Spring; 12Gregory Creek; 13Rim Rock Spring; 14Burns; 15Newberry; 16Big Obsidian Flow; 17Sqaw Ridge; 18Cougar Mountain; 19Quartz Mountain; 20Glass Buttes 1; 21Tank Creek; 22Horse Mountain; 23Silver Lake/Sycan Marsh; 24Spodue Mountain; 25Drews Creek; 26Paulina Lake.

63 Table 8: Chronometric and inferred dates associated with obsidian artifacts. Conventional age Calibrated ID Code Lab# Material Source (BP) 2σ (BP) 17 ElSx-1 GAK-3122 Charcoal 680+/-90 768-522 Carlson 1996 17 ElSx-1 WSU-1942 Charcoal 1405+/-120 1554-1060 Carlson 1996 17 ElSx-1 GAK-3119 Charcoal 2440+/-100 2751-2315 Carlson 1996 17 ElSx-1 SFU-17 Charcoal 3280+/-100 3728-3324 Carlson 1996 17 ElSx-1 GAK-2717 Charcoal 4290+/-120 5088-4527 Carlson 1996 17 ElSx-1 WAT-451 Charcoal 5170+/-90 6185-5717 Carlson 1996 17 ElSx-1 WSU-1941 Charcoal 6060+/-100 7171-6672 Carlson 1996 17 ElSx-1 GAK-3120 Charcoal 7800+/-200 9135-8185 Carlson 1996 17 ElSx-1 WAT-516 Charcoal 8570+/-90 9794-9405 Carlson 1996 21 EeSu-5 GAK-3901 Charcoal 2540+/-90 2776-2358 Chapman 1974 36 EdSn-35 N/A N/A 6250+/-110 7343-6990 Archaeology Branch 2015a 37 EcRq-1 Beta-147521 Charred material 2520+/-110 2796-2343 Witt et al. 2001 39 EaSe-76 Beta-261183 Charcoal 2180+/-50 2330-2040 Springer et al. 2013 39 EaSe-76 Beta 319135 Bone collagen 2630+/-30* 2046-1741 Springer et al. 2013 39 EaSe-76 Beta-314096 Charred material 1300+/-30 1290-1180 Springer et al. 2013 39 EaSe-76 Beta-314098 Charred material 830+/-30 790-690 Springer et al. 2013 42 EaSe-18 Beta-342741 Shell 3350+/-30** 2838-2502 Springer et al. 2013 43 EaRu-5 N/A N/A 1210+/-35 1189-1057 Reimer 2014 43 EaRu-5 N/A N/A 75+/-35 142-23 Reimer 2014 47 DlRt-9 Beta-227280 Charcoal 1390+/-40 1375-1263 Reimer 2006, 2014 48 DkSf-2 N/A N/A 8300+/-200 9635-8644 Archaeology Branch 2015b 50 DkSc-15 263330 Charred material 2690+/-40 2860-2750 Springer et al. 2013 51 DkRs-6 N/A N/A 4000+/-60 4629-4285 Reimer 2014 51 DkRs-6 N/A N/A 2270+/-60 2379-2119 Reimer 2014 52 DkRs-19 N/A Charcoal 6761+/-32 7666-7576 Ritchie and Sellers 2016 55 DkRr-4 N/A N/A 7130+/-40 8019-7922 Reimer 2014 58 DkRr-1 N/A N/A 2850+/-40 3076-2854 Reimer 2014 62 DjSf-13 GAK-7347 Shell 2640+/-90** 2110-1536 Mitchell 1974 62 DjSf-13 GAK-7348 Shell 2770+/-90** 2288-1704 Mitchell 1974 66 DjRi-3 S-47 Charcoal 8150+/-310 9790-8346 Borden 1961; Mitchell and Pokotylo 1996 67 DiSh-6 Beta-108539 Charcoal 2630+/-50 2860-2699 Mason et al. 1998 72 DiRu-19 N/A N/A 2050+/-90 2185-1822 Reimer 2014 73 DiRu-15 N/A N/A 2690+/-70 2965-2714 Reimer 2014 74 DiRt-11 N/A N/A 1190+/-120 1330-906 Reimer 2014 75 DjRr-2 N/A N/A 640+/-26 607-556 Ritchie and Sellers 2016 75 DjRr-2 N/A N/A 891+/-23 804-736 Ritchie and Sellers 2016 75 DjRr-2 N/A N/A 920+/-26 921-782 Ritchie and Sellers 2016

64 Conventional age Calibrated ID Code Lab# Material Source (BP) 2σ (BP) 83 DhSe-2 GAK-5107 Charcoal 1730+/-80 1828-1516 McMillan and St. Claire 1982 83 DhSe-2 GAK-5104 Charcoal 2860+/-90 3218-2776 McMillan and St. Claire 1982 84 DhRt-6 Beta-69094 Charred material 1630+/-80 1708-1354 Brolly and Muir 1993 85 DhRs-1 N/A Charcoal 1510+/-90 1572-1280 Archaeology Branch 2015c 85 DhRs-1 N/A Charcoal 2350+/-60 2540-2301 Archaeology Branch 2015c 90 DhRq-22 Beta-80773 Charcoal 300+/-50 488-284 Spurgeon 1994 91 DhRp-17 Beta-153919 Charcoal 1600+/-60 1617-1354 Carmichael et al. 2001 92 DhRl-16 WSU-5050 Charcoal 1310+/-45 1309-1173 Lepofsky et al. 2000a, 2000b 92 DhRl-16 WSU-5019 Charcoal 1080+/-70 1179-901 Lepofsky et al. 2000a, 2000b 92 DhRl-16 Beta-56217 Charcoal 1190+/-70 1268-969 Lepofsky et al. 2000a, 2000b 92 DhRl-16 WSU-5052 Charcoal 1850+/-50 1893-1692 Lepofsky et al. 2000a, 2000b 92 DhRl-16 CAMS-61998 Charcoal 2250+/-70 2379-2043 Lepofsky et al. 2000a, 2000b 93 DhRk-8 GAK-4919 N/A 4220+/-100 4982-4511 Lepofsky et al. 2009 94 DgRw-4 GAK-2754 Charcoal 1679+/-90 1815-1394 Archaeology Branch 2015d 96 DgRs-1 Beta-83117 N/A 3900+/-60 4445-4152 Brolly et al. 1996 97 DgRr-2 WSU-2840 Charcoal 3380+/-70 3735-3459 Ham et al. 1986 99 DfRu-8 GAK-3202 Charcoal 700+/-110 801-513 CARD 2016 99 DfRu-8 GAK-3204 Charcoal 640+/-90 728-512 CARD 2016 99 DfRu-8 GAK-4935 Charcoal 1370+/-85 1415-1069 CARD 2016 99 DfRu-8 GAK-4936 Charcoal 1120+/-100 1275-901 CARD 2016 99 DfRu-8 GAK-4937 Charcoal 2110+/-105 2338-1877 CARD 2016 99 DfRu-8 SFU-623 Charcoal 3690+/-70 4184-3843 CARD 2016 99 DfRu-8 GAK-3201 Charcoal 3980+/-130 4828-4141 CARD 2016 102 DeRw-18 Beta-231891 Charcoal 1270+/-40 1288-1172 McLay et al. 2009 103 DeRt-2 RIDDL-105 Human bone 4070+/-150*** 4092-3295 Carlson 1986 110 45WH47 Inferred Diagnostics Locarno**** 3500-2400/2000 Reed et al. 2010 121 45LE222 WSU-3562 Charcoal 780+/-80 834-636 Daugherty et al. 1987 124 45LE428 Beta-130836 Charred material 520+/-60 653-484 Wessen 2000 142 45KI428 Beta-58023 Charcoal 2940+/-70 3257-2919 Larson and Lewarch 1995 143 45SJ169 Beta-170654 Charred material 2350+/-60 2540-2301 Walker et al. 2003 145 DgRn-23 SFU-888 N/A 4490+/-70 5315-4955 Lepofsky et al. 2009 145 DgRn-23 Beta-77758 N/A 4840+/-110 5759-5316 Lepofsky et al. 2009 147 DjRi-5 M-1511 N/A 570+/-100 704-431 Borden 1983 150 EaSe-71 Beta-285595 Shell 1140+/-40** 488-253 Springer et al. 2013 Note: *A marine correction of 421+/-54 with an assumed 90% marine-based protein diet (Chisholm 1986) was applied to the bone date using Calib 7.1 (Stuiver et al. 2017). **A marine correction of 421+/-54 was applied to shell dates using Calib 7.1 (Stuiver et al. 2017). ***A marine correction of 408+/-59 with an assumed 90% marine-based protein diet (Chisolm 1986) was applied to the bone date using Calib 7.1 (Stuiver et al. 2017). ****Locarno Phase 3500-2400/2000 years BP (Clark 2010; Williams 2013)

65 3.3.1. Spatial Distribution of Sourced Toolstone within the Salish Sea Region

The mapped sites and sources in our dataset show a wide but uneven distribution of obsidian toolstone in the Salish Sea region (Figures 11-13). As expected, the lower quality obsidians have relatively restricted distributions (Figures 12 and 13a). Also, as expected, higher quality toolstone from the north and south of the region follow drop-off patterns moving away from sources (Figures 11a, 11c, and 13b). However, the extent of each varies, with Oregon toolstone showing the widest distribution and Washington and Nch’kay toolstone, the most restricted.

Obsidian toolstone from B.C. sources (n=858) accounts for 54.4% of our total sample (Table 4, Figures 11-12). Of this, 419 (48.8% of B.C. sample; 26.5% of total sample) artifacts were recovered from sites within the Salish Sea region. Three of the seven known B.C. sources represent the majority of B.C. obsidian in our sample found in the Salish Sea region: Anahim (n=37, 8.8%); Kingcome (n=63, 15.0%); and Nch’kay (n=297, 70.8%). Although Nch’kay obsidian is the clear majority, its distribution is largely restricted to sites on the mainland side of the Salish Sea (Figure 12c). Nch’kay is the only known source for obsidian within the Salish Sea region; however, the circumscribed nature of its distribution suggests that it was either an undesirable toolstone due to its poor quality (e.g., Carlson 1994:323) and seasonal availability as an alpine source, or that access to the source was controlled (e.g., Reimer 2012a). Obsidian toolstone from Anahim and Kingcome have much wider distributions on the Pacific Northwest Coast of B.C. compared with Nch’kay (Figures 11c and 12b), but within the Salish Sea specifically, they are largely confined to the northern end of the region. This is likely due, in part, to drop-off with increased distance from the source but it may also be a reflection

Figure 11: General locations of archaeological sites (circles) with obsidian artifacts sourced to Mount Edziza (11a), Ilgachuz (11b), Anahim Peak (11c), and MacKenzie (11d). ID numbers provide additional information in Table 4 (archaeological sites) and Table 5 (toolstone sources). The dotted line represents the approximate boundary of Coast Salish languages (maps generated from ESRI basemap catalogue using ArcMap 10.3).

Figure 12: General locations of archaeological sites (circles) with obsidian artifacts sourced to Central Coast Unknown B (12a), Kingcome/UTA (12b), and Nch’kay (12c). ID numbers provide additional information in Table 4 (archaeological sites) and Table 5 (toolstone sources). The dotted line represents the approximate boundary of Coast Salish languages (maps generated from ESRI basemap catalogue using ArcMap 10.3).

Figure 13: General locations of archaeological sites (circles) with obsidian artifacts sourced to toolstone deposits located in Washington State (13a), and Oregon and Idaho States (13b). ID numbers provide additional information in Tables 4, 6, and 7 (archaeological sites) and Table 5 (toolstone sources). The dotted line represents the approximate boundary of Coast Salish languages (maps generated from ESRI basemap catalogue using ArcMap 10.

69 of the lesser quality of the toolstone from these sources, especially Kingcome, when compared with toolstone from Oregon.

Oregon obsidian (n=629) accounts for 39.8% of the total sample and is found as far north as the Central Coast of B.C. (Figure 13). Within the Salish Sea region specifically, Oregon sources represent 25 of the 37 known sources (including CCUB and UTA) associated with archaeological sites in the region. Four of the 25 sources provided 89.8% (n=405) of the Oregon sample found at Salish Sea sites (n=451): Obsidian Cliffs (n=116, 25.7%); Whitewater Ridge (n=86, 19.1%); Newberry (including the four pieces sourced to Paulina Lake, n=78 pieces, 17.3%); and Glass Buttes (Glass Buttes 1, n=115 pieces, Glass Buttes 3, n=10 pieces, n=125 pieces, 27.7%). The wide distribution of these four sources into B.C. has been described by other researchers as a consequence of the importance of Oregon toolstone in “a regional exchange system that extended well into British Columbia” (Connolly et al. 2015:185). We agree with this position and further suggest that these four sources were likely preferred because the toolstone was of a higher quality, especially in comparison to the locally available Nch’kay obsidian and the geographically closer Washington sources to the south, and most B.C. sources north of the region (e.g., Carlson 1994:323).

3.3.2. Spatial Distribution of Sourced Toolstone by Cultural Group within the Salish Sea Region

Our analysis of the distribution of obsidian by the four cultural groups within the Salish Sea region indicates only some support for the distance-decay model. Oregon obsidian follows the expected pattern of declining total numbers moving northward; however, the number of Oregon sources represented in the Salish Sea archaeological record varies from this pattern (Figures 14 and 15). Specifically, there are more Oregon sources associated with the FV group (n=13) of sites than with any of the other groups (NCS, n=6; CCS, n=6; SCS, n=9). This is true despite the fact that the total CCS sample of obsidian (n=407), and Oregon obsidian specifically (n=191), is greater than the other three groups (NCS: total n=142, Oregon n=28; FV: total n=256, Oregon n=165; SCS: total n=73, Oregon n=67; Figure 15).

The southward pattern also deviates from the distance-decay null model. In particular, there is an almost complete drop-off of northern obsidian among the CCS and

70 FV groups with none found at SCS sites (Figure 15). This suggests some kind of socioeconomic barrier to southward movement. In the case of Kingcome and CCUB obsidians (Figures 12a, 12b, and 15), we hypothesize that their truncated southward movement is related to their lower quality (i.e., the presence of phenocrysts; Carlson 1994:323). However, the relative absence of the higher quality obsidians (i.e., Anahim and Edziza) among the CCS and FV groups, and their total absence from the SCS group suggests that social factors were also at play.

For the Nch’kay and Washington obsidians, we suspect their poor quality and seasonal availability (Carlson 1994:323; Galm 1994:281; Rorabaugh and McNabb 2014:376, 382) were primary factors in their highly localized distributions (Figures 12c, 13a, and 15). Nch’kay is largely confined to CCS territory, and is almost entirely found on the mainland, close to the source (Figures 12c and 14). Reimer (2012a) has argued that the Nch’kay source was controlled and associated with spiritual power, which would affect its distribution. Interestingly, although overlapping social networks regionally connected the Central Coast Salish, there is little evidence for the presence of other obsidian toolstone at sites in close proximity to Nch’kay (Figure 14). Given this, as argued by Reimer (2012a), Nch’kay toolstone may have played a symbolic role in the expression of social and cultural capital among the ancestral Central Coast Salish-

Figure 14: General locations of archaeological sites with obsidian artifacts sourced to toolstone deposits south of the Salish Sea region (crosses), north of the Salish Sea region (black circles), north and south of the Salish Sea region (triangles), north and south of the Salish Sea region, and Nch’kay (stars), and Nch’kay (open circles). The ID numbers provide additional information in Table 4. The dotted lines are approximate representations of linguistic boundaries of Northern (NCS), Central (CCS), Southern (SCS), and Southwestern (SWCS) Coast Salish peoples (Suttles 1990) (map generated from ESRI basemap catalogue using ArcMap 10.3).

Figure 15: Percent obsidian abundance by source. Sources organized from north (Left) to south along the x-axes; four Coast Salish areas organized north (Top) to south on the y-axis. The graphs illustrate deviations from distance-decay null model. Numbers on top of bars represent total counts.

73 Squamish with direct access to the source. We suggest that the poor quality and seasonal accessibility of Nch’kay obsidian likely overrode its symbolic importance among communities with indirect access to the source. Similarly, we suspect the all but complete absence of Washington obsidian anywhere in the Salish sea region is also related to concerns with quality and accessibility (Galm 1994:281).

3.3.3. Temporal Distribution

Our assemblage includes 676 pieces from dated contexts with the majority (n=301) associated with two sites: Namu (n=194, ID 17) on the Central Coast of B.C.; and Helen Point (n=107, ID 99) on Mayne Island, part of the San Juan/Gulf Islands Archipelago in the Salish Sea region. Keeping this in mind, only provisional statements are possible regarding the intra and interregional movement of obsidian through time on the Pacific Northwest Coast in general and the Salish Sea region specifically.

Obsidian toolstone from the Anahim source in B.C. is the earliest known on the coast (Figure 16b). Oregon obsidian becomes available among the FV group in the Salish Sea region from at least 8,000 years BP (Figure 16a) but does not appear again in dated contexts until 6,000-5,000 years BP (Figure 17b). From 8,000-6,000 years BP dated toolstone on the coast is from B.C. sources exclusively (Figures 17c and 17d). The number of southern toolstone sources represented in the Salish Sea region steadily increases from 6,000-5,000 years BP, with the four main Oregon contributors—Obsidian Cliffs, Newberry, Whitewater Ridge, and Glass Buttes—all present by 5,000 years BP (Figures 17a and 17b, Figure 18, Table 5). The known B.C. sources, except for Edziza, are represented in the Salish Sea region by 5,000-4,000 years BP (Figure 17a). The earliest known use of Edziza obsidian in the region is between 2,000-1,000 years BP in the FV area. After approximately 600 years BP, obsidian toolstone all but drops out of the Salish Sea archaeological record.

Although limited by sample size, our temporal data are consistent with some form of down-the-line trade throughout the Holocene. For example, during the early Holocene, the great distances between the dated provenience of sourced artifacts and the provenance of the toolstone material (e.g., Figures 16a and 17b, Tables 5, 7, and 8 IDs 66 and [17], and 145 and [21], ~675 km between the Milliken site and the Sqaw Ridge source, and ~642 km between the Hatzic Rock site and Glass Buttes source,

74 respectively), suggest down-the-line trade through marriage ties was likely occurring. We suggest that these relationships formed the basis for more formalized social networks that continued to facilitate down-the-line trade in the mid to late Holocene when linguistic and cultural barriers became more prominent with increasing populations across the Salish Sea region (e.g., Ritchie et al. 2016).

After 600 years BP, the drop off in obsidian within the Salish Sea region suggests a combination of environmental and social processes restricted access to obsidian sources. The onset of the Little Ice Age (LIA) circa 750-100 years BP (Clague and Mathewes 1996; Smith and Desloges 2000), and the consequent increase in snowfall, could explain the reduced representation of high elevation obsidian (e.g., Nch’kay, Kingcome, and CCUB sources; Figure 19). However, this does not explain the concurrent reduction in frequency of low-elevation obsidian sources. Thus, there may also have been social reasons for the change in distribution, such as a shifted focus to other material goods (e.g., Carlson 1994:319). By at least the late 18th-century AD, the introduction of European metals and glass (e.g., Cole and Darling 1990:120; Lamb 1984:527, 581, 585, 606) all but ended the procurement of obsidian toolstone in the Salish Sea region.

Figure 16: General locations of archaeological sites (circles) with obsidian artifacts sourced to represented toolstone sources (stars), dating from 9,000-8,000 years BP (16a) and 10,000-9,000 years BP (16b). ID numbers provide additional information in Tables 4 and 8 (archaeological sites) and Table 5 (toolstone sources). The dotted line represents the approximate boundary of Coast Salish languages (maps generated from ESRI basemap catalogue using ArcMap 10.3).

Figure 17: General locations of archaeological sites (circles) with obsidian artifacts sourced to represented toolstone sources (stars), dating from 5,000-4,000 years BP (17a), 6,000-5,000 years BP (17b), 7,000- 6,000 years BP (17c), and 8,000-7,000 years BP (17d). ID numbers provide additional information in Tables 4 and 8 (archaeological sites) and Table 5 (toolstone sources). The dotted line represents the approximate boundary of Coast Salish languages (maps generated from ESRI basemap catalogue using ArcMap 10.3).

Figure 18: General locations of archaeological sites (circles) with obsidian artifacts sourced to represented toolstone sources (stars), dating from 1,000-100 years BP (18a), 2,000-1,000 years BP (18b), 3,000- 2,000 years BP (8c), and 4,000-3,000 years BP (18d). ID numbers provide additional information in Tables 4 and 8 (archaeological sites) and Table 5 (toolstone sources). The dotted line represents the approximate boundary of Coast Salish languages (maps generated from ESRI basemap catalogue using ArcMap 10.3).

Figure 19: Probability distribution of radiocarbon dated archaeological sites with obsidian artifacts that have been sourced to Nch’kay, Kingcome, and CCUB. Probability distributions based on calibrated 2σ range. Bars at bottom of figure represent approximate timelines for LIA and sociopolitical unrest that may have affected access to toolstone deposits. ID numbers (y-axis) provide additional information in Tables 4 and 8. Figure generated using Calib 7.1 (Stuiver et al. 2017) and Adobe Illustrator.

79 3.4. Discussion

The spatial and temporal patterning of obsidian toolstone in the Salish Sea region provides an opportunity to investigate ancestral Coast Salish social networks and explore their relationship with past territorial and tenurial claims. Our results show that while quality of toolstone played an important role in the movement of different types of obsidian, social factors must have contributed to the distribution seen archaeologically. This is suggested by the number of Oregon sources represented at FV sites and the relative absence of high-quality northern obsidian south of the NCS group. On the other hand, low-quality obsidians from Nch’kay and the Washington sources at Elk Pass and Indian Rock tend to have highly localized distributions, despite the extra-local reach of social networks.

We suggest that relatively higher numbers of Oregon sources found at FV sites reflects the existence of social networks with stronger ties to the Oregon obsidian trade along the Columbia River drainage (Sobel 2006, 2012; Figure 10). Obsidian traded east along the Columbia River through the major trading centre at the Dalles would have been moved into the mid and lower Fraser Valley through Interior Salishan networks by way of important interregional trading hubs in Washington state and B.C. at Kettle Falls, and Okanagan Falls and Kamloops, respectively (Stern 1998:641-643; Teit 1900:258- 259, 1909:494, 536, 1930:250-253; see Figure 10). Trade between Interior Salish and Coast Salish groups at the north end of the lower Fraser Valley also occurred (Teit 1930:254). Additional support for the lower Fraser Valley as a trade route for Oregon toolstone into the Salish Sea region is the discovery at site DhRl-29 (Figure 13b, Tables 4 and 7, ID 148) of a cache of 36 unmodified nodules sourced to the Glass Buttes source in Oregon (Figure 13b, Tables 5 and 7, ID [21]; Reimer 2012b). This is suggestive of FV communities having access to and curating Oregon toolstone in raw form for either future trade as a raw material or for reduction into blanks and tools for trade. This strategy of curating raw southern toolstone for trade has also been noted among Chinookan households along the lower Columbia River Valley (Sobel 2006, 2012), which would have had much more direct access to Oregon obsidian than households in the Fraser Valley.

The development of social networks that facilitated trade in the past is illustrated by overlaying historic Coast Salish marriage patterns onto the spatial distribution of

80 Oregon obsidian within the Salish Sea (Figure 20). This map of Central (Squamish, Musqueam, Nooksack, and Lummi) and Southern (Twana) Coast Salish marriage patterns shows the significant overlap of territories that occurs as a consequence of group exogamy. This overlap highlights how social networks, developed out of marriage ties, would have functioned as socioeconomic conduits for preferred resources from distant locales, including toolstone. Moreover, the map illustrates how FV communities could have used far-reaching networks to establish their role as primary traders in the distribution of Oregon obsidian through tenurial claims. The social networks that moved the obsidian would have been maintained by gifting between families tied by marriage (Suttles 1960) and through oratory, gifting, and performance during potlatches. Gifting between closely tied families and claims to access expressed at multihousehold events are acts of territoriality that communicate tenurial claims, claims that are realized through the social networks.

Obsidian has long played an important role in the daily lives of ancestral Pacific Northwest Coast populations. Toolstone from a variety of sources was widely distributed, and knowledge of, and access to, source locations existed for millennia. Among populations within the Salish Sea region, the relationship with obsidian began as early as 8000 years BP. The most likely mechanism that brought the obsidian into the region was down-the-line trade initially practiced through relationships formed through group exogamy and later through social networks. We have suggested that the presence of these networks can be inferred from the spatial and temporal distribution of obsidian

Figure 20: General locations of archaeological sites (crosses) in the Salish Sea region with obsidian artifacts sourced to toolstone deposits located in Oregon State. Historic Squamish, Musqueam, Nooksack, Lummi, and Twana marriage patterns derived from Kennedy (2007). Arrows show hypothesized routes through which obsidian toolstone was moved into the Salish Sea region via social networks formed through affinal relationships and kinship. ID numbers provide additional information in Tables 4 and 7 (map generated from ESRI basemap catalogue using ArcMap 10.3).

82 toolstone and through evidence that particular sources were preferred. Furthermore, social networks enabled the expression of territorial and tenurial claims as part of ongoing practices associated with gaining, maintaining, and legitimizing access to resources, including toolstone, beyond the local.

3.5. Conclusion

Social networks framed interaction for many ancestral communities at various cultural and spatial scales. Yet, despite the importance of social networks, their archaeological footprint can only be effectively inferred through widely traded material remains such as toolstone. When combined with ethnographic data, the spatial and temporal distributions of toolstone from distant locales can support inferences of network directionality and the strength of those networks across space and through time. Ethnography necessarily underpins suppositions about the existence of ancestral networks since archaeological materials alone—such as toolstone—cannot tell us much about the social framework within which materials were moved from source to site.

Our use of ethnography and the distribution of sourced obsidian toolstone within the Salish Sea region allowed us to examine the relationship between social networks and abstract notions of territory and tenure among the ancestral Coast Salish. Ethnography provided insight into Coast Salish kinship and marriage practices that both flow from and furnish the means for forming, maintaining, and negotiating social networks. Once formed, these networks functioned as, among other things, social fields for facilitating, asserting, and legitimizing access to nonlocal places and resources. Considering the distribution of obsidian as the physical remains of these networks, the combination of ethnographic and archaeological data gives substance to ancestral Coast Salish social practices that would be otherwise invisible in deep time.

83 Chapter 4. Conflict and Territoriality: An Archaeological Study of Ancestral Northern Coast Salish-Tla’amin Defensiveness in the Salish Sea Region of SW British Columbia

In both state and non-state societies, territorial and tenurial claims are enacted daily and form the basis for socioeconomic interactions at local and extra-local scales (Blackburn 1986; Chou 1997; Ingold 1986; Mearns 1993). Generally, such claims and the socioeconomic interactions they facilitate, are peaceful, mutually beneficial arrangements. However, the potential for conflict is always present, and, when unchecked, can turn violent (Matter 2010; Toft 2003).

Over two decades ago, Keeley (1996) demonstrated the extent to which violent conflict and its threat permeated non-state societies. Despite his extensive review, biases persisted in perceived notions about conflict in non-state societies. Such was the case among the Coast Salish of the Northwest Coast of North America, who, until relatively recently, were considered anomalously peaceful relative to other Northwest Coast groups (Angelbeck 2007). Detailed archaeological and ethnographic reviews reveal that violent conflict was well integrated into Coast Salish social systems, and that not recognizing such conflict denied the Coast Salish an important part of their history (Angelbeck 2009, 2016; Angelbeck and Grier 2012; Angelbeck and McLay 2011; Keddie 1984, 1996; Schaepe 2006).

Coast Salish archaeological and ethnographic records indicate the degree to which all interactions, including conflict, were intertwined with a well-developed social system based on group exogamy and bilateral kinship (Angelbeck 2016; Collins 1979; Elmendorf 1971; Suttles 1963). These foundational affinal and kin relations created and maintained widespread social networks, including those involving tenurial and territorial claims. Consequently, such networks played a significant role in planning and enacting conflict in both offensive and defensive scenarios (Angelbeck and McLay 2011).

Coast Salish ethnohistory describes how various locations associated with settlements were used for defence. Access to, and the connections among, these locations mirror the social networks that define Coast Salish social organization (Angelbeck 2016:12-14). During times of conflict, these linked places formed defensive

84 networks that functioned to maximize defensibility at settlement and allied settlement scales (Schaepe 2006; cf., Supernant 2014)—the intrinsic and extrinsic components of defensibility, respectively (Bocinsky 2014:167). These networks manifested in the Coast Salish natural and built environments as any combination of defensive settlements, lookouts, redoubts, subterranean refuges, and trench embankments (Angelbeck 2009, 2016). Examining the distribution of such defensive networks in time and space provides insights into the role of conflict in broader social contexts including its relationship to territorial and tenurial claims expressed through territoriality.

In this paper, we explore the relationship between territorial and tenurial claims, and conflict among the ancestral Tla’amin-Northern Coast Salish (Figure 21). From a landscape perspective, we examine the temporal and spatial distribution of defensive networks as expressions of territoriality (i.e., defensive territoriality) in geographically distinct sub-areas of Tla’amin territory. Analyzing sites collectively as part of defensive networks rather than as discrete locations, reveals that defensive strategies were in place in Tla’amin territory from at least 900 years BP. This in turn suggests the longevity of territorial and tenurial claims expressed through defensive territoriality, as noted in the ethnohistoric record. Our study is aligned with other research that considers the role of landscape modification and settlement placement in defensibility (e.g., Bocinsky 2014; Cookson 2013; Sakaguchi et al. 2010).

Figure 21: General locations of places, landforms, and bodies of water referred to throughout the text. Outer dotted line represents the approximate extent of Tla’amin traditional territory. Map generated from the ESRI basemap catalogue using ArcMap 10.3.

4.1. Territory, Tenure, and Territoriality Among the Northern Coast Salish

The interrelated concepts of territory, tenure, and territoriality provide a starting point for our investigation into conflict and defensive networks in ancestral Tla’amin territory. These concepts facilitate bridging arguments between ethnographic

86 descriptions of Northern Coast Salish defensive strategies and archaeological settlement patterns. This in turn allows for inferences on the potential role of ancestral network strategies in the defense of territorial and tenurial interests.

We use the term territory to refer to nested but permeable spaces with which an individual or group identifies through some combination of spiritual, familial, affinal, and economic relationships. Territories are both physical (natural and/or culturally modified places) and abstract (e.g., hunting ranges, trade relations, water routes, and sociopolitical connections) entities that shift in extent and significance in concert with changing kinship and other socioeconomic relationships (e.g., Ingold 1986:138). This understanding of territory as a somewhat fluid concept is reflected in Northern Coast Salish network-based social structure. Traditionally, individuals were linked with a particular place, community, and family through kinship and membership in a descent group that was legitimized through a combination of kin connections, the acquisition of an ancestral name, residency, and the investment of labour (Kennedy 2007:23-25, 28- 29). Yet, the practice of group exogamy within a bilateral kinship structure allowed for membership in far-reaching socioeconomic networks (Barnett 1955:182; Kennedy 2007:6), including networks for defense (Angelbeck 2016).

Tenure defines the communication of relationships that form social networks within a continuously evolving social sphere (Ingold 1986: 130, 136-137). Fundamental characteristics of tenure include the public assertion and recognition of access and rights to land and resources at local and extra-local scales, the generational transfer of prerogatives to land and resources, and the knowledge and social protocols that inhere in tenurial relations. During times of conflict, tenurial relations that flowed through social networks within and between Coast Salish communities became a means through which groups could cohere against a common threat (Angelbeck 2016; Angelbeck and McLay 2011).

Territoriality refers to the practices that serve to communicate territorial and tenurial interests (Keeley 1996:55-58; Taçon 2008:224); it is the instrument of tenure (Ingold 1986:141). Thus, whereas tenure refers to the abstract communication of territorial interests, territoriality is the physical manifestation of those interests either through direct action between individuals or groups (e.g. conflict, trade) or through the manipulation of the environment. It is best inferred archaeologically from the act of

87 marking spaces such that they become meaningful places to those who have claimed the space (e.g., Zedeño and Bowser 2009).

4.2. The Tla’amin-Northern Coast Salish and Their Neighbours

The Tla’amin are closely tied through language, kinship, and marriage to their northern neighbours the Northern Coast Salish-Klahoose and Homalco, and, to a lesser degree, their southern neighbours the Northern Coast Salish-Shishálh (Kennedy and Bouchard 1983:52; Washington 2004:587). In the recent past, the close relations between the Tla’amin, Klahoose, and Homalco extended to their winter settlements, which they shared at Grace Harbour in Malaspina Inlet (Figure 21; Barnett 1955:25; Kennedy and Bouchard 1983:89, 1990:447).

Tla’amin traditional territory encompasses seascapes and complex topography with large bodies of water and the Coastal and Insular Mountain ranges circumscribing traditional use areas (Figure 21). Prior to the last century, travel was typically by way of canoe but trails also crosscut the territory joining many settlements by land routes. Such travel facilitated the maintenance of social networks that defined the broader Northern Coast Salish cultural region. Both descendent and ancestral Tla’amin settlement patterns focused on the coastline of the mainland, the heads of bays, coves, and inlets, and along island shorelines.

The Tla’amin and other Northern Coast Salish groups, while distinct linguistically, share cultural traits with all Coast Salish peoples—bilateral kinship, group exogamy, place-based identity, and behavioural norms that emphasize voluntary association, network organization, and mutual aid (sensu Angelbeck and Grier 2012). The intra and intervillage social network organization that joined Northern Coast Salish communities was initiated through marriage alliances and formalized through kinship, which enabled local and extra-local interaction throughout the Coast Salish world and beyond (e.g., Blake 2004; Morin 2012; Springer et al. 2018). They were also important during periods of conflict for bringing together allied groups (Angelbeck and McLay 2011) and for warning others within a common network (Angelbeck 2009:116; Suttles 1951:322; Tollefson 1996:155).

88 Although widely connected, tensions still existed between allied Salishan groups (Kennedy and Bouchard 1983:89) and non-Salishan groups with whom socioeconomic links were more fragile. In recent times, of particular concern to the Tla’amin were the Southern Kwakwaka’wakw-Laich-kwil-tach whose territory abutted the northern extent of the Coast Salish world. From approximately 250 to 100 years BP, the Laich-kwil-tach were in a continuous state of reciprocating violent conflict with Northern and Central Coast Salish groups (Angelbeck 2007, 2009; Angelbeck and McLay 2011; Barnett 1955:267; Codere 1990:359-360; Kennedy and Bouchard 1990:443-444; Taylor and Duff 1956). Earlier periods of violence were part of a cycle of conflict from 1600 to 500 years BP that has been noted in the Salish Sea region (e.g., Angelbeck 2007, 2009, 2016; Angelbeck and Grier 2012; Keddie 1984, 1996; Schaepe 2006) and across the Pacific Northwest Coast (e.g., Cybulski 2014; Maschner 1997; Moss and Erlandson 1992).

Tla’amin oral histories suggest that raiding by the Laich-kwil-tach was a major impetus for defensive networks. However, the broader ethnohistoric record documents internecine conflict and attacks from foreign enemies (Angelbeck 2009:227-241; Barnett 1955:267; Carlson 2001:48-49; Elmendorf 1992:465; Suttles 2000:200-202), as well as allied Northern Coast Salish/Laich-kwil-tach attacks on other Coast Salish groups (Barnett 1955:267; Kennedy and Bouchard 1990:443-444). The focus on the Laich-kwil- tach in the Tla’amin ethnohistoric record stems from the 19th century AD when Laich- kwil-tach raiding turned to territorial expansion southward into Northern Coast Salish territory. This expansionism was bolstered by the Laich-kwil-tach’s relatively greater acquisition of European firearms (Angelbeck 2007:269-270; Duff 1964:59-60), and the relatively greater decline in Coast Salish populations as a result of European introduced diseases (Harris 1994).

A number of different measures were used by the Tla’amin to maximize tactical advantage and increase defense during times of conflict. Engineering efforts included building defensive settlements (Barnett 1935-1936:6:47, 84-85; Kennedy and Bouchard 1983:90; Lamb 1984:604), subterranean refuges (Barnett 1935-1936:6:44, 201, 1944:266-267, 1955:49-52), and trench embankments parallel to the beach at or near settlements to act as a wall or foundation for a palisade (Tla’amin Traditional Use Study 1996). Redoubts and lookouts were also used as part of a broader defensive network strategy. Placed on high points of land, lookouts significantly increased visibility from

89 coastline locations and along narrower channels and inlets (Angelbeck 2009:256). Also situated on high points of land, redoubts functioned as refuges from attack or places from which defenders could safely assail would be invaders. Although similar to lookouts in positioning above sea level, the three known redoubts in Tla’amin territory are integral components of the associated settlements, have restricted viewsheds, and are mostly hidden from view. Lookouts, on the other hand, are located at or within 3 km of a settlement on high points of land with unrestricted visibility (Table 9).

4.3. Methods

4.3.1. Site Selection

To examine the relationship between conflict and ancestral Tla’amin territoriality, we first assembled all available ethnohistoric and archaeological information on shell midden sites (n=98) in Tla’amin territory (Figure 21). Using descriptions from Tla’amin ethnohistory and site morphology, we grouped the sample of 98 sites and features into one of four categories: settlement, lookout, trench embankment, or redoubt (Figure 22; Table 9-11). Archaeological sites and features not defined in Tla’amin ethnohistory as settlements, lookouts, trench embankments, or redoubts, were designated as such by considering various attributes including height above sea level, site extent, associated features, midden composition (where possible), and location (Table 9).

Ethnohistoric data were derived from ethnographic documents (Barnett 1955; Kennedy and Bouchard 1983), a Tla’amin traditional use study (1996), two place names studies (Finnamore et al. 2016; Tla’amin Place Names 2002), and explorer journals (Lamb 1984; Menzies 1923). Archaeological and site setting data were collected as part of the Simon Fraser University (SFU)-Tla’amin First Nation Archaeology and Heritage Stewardship Project, 2008-2013. These data were then augmented with spatial data and site reports housed in the Provincial Archaeology Branch’s Remote Access to Archaeological Data (RAAD). All of the radiocarbon samples were collected during the SFU-Tla’amin initiative (Table12).

Figure 22: General locations of archaeological sites referred to throughout the text. The ID numbers provide additional information in Tables 10 and 11. Map generated from ESRI basemap catalogue using ArcMap 10.3.

91 Table 9: Descriptions of archaeological site types1 Type Location Description

Settlement Typically, 1-5 masl, Stratified middens, house platforms, sited at heads of may have associated intertidal inlets, bays, coves, or components (e.g., canoe runs, clam along open shoreline, gardens, fish traps), may have variable viewshed, associated burials close proximity to potable water source

Trench embankment 1-5 masl, usually Earth berm running parallel to directly associated shoreline, used as foundations for with a settlement, may palisades or dry-stacked stone walls be up to within 1 km (Tla’amin Traditional Use Study 1996) away

Lookout ≥5 masl, sited on Thin midden deposits, possible bluffs with wide evidence for a variety of activities viewshed, directly (e.g., tool production, resource associated with or processing) within 3 km of settlements

Redoubt >5 masl, directly Flat-topped, bedrock outcrop with associated with a steeply sloped sides, narrow access settlement, variable point, may have house platforms, may viewshed have thin midden deposits, typically limited surface area Note: 1See Tables 10-15, and Figures 22, 24-29 for information on specific sites.

92 Table 10: Tla’amin habitation sites in ethnohistoric and archaeological records, arranged from largest to smallest by area. ID1 Code2 Location (see Tla’amin Area3 Comments Appendix A) place name (m2)/masl (see Appendix A) 1 DlSd-16 Powell River tiskʷat -/5 Historic village, community relocated to current location (ID 16) in 1910 2 EaSf-11 Cortes Island p̓ aq̓ iʔaǰɩm -/2 Historic village abandoned ca. 1920 3 EaSf-7 Cortes Island saʔyiɬit -/4 Historic village abandoned (Gorge Harbour) before 1900 4 - Blubber Bay t̓atlaχʷnač -/2 Historic village (Texada Island, sayayɩn) 5 - Mary Point gi: t̓aχʷ -/5 Historic village abandoned before 1900 6 - Pocahontas Bay šɛtɛqʷən -/2 Historic village (Texada Island, sayayɩn) 7 - Vananda Bay lɛχʷamɛn -/2 Historic village, place name (Texada Island, translates as ‘being sayayɩn) infected’ suggesting abandonment following early epidemic, ca. late 1700s/early 1800s 8 DkSd-1 Grief Point χakʷum 296070/8 Historic village, possible subterranean refuges used at this location 9 DkSc-3 Myrtle Point kʷʊθaysqɛn 52800/8 Historic village 10 EaSf-2 Smelt Bay (Cortes kʊmaχən 47600/2 Historic village, Island) subterranean bunkers

93 ID1 Code2 Location (see Tla’amin Area3 Comments Appendix A) place name (m2)/masl (see Appendix A) beneath plankhouses may have been used here 11 EaSg-3 Marina Island - 46152/2 Historic village abandoned (šɛtqaǰɛ) in late 1800s, site of historic battle with Southern Kwakwaka’wakw-Lekwiltok First Nation 12 DlSe-1 Keefer Bay - 31680/2 Ancestral village (Savary Island, qɛyɛ qʷən) 13 EaSe-6/8 Penrose/Trevenen ƛaqəmayɩn 30600/4 Ancestral village Bays (Isthmus connecting Coode Island to Malaspina Peninsula) 14 EaSg-6 Gorge Harbour yipikʷʊ 28800/30 Ancestral village, high on (Cortes Island) bluff east side of entrance to Gorge Harbour 15 DkSc-4 Brew Bay - 24245/2 Historic village 16 DlSd-11 Current Tla’amin t̓išosəm 20800/4 Main Tla’amin community community since 1910 17 EaSg-17 Marina Island - 17200/2 Ancestral village (šɛtqaǰɛ) 18 DkSb-16 Stillwater Bay qʷoqʷnəs 17063/10 Historic village, transferred to Coast Salish-Shíshálh First Nation through marriage in 1800s 19 EaSe-23 Portage qɛgiyɩn 16500/4 Ancestral village/historic Cove/Wooton Bay campsite, used as a

94 ID1 Code2 Location (see Tla’amin Area3 Comments Appendix A) place name (m2)/masl (see Appendix A) portage between Lancelot Inlet and Desolation Sound/Homfray Channel (θiyčɛmayič) 20 DlSe-13 Emmonds Beach šɛʔaystən 15432/4 Historic village abandoned in late 1880s, later used as a campsite, trail connecting with toχʷnač 21 EaSg-16 Marina Island - 15280/6 Ancestral village (šɛtqaǰɛ) 22 EaSd-7 Theodosia Inlet toqʷanan 13523/2 Historic village abandoned (head) in the 1920s 23 DlSd-36 Harwood Island maloʔhom 13289/20 Ancestral village/historic (ʔaʔgayqsən) campsite 24 DlSf-2 Stag Bay ɬay ta ƛač 12456/2 Ancestral village (Hernando Island, kʷʊp ƛač) 25 EaSe-76 Cochrane Bay qɛqɛgɩš 12265/5 Ancestral village, possibly used as a refuge site, trail connecting with ǰɛǰɩšʔsiʔəm (ID 37) 26 EaSe- Grace Harbour q̓ a qɛy q̓ ay 12200/5 Historic village abandoned 11/65 ca. 1960 27 DlSe-48 Duck Bay (Savary - 11900/10 Ancestral village Island, qɛyɛ qʷən) 28 DlSf-12 Ashworth Point - 11883/2 Ancestral village (Hernando Island, kʷʊp ƛač) 29 DlSd-6 Scuttle Bay ƛɛkʷanəm 10985/5 Ancestral village/historic campsite, possible

95 ID1 Code2 Location (see Tla’amin Area3 Comments Appendix A) place name (m2)/masl (see Appendix A) subterranean refuges used at this location, now part of main Tla’amin community 30 EaSe-53 Stopford Point - 9954/4 Ancestral village 31 DkSb-8 McRae Cove - 9000/3 Ancestral village 32 DkSb-7 Frolander Bay - 8262/2 Ancestral village 33 EaSf-36 Blind Creek ǰɛmoθən 7000/4 Historic village abandoned (Cortes Island) before 1900 34 DlSf-18 Indian Point θatɛq 5166/2 Ancestral village (Savary Island, qɛyɛ qʷən) 35 DkSd-10 Raven Bay - 5134/20 Ancestral village (Texada Island, sayayɩn) 36 EaSe-18 Head of unnamed - 4500/4 Ancestral village bay on east side of Lancelot Inlet 37 EaSe-2 Bliss ǰɛǰɩšʔsiʔəm 4000/5 Ancestral village/historic Landing/Turner campsite, trail connecting to Bay qɛqɛgɩš (ID 25) 38 EaSe-4 Parker čɛn 3960/2 Historic village, possible Harbour/Hinder subterranean refuges used Creek at this location 39 DlSe-14 Okeover Inlet toχʷnač 3874/4 Historic village abandoned (head) ca. 1920, trail to šɛʔaystən (ID 20) 40 EaSe-25 Thor’s Cove - 3766/2 Ancestral village/historic campsite likely associated with EaSe-24 (ID 45)

96 ID1 Code2 Location (see Tla’amin Area3 Comments Appendix A) place name (m2)/masl (see Appendix A) 41 DlSd-8/9 Klahanie čuχʷo θɛn 3640/5 Historic village located between t̓išosəm (ID 51) and ƛɛkʷanəm, now part of main Tla’amin community 42 EaSe-100 Tenedos Bay kʷʊmqɛn 3150/2 Ancestral village, historic campsite 43 DkSc-15 Lang Bay - 2484/5 Ancestral village 44 EaSe-34 Isabel Bay - 2386/5 Ancestral village (Lancelot Inlet) 45 EaSe-24 Entrance to - 2165/2 Ancestral village Theodosia Inlet 46 DlSe-26 Rasmussen Bay - 2100/5 Historic village, may be palisaded village referred to by Barnett (1955:50) 47 EaSe-21 Galley Bay kʷišitəm 2040/2 Ancestral village 48 EaSd-3 Roffey Bay - 2022/3 Historic village abandoned (Prideaux Haven, ca. mid-1700s, described mačɛnay) by Vancouver (Lamb 1984:604) 49 DlSe-7 Lund ƛaʔamɛn 1942/2 Historic village, now part of modern Lund village/harbour 50 EaSf-3 Manson Bay ɬaytoθɛn 1647/2 Historic village abandoned ca. late 1700s 51 DlSd-10 Current Tla’amin t̓išosəm 1600/5 Historic village, now part of community main Tla’amin community 52 DlSe-36 Keays Bay χʷoχʷ ǰusɛm 859/4 Ancestral village/historic campsite likely associated with šɛʔaystən (ID 20)

97 ID1 Code2 Location (see Tla’amin Area3 Comments Appendix A) place name (m2)/masl (see Appendix A) 53 EaSd-1 Laura Cove χɛpǰɛʔqɛn 660/3 Ancestral village/historic (Prideaux Haven, campsite mačɛnay) 54 EaSd-2 Melanie Cove muθkʷamɛn 600/3 Ancestral village/historic (Prideaux Haven, campsite mačɛnay) Note: 1Use IDs to find additional information in Tables 12-15, and Figures 22-29; 2Borden code assigned to sites registered with the Archaeology Branch of British Columbia; 3Area determined from polygons on file with the Archaeology Branch/meters above sea level.

98 Table 11: Tla’amin lookouts, trench embankments, and redoubts in ethnohistoric and archaeological records, arranged by type from largest to smallest by area. ID1 Code2 Location Tla’amin Area3 Comments (Appendix A) place name (m2)/masl (Appendix A) Lookouts 55 DkSc- Myrtle Rocks kʷʊθaysqɛn 9529/10 Historic lookout 13 associated with kʷʊθaysqɛn village (ID 9) 56 EaSe- Scobell Island - 5109/20 Ancestral lookout, likely 116 associated with one or all three Prideaux Haven (mačɛnay) village sites (IDs 48, 53, 54) 57 DlSd- Willingdon ʔahʔǰumɩχʷ 2032/20 Historic lookout 17 Beach 58 EaSe- Malaspina - 1672/6 Ancestral lookout on islet 61 Inlet at entrance to Trevenen Bay, likely associated with ƛaqəmayɩn village (ID 13) 59 DlSe- Okeover Inlet - 1445/8 Ancestral lookout on top 53 of marine terrace abutting toχʷnač village (ID 39) 60 EaSe- Scobell Island - 844/18 Ancestral lookout, likely 114 associated with one or all three Prideaux Haven (mačɛnay) village sites (IDs 48, 53, 54) 61 DlSe- Savary Island - 809/15 Ancestral lookout likely 12 (qɛyɛ qʷən ) associated with either or both Keefer (ID 12) and Duck (ID 27) Bay villages 62 EaSe- Malaspina - 792/30 Ancestral lookout on 67 Inlet bluff at entrance to Grace Harbour, likely associated with q̓ a qɛy q̓ ay village (ID 26) 63 DlSe- Okeover Inlet - 692/8 Ancestral lookout on top 15 of marine terrace just NE of toχʷnač village (ID 39) 64 DlSe- Rasmussen - 600/15 Historic lookout 26 Bay associated with historic village at Rasmussen Bay (ID 46) 65 EaSe- Tenedos Bay - 583/10 Ancestral lookout likely 102 associated with kʷʊmqɛn village (ID 42)

99 ID1 Code2 Location Tla’amin Area3 Comments (Appendix A) place name (m2)/masl (Appendix A) 66 DlSe- Keays Bay - 561/15 Ancestral lookout on 36 bluff abutting χʷoχʷ ǰusɛm village (ID 52), likely associated with χʷoχʷ ǰusɛm and šɛʔaystən (ID 20) villages 67 DlSe- Emmonds - 531/20 Ancestral lookout on 13 Beach bluff abutting šɛʔaystən village (ID 20), likely associated with šɛʔaystən and χʷoχʷ ǰusɛm (ID 52) villages 68 EaSd-8 Laura Cove - 522/20 Ancestral lookout likely (Prideaux associated with Haven, χɛpǰɛʔqɛn ancestral mačɛnay) village (ID 53) 69 DlSf-6 Savary Island t̓i: t̓i: may 453/20 Historic lookout, also (qɛyɛ qʷən) used as a resource gathering place, expansive southerly/southwesterly view of the Salish Sea 70 DlSe- - - 297/25 Ancestral lookout on a 44 bluff SE of šɛʔaystən village (ID 20) likely associated with šɛʔaystən and χʷoχʷ ǰusɛm (ID 52) villages 71 EaSg-6 Gorge - 288/30 Ancestral lookout on Harbour bluff on east side of (Cortes entrance to Gorge Island) Harbour, likely associated with yipikʷʊ village (ID 14) 72 EaSe- Malaspina - 280/8 Ancestral lookout on islet 72 Inlet at entrance to Malaspina Inlet, likely associated with čɛn (ID 38) village site 73 EaSg- Marina Island - 279/30 Ancestral lookout likely 30 (šɛtqaǰɛ) associated with EaSg-16 village (ID 21) 74 DlSe- Savary Island - 240/10 Ancestral lookout likely 46 (qɛyɛ qʷən) associated with either or both Keefer (ID 12) and Duck (ID 27) Bays villages

100 ID1 Code2 Location Tla’amin Area3 Comments (Appendix A) place name (m2)/masl (Appendix A) 75 DlSe- Malaspina - 237/5 Ancestral lookout NW of 34 Peninsula χʷoχʷ ǰusɛm (ID 52) village, likely associated with χʷoχʷ ǰusɛm and šɛʔaystən (ID 20) villages 76 EaSe- - - 185/8 Ancestral lookout NE of 98 qɛgiyɩn village (ID 19), likely associated with qɛgiyɩn 77 DlSd- Gibson’s qʷɛqʷiqʷɛy 178/10 Historic lookout likely 12 Beach associated with t̓išosəm village (ID 16) 78 DlSe-7 Lund - 178/10 Historic lookout likely associated with ƛaʔamɛn village (ID 49) 79 DlSe- Okeover Inlet - 175/20 Ancestral lookout likely 21 associated with ƛaqəmayɩn village (ID 13) 80 DlSf-16 Savary Island - 172/20 Ancestral lookout (qɛyɛ qʷən) 81 DlSd-2 Gibson’s qʷɛqʷiqʷɛy 171/10 Historic lookout likely Beach associated with either or both tiskʷat (ID 1) and t̓išosəm (ID 16) villages 82 DkSc-9 Albion Point, - 168/10 Historic lookout likely Brew Bay associated with DkSc-4 village in Brew Bay (ID 15) 83 EaSe- Theodosia - 158/6 Ancestral lookout on 17 Inlet islet, likely associated with toqʷanan village (ID 22) 84 DlSd- Harwood - 145/40 Ancestral lookout likely 36 Island associated with (ʔaʔgayqsən) maloʔhom village (ID 23) 85 DlSd-1 Gibson’s qʷɛqʷiqʷɛy 127/10 Historic lookout likely Beach associated with either or both tiskʷat (ID 1) and t̓išosəm (ID 16) villages 86 DlSe- Dinner Rock qʷaqʷtɛm 107/10 Historic lookout NW of 37 χʷoχʷ ǰusɛm (ID 52) village, likely associated with χʷoχʷ ǰusɛm and šɛʔaystən (ID 20) villages

101 ID1 Code2 Location Tla’amin Area3 Comments (Appendix A) place name (m2)/masl (Appendix A) 87 DlSd- - - 89/28 Ancestral lookout 25 between χakʷum (ID 8) and tiskʷat (ID 1) villages, likely associated with tiskʷat 88 DkSb- McRae Cove qʷoqʷnəsəm Ancestral lookout on islet 107 islets at entrance to McRae Cove, likely associated with DkSb-8 village (ID 31) at head of cove Trench Embankments 89 DlSf-3 Hernando - 15095/5 Ancestral earthwork Island likely associated with (kʷʊp ƛač) either or both ɬay ta ƛač (ID 24) and Ashworth Point (ID 28) villages 90 DlSd-7 Klahanie - 4847/2 Historic earthwork likely (čuχʷo θɛn) associated with ƛɛkʷanəm (ID 29), čuχʷo θɛn (ID 41), and t̓išosəm (ID 16) villages 91 DlSe- Emmonds (šɛʔaystən) 4032/2 Historic earthwork 13 Beach associated with šɛʔaystən village (ID 20) 92 EaSg-1 Marina Island - 1941/15 Ancestral earthwork (šɛtqaǰɛ) likely associated with EaSg-16 village (ID 21) 93 EaSf-1 Manson’s - 1121/1 Historic earthwork likely Landing associated with ɬaytoθɛn (Cortes village (ID 50) Island) 94 DlSf-5 Hernando - 1015/5 Ancestral earthwork Island likely associated with (kʷʊp ƛa) either or both ɬay ta ƛač (ID 24) and Ashworth Point (ID 28) villages 95 DlSf-4 Hernando - 665/5 Ancestral earthwork Island likely associated with (kʷʊp ƛa) Ashworth Point (ID 28) village Redoubts 96 EaSd-3 Roffey Cove - 860/10 Granite outcrop (Prideaux associated with historic Haven, EaSd-3 village (ID 48) mačɛnay) 97 EaSd-1 Laura Cove χɛpǰɛʔqɛn 370/8 Granite outcrop associated with

102 ID1 Code2 Location Tla’amin Area3 Comments (Appendix A) place name (m2)/masl (Appendix A) (Prideaux ancestral χɛpǰɛʔqɛn Haven, village (ID 53) mačɛnay) 98 EaSd-2 Melanie Cove muθkʷamɛn 276/8 Granite outcrop (Prideaux associated with Haven, ancestral muθkʷamɛn mačɛnay) village (ID 54) Note: 1Use IDs to find additional information in Tables 12-15, and Figures 22-29; 2Borden code assigned to sites registered with the Archaeology Branch of British Columbia; 3Area determined from polygons on file with the Archaeology Branch/meters above sea level.

We focused our analyses on three sub-areas—open shoreline, inlets, and coves. The three sub-areas collectively represent the topographic variation within Tla’amin territory, and the distinctive geography of each influences access, exposure, and viewshed. The open shoreline subarea is characterized by large, ocean-facing settlements with relatively unrestricted access. By contrast, access to inlets is highly canalized, restricting travel to a head-on approach giving strategic advantage to settlements sited at inlet heads. This strategic advantage is further amplified in coves since many have narrow openings fronted by islands and islets that limit choices of approach and effectively hide cove sites from external observers.

Within the three sub-areas, we identified a sample of the known sites from each of the four categories of settlements, lookouts, trench embankments, and redoubts. This sample was used to assess the defensibility of sites within the three sub-areas. Based on proximity (see Table 9), ethnohistoric information on conflict (Kennedy and Bouchard 1983; Lamb 1984; Menzies 1923), and regional culture history (Barnett 1955), we grouped the sites into site complexes. These groups of sites are considered associated based on their spatial relationships and, in some cases, temporal relationships. Trench embankments and redoubts are components of settlements, in close proximity to house platforms. In some cases, lookouts are also directly associated with settlements, in other cases they are distant. For the latter, we argue that having outlying lookouts increased the combined viewsheds of the complexes, and allowed for early warning either by signal fire (Angelbeck 2009:116), runners (Suttles 1951:322; Tollefson 1996:155), or canoe. We also recognize that an outlying lookout could have been used by allied settlements. For each site and feature that comprise the site complexes, we calculated viewsheds as well as defensibility indices. We assess a site complex’s overall

103 defensibility by considering a combination of these measures as well as potential alliances with neighbouring complexes.

4.3.2. Open Shoreline Site Complexes

Emmonds Beach (Figures 21-22; Table 10, ID 20) and Keays Bay (Figures 21-22; Table 10, ID 52)

We consider Emmonds Beach and Keays Bay to be the same settlement given their close proximity and refer to them collectively as Emmonds. There are five lookouts (Figure 22; Tables 11 and 13, IDs 66, 67, 70, 75, 86) and a single trench embankment (Figure 22; Tables 11 and 13, ID 91) associated with the Emmonds settlement.

The setting of Emmonds affords it some degree of protection. It backs a south- facing, exposed beach; however, the surrounding land to the north and west rises steeply. A high bedrock outcrop stands ~20 masl between Emmonds Beach proper and the adjoining, Keays Bay. In the recent past, a trail ran northwest from Emmonds to connect with the settlement at the head of Okeover Inlet.

The midden burials and aboveground box burials found at Emmonds indicate mid-Holocene and historic occupations (Table 13). Based on local culture history, the associated trench embankment could date to at least 1600 years BP, but it is described in Tla’amin oral history as having been used to barricade the settlement against northern raiders in the mid to late 18th century AD (Tla’amin Traditional Use Study 1996). The place name for Emmonds (šɛʔaystən, ‘steep’) may refer to this trench embankment (Finnamore et al. 2016; Tla’amin Place Names 2002).

104 Table 12: Dated sites from sample used in defense index and viewshed analyses and associated ancillary sites. ID1 Lab Code Site type Material Conventional 1-σ cal. BP (probability, 2-σ cal. BP Median number radiocarbon calendar date) (probability, probability age (BP) calendar date) cal. BP (calendar date) Bliss Landing (ǰɛǰɩšʔsiʔəm) 37 SFU- EaSe-2 settlement bone 4000±60 3647-3452 (100%, 1698- 3767-3372 (100%, 3562 (1613 1108 collagen2 1503 BC) 1818-1423 BC) BC) (human) 37 UCI- EaSe-2 settlement bone 3595±20 3149-2989 (100%, 1200- 3224-2911 (100%, 3068 (1119 132982 collagen2 1040 BC) 1275-962 BC) BC) (dog) Okeover Inlet (toχʷnač) 59 UCI- DlSe-53 lookout shell3 1025±15 167-145 (12%, AD 1652- 367-76 (100%, AD 240 (AD 1710) 163679 1777), 298-173 (88%, AD 1553-1874) 1783-1805) 63 Beta- DlSe-15 lookout shell3 1700±30 895-761 (100%, AD 1055- 937-688 (100%, AD 823 (AD 1127) 342735 1189) 1013-1262) - Beta- DlSe-54 fish trap wood4 90±30 260-220 (30%, AD 1700- 270-210 (27%, AD 108 (AD 1842) 305568 1720), 140-30 (70%, AD 1680-1740), 140-20 1820-1920) (73%, AD 1810- 1930) Portage Cove (qɛgiyɩn) - Beta- EaSe- fish trap wood4 360±30 490-430 (57%, AD 1460- 500-310 (100%, AD 414 (AD 1536) 263329 137 1520), 370-320 (43%, AD 1450-1640) 1580-1630) Prideaux Haven (mačɛnay) 53 Beta- EaSd-1 settlement charred 650±40 660-630 (45%, AD 1290- 670-550 (100%, AD 605 (AD 1345) 263674 wood 1320), 600-560 (55%, AD 1280-1400) 1350-1390) 53 Beta- EaSd-1 fish trap wood4 380±40 500-430 (74%, AD1450- 510-310 (100%, AD 438 (AD 1512) 263326 1520), 360-330 (26%, AD 1440-1640) 1590-1620 54 Beta- EaSd-2 fish trap wood4 430±50 520-480 (100%, AD 1430- 540-430 (79%, AD 481 (AD 1469) 263327 1470) 1420-1520), 370-

105 ID1 Lab Code Site type Material Conventional 1-σ cal. BP (probability, 2-σ cal. BP Median number radiocarbon calendar date) (probability, probability age (BP) calendar date) cal. BP (calendar date) 320 (21%, AD 1580- 1630) 48 Beta- EaSd-3 fish trap wood4 310±40 440-350 (75%, AD 1500- 490-290 (100%, AD 387 (AD 1563) 263325 1600), 340-300 (25%, AD 1460-1660) 1610-1650) - Beta- EaSe- fish trap wood4 540±50 620-610 (35%, AD 1330- 650-580 (44%, AD 558 (AD 1392) 263328 118 1340), 560-520 (65%, AD 1300-1370), 570- 1400-1430) 510 (56%, AD 1380- 1440) Note: 1Use IDs to find additional information in Tables 10-11 and 13-15, and Figures 22-29; 2a marine correction of 421+/-54 with an assumed 90% marine-based protein diet (Chisholm 1986) was applied to the bone date using Calib 7.1 (Stuiver et al. 2017); 3a marine correction of 421+/- 54 was applied to the shell dates using Calib 7.1 (Stuiver et al. 2017); 4waterlogged. The marine correction used is an average derived from local Delta-R dates from five locations in the Salish Sea: Savary Island (510+/-50); Departure Bay (440+/-50); Burrard Inlet (410+/-40); Comox (380+/- 50); and Nanoose Bay (370+/-50).

106 Table 13: Summary of defensiveness of site complexes in three sub-areas. Site Ethnohistory lookout Redoubt Trench Trail Min/max Min/max Min/max Chronometric complexes of conflict embankment viewing viewshe defence and inferred (Y/N) distance1 d area2 index3 dates4 (km) (km2) Open shore Emmonds, Y 5 - 1 To 6.9/18.4 48.6/370 0.55/2.39 5400-1500 (MB), (šɛʔaystən) Okeover, 500-200 (BB), toχʷnač 1600-100 (DF) Bliss N - - - To 6.9 24.8 0.51 3767-2911 (BP) Landing, Cochrane 5420-3500 (DA), (ǰɛǰɩšʔsiʔəm) Bay, 1200-250 (DA) qɛqɛgɩš Inlets Okeover, N 2 - - To 8.5/11.1 13.1/16.2 0.91/2.56 937-20 (BP) (toχʷnač) Emmonds , šɛʔaystən Portage N 1 - - - 8.5/11.1 24/24 1.05/1.39 500-310 (BP), (qɛgiyɩn) 2000-250 (DA) Coves Prideaux Y (Roffey 3 3 - - 7.7/16.6 0.5/93.2 0.66/2.47 670-290 (BP) Haven Cove) (mačɛnay) Note: 1See Table 14 viewing radius calculations (for Bliss Landing there is only one viewpoint so only one viewing distance); 2See Table 14 viewshed area calculations; 3See Table 15 for defense index calculations; 4See Table 12 for additional information on chronometric (BP) and inferred dates for midden burials (MB), box burials (BB), defensive features (DF), and diagnostic artifacts (DA); 5Approximate time frames for midden and aboveground box burials in the Salish Sea region, respectively (Mathews 2014:92-94); 6Approximate time frame for defensive features in the Salish Sea region (Angelbeck and Grier 2012:564); 7Approximate time frames for diagnostic artifact assemblages recovered from Bliss Landing (Beattie 1972:28-32); 8Approximate time frame for diagnostic artifact recovered from Portage Cove midden (Clark 2010:70-74).

107 Bliss Landing (Figures 21-22; Table 10, ID 37)

There is no ethnohistoric record of conflict at the Bliss Landing settlement, and it has no obvious associated archaeological defensive features. The settlement is located ~15 km north of Emmonds and sits above the southwest facing beach of Turner Bay with a year-round creek running along the southeast edge. Although open to the Salish Sea, it is partially protected on the southwest from the northern group of the Copeland Archipelago and to the northwest by a group of islets that front the bay. The land to the north and south rises gently to low bluffs while the back of the settlement grades into swampy ground and forest. The house platforms, midden burials, and site extent indicate that this was a major settlement in the past but Tla’amin oral history describes Bliss Landing as a seasonal campsite for procuring intertidal resources, fish, roots, and marine and terrestrial mammals.

Radiocarbon and inferred dates for Bliss Landing range from mid-late Holocene and later pre-contact (Tables 12 and 14), which overlaps with Emmonds. An historic trail between Bliss Landing and the comparably aged Cochrane Bay settlement, indicate the possibility of a connection between the two settlements not only in the recent past but also several millennia ago.

4.3.3. Inlet Site Complexes

Okeover (Figures 21-22; Table 10, ID 39)

The Okeover settlement is located at the head of Okeover Inlet between two ancient marine terraces with associated lookouts (Tables 10 and 11, IDs 59 and 63). A year-round creek flows into the inlet east of the settlement and the remains of an early 19th century AD fish trap, skirt the creek mouth (DlSe-54). Although not remembered as a place of conflict, dates from the two lookouts (Tables 12 and 13) place this site complex in the last half of the earlier cycle of conflict (1600-500 years BP) and at the beginning of the more recent surge between 250- and 100-years BP.

Portage Cove (Figures 21-22; Table 10, ID 19)

The Portage Cove settlement is located on an isthmus joining Gifford Peninsula and the mainland. The land to the northeast and southwest of the settlement rises fairly steeply making water access the best option from either Lancelot Inlet to the southeast

108 or via Portage Cove to the northwest. The place name for the settlement (qɛgiyɩn, ‘walk across’) refers to its use as a portage between Wooton Bay (at the head of Lancelot Inlet) and Portage Cove (Finnamore et al. 2016; Kennedy and Bouchard 1983:162-163; Tla’amin Place Names 2002), reducing travel between the two areas by 10-15 km.

Historically, Portage Cove was a temporary campsite but, in the past, it was likely

2 a more permanent settlement given the extent of the midden (16,500 m , over 2 m deep). Temporal data suggest a mid-late Holocene and late pre-contact occupation of the settlement that coincides with occupations at Emmonds, Bliss Landing, and Okeover (Tables 12 and 13). There is no oral history associating Portage Cove with periods of conflict but its position as a short-cut between Lancelot Inlet and Desolation Sound/Homfray Channel, and its associated lookout (Figure 22; Tables 11 and 13, ID 76) make it a strategic location.

4.3.4. Cove Site Complexes

Laura Cove (Figures 21-22; Table 10, ID 53), Melanie Cove (Figures 21-22; Table 10, ID 54), and Roffey Cove (Figures 21-22; Table 10, ID 48)

The three settlements in our Cove subsample comprise what is now known locally as Prideaux Haven. The settlements are located at the cove heads and have house platforms situated on and around bedrock outcrops. There are wooden fish traps at stream mouths in each of the coves. The settlements at Laura and Melanie Coves would have been all but hidden from canoes passing by along Homfray Channel. The hiddenness of Laura Cove is highlighted in its place name χɛpǰɛʔqɛn, which translates as ‘to go back’ (Finnamore et al. 2016; Tla’amin Place Names 2002). The name refers to the fact that Laura Cove cannot be seen when travelling down Homfray Channel, only when going up (north by northeast). It would appear that Bocinsky’s (2014) suggestion in his study of defensiveness in the Central Coast Salish territory of the Salish Sea that, “[i]nvisibility may be as important as visibility” (2014:167), may well be exemplified in the Northern Coast Salish case of Melanie and Laura Coves. Roffey Cove is more visible to passersby but only at close range. This general morphology suggests the coves functioned as defensive settlements using the outcrops as redoubts during increased periods of conflict. This assumption is given additional support by the three associated lookouts (Figure 22; Tables 11 and 13, IDs 56, 60, 68). Radiocarbon dates place the

109 three settlements at the end of the early cycle of conflict and near the beginning of the more recent increase in violence (Tables 12 and 13).

The historic record for Prideaux also indicates a defensive character. The defensive features of the settlement at the head of Roffey Cove were described by Captain Vancouver and his botanist Archibald Menzies in AD 1792 (Lamb 1984:604; Menzies 1923:66-68). Both descriptions were subsequently used as source material for a water colour by the artist William Alexander, who painted the settlement sometime after Vancouver’s return to England (Figure 23).

4.3.5. Viewshed Analyses

We conducted viewshed analyses both on the complete data set from Tla’amin territory and on our sub-set of five site complexes to determine visibility of sites from approaching canoes and visibility of approaching canoes from the sites and features within each of the five site complexes. These individual analyses gave us a minimum and a maximum viewing distance and area for each site complex.

We used ArcMap on the ArcGIS 10.3 platform to calculate viewsheds from the point-of-view of observers standing at each of the four site and feature types, as well as from points on the water. The former represents the position of defense and the latter represents the perspective of the attacking force from a canoe(s). Our analysis assumes a daytime attack, and that sea and weather conditions are optimal for visibility. However, attacks did sometimes occur at night or in the early hours of the morning (Jenness 1934, cited from Angelbeck 2009:86, 171-172). For land-based observers, viewsheds were calculated using an eye height of 1.7 m (Supernant 2014). For canoe-based observers this was reduced to 0.5 m, assuming a sitting position in a canoe (Supernant 2014).

110

Figure 23: “Deserted Indian Village” by William Alexander (1798). Thought to be of the Roffey Cove settlement in Prideaux Haven. (Used with permssion from the copyright holder: The Newberry Library, Chicago. Call # VAULT oversize Ayer Art Alexander, Drawing no. 17).

111 We constrained our viewshed analyses with a maximum viewing distance or radius. This maximum was derived from the maximum distance to the horizon from a given location, which was determined using an equation that accounted for observer eye-height, height above sea level, and the radius of the Earth (Dalager 2005) (Table 14). These constraints allowed for the determination of what was potentially visible from a given point and precluded issues with edge effects on the calculated viewsheds. In some cases, the calculated viewing distances resulted in viewsheds with significant coverage. Even in optimal conditions it would be difficult for an unaided eye to distinguish much detail beyond 2.5 to 3 km (Krisciunas and Carona 2015), even with added height above sea level. However, since context and repetitive motion play a significant role in object identification (Bar 2004; Oliva and Toralba 2007; Palmer 1975; Polana and Nelson 1994) even distant approaching objects on the water would likely have been interpreted as canoes with paddlers.

The digital elevation models (DEM) for the viewshed analyses were downloaded as 20 m x 20 m (i.e., 19.06737202 m x 19.06737202 m) paired tiles (east-west) from the Natural Resources of Canada GeoGratis website (GeoGratis 2015). The tiles were mosaicked into a single DEM and projected to a NAD 1983 UTM Zone 10 N Transverse Mercator projection. The resulting DEM is a rectilinear polygon measuring 5,455,705 and 5,597,223 m north, and 318,061 and 429,090 m east in UTM Zone 10. To align properly with the DEM, site shapefiles downloaded from RAAD were re-projected from Albers to a NAD 1983 UTM Zone 10 N Transverse Mercator projection.

112 Table 14 Summary of viewing radius (VR) and area calculations for viewshed analyses. ID1 Code/location Site type Place name Site Observer VR5 or Viewshed (see elevation3 height4 (m) distance area (km2) Appendix A) (masl) (km) 20 DlSe-13/Emmonds settlement šɛʔaystən 4 1.7 8.5 106.3 52 DlSe-36/Emmonds settlement χʷoχʷ ǰusɛm 4 1.7 8.5 48.6 86 DlSe-37/Emmonds lookout qʷaqʷtɛm 10 1.7 12.2 192.5 75 DlSe-34/Emmonds lookout - 5 1.7 9.2 92.8 70 DlSe-44/Emmonds lookout - 25 1.7 18.4 370 67 DlSe-13/Emmonds lookout šɛʔaystən 20 1.7 16.6 307.8 66 DlSe-36/Emmonds lookout χʷoχʷ ǰusɛm 15 1.7 14.6 275.9 91 DlSe-13/Emmonds earthwork2 šɛʔaystən 2 1.7 6.9 56.5 37 EaSe-2/Bliss settlement ǰɛǰɩšʔsiʔəm 2 1.7 6.9 24.8 59 DlSe-53/Okeover lookout - 8 1.7 11.1 13.1 63 DlSe-15/Okeover lookout - 8 1.7 11.1 16.2 39 DlSe-14/Okeover settlement toχʷnač 4 1.7 8.5 15.2 19 EaSe-23/Portage settlement qɛgiyɩn 4 1.7 8.5 24 76 EaSe-98/Portage lookout - 8 1.7 11.1 24 53 EaSd-1/Prideaux settlement χɛpǰɛʔqɛn 3 1.7 7.7 2.7 54 EaSd-2/Prideaux settlement muθkʷamɛn 3 1.7 7.7 1.7 48 EaSd-3/Prideaux settlement - 3 1.7 7.7 23.4 60 EaSe-114/Prideaux lookout - 18 1.7 15.8 93.2 56 EaSe-116/Prideaux lookout - 20 1.7 16.6 90.4 68 EaSd-8/Prideaux lookout - 20 1.7 16.6 31.6 96 EaSd-3/Prideaux redoubt mačɛnay 10 1.7 12.2 14.5 97 EaSd-1/Prideaux redoubt χɛpǰɛʔqɛn 8 1.7 11.1 0.5 98 EaSd-2/Prideaux redoubt muθkʷamɛn 8 1.7 11.1 1.4 - - canoe - 0 0.5 2.5 15.2 Note: 1Use IDs to find additional information in Tables 9-14 and Figures 22-29; 2earthwork refers to trench embankment at Emmonds; 3meters above sea level; 4average height of a standing observer and of an observer sitting in a canoe (Supernant 2014:502-503); 5viewing radius calculated using the formula d=√(h[2r+h]), where d = distance to horizon, h = height above sea level (observer height + elevation above sea level), and r = the radius of Earth (Dalager 2005); for r, the volumetric mean radius of the Earth (6371.0 km) was used (NASA 2013).

113 The shapefiles were then converted from polygon features to closed polyline features using the drawing tool in ArcMap. We chose polylines on the assumption that observers would typically be standing on the boundaries of settlements and lookouts to view something out on the water. However, it should be noted that lookouts were sometimes positioned on rooftops during night watches (Jenness 1934, cited from Angelbeck 2009:86, 171-172). The canoe viewsheds were run using point features, representing a person sitting in a canoe. Viewshed areas for both land and water-based observers were calculated using the Zonal Geometry as Table tool in ArcMap 10.3. Area was calculated to allow for comparison of the horizontal space potentially visible from a particular site type (e.g., lookout vs. settlement).

Although we calculate the viewshed of the individual features and sites within each site complex, we evaluate the defensibility of a site complex based on the whole rather than its component parts. The value of considering site complexes, rather than individual locations is that it recognizes that occupants of most settlements could take advantage of a number of different locations (lookouts, hideouts, escape routes) in times of conflict. Thus, for each settlement complex, we consider its most expansive viewshed to be the most meaningful when discussing overall settlement complex defensibility. Given the potential viewshed distance (2.5 km) and area (15.2 km2) from a canoe, we consider those site complexes with a viewshed distance ≥3 km and an area ≥16 km2 to be relatively more defensible from marine-based assault.

Defense Index

We calculated a defense index (DI) for component parts of site complexes; providing us with an overall assessment of defensibility. The DI quantitatively assesses the defensiveness of a given site or feature based on the sum of four variables: visibility, elevation, accessibility, and area (Martindale and Supernant 2009). Calculations for each variable are made on a scale of 0- 1, with 0 equalling no defensiveness and 1 equalling maximum defensiveness. Visibility is defined by what can be seen beyond 100 meters from a location over land and water (Martindale and Supernant 2009:194). Elevation is a calculation based on the location’s height above sea level, degree of slope, and the width of the arc where elevation provides the greatest defensive advantage (Martindale and Supernant 2009:195). Our calculations for visibility and elevation do not consider the heights of defensive features, which may have increased the defensibility of a given site. However, this would not have affected our overall results as the viewshed analyses make up for any lack in the DI calculations. Accessibility refers to the

114 proportion of a site’s circumference that allows access given associated natural and architectural constraints (Martindale and Supernant 2009:196). Area is the ratio between site size and 1,000,000 m2, the assumed maximum of a site on the Pacific Northwest Coast (Martindale and Supernant 2009:196). The summed variables, results in a defense index on a scale of 0-4 (Tables 13 and 15), with 4 representing the greatest defensibility. Additional details on how the defense index is calculated are given in Martindale and Supernant (2009). Calculations for features and sites within each site complex, results in a minimum and maximum number for that site complex. As with the viewsheds, we consider the maximum DI to be the most meaningful calculation. We consider any location with a defense index greater than 2 to be moderately to highly defensible.

4.4. Results

The combined viewsheds for the total data set (n=98) illustrate the degree to which the lives of the ancestral Tla’amin were integrated with their marine environment (Figure 24). Not surprisingly, visibility from any location was generally best toward the water. Taking a closer view with the subsample, there are some notable differences in defensibility among the four site complexes (Figures 25-29; Tables 13-15). Viewshed distances and areas range from 6.9 km to 18.4 km and 0.5 km2 to 370 km2, respectively. While not all locations within the site complexes satisfy our minimum requirement of defensibility, the combination of viewsheds from various positions balances out any limitations of component parts. As we explore further below, this strategy of associating lookouts and settlements and allying with neighbouring settlements created an effective defensive network (e.g., Schaepe 2006) for warning of the presence of belligerents well in advance of their arrival at a given settlement.

115 Table 15: Summary of variables and defense index (DI) calculations for sample of sites used in viewshed analyses.

ID land degrees: in Approach water degrees: in Approach (lan >100m in degrees Visibility (water) >100m in degrees Visibility level) above sea (meters Elevation access Degrees Area (m Site radius degrees in angle Elevation ( Visibility ( Elevation ( Approach/access Area ( DI =

)

Emmonds Beach (šɛʔaystən)2 20 170 110 0 115 4 170 15538 162.5 1.41 0.41 0.02 0.31 0.015538 0.75 52 180 75 0 75 4 255 859 23.5 9.66 0.29 0.11 0.15 0.000859 0.55 86 0 360 0 360 10 1 107 7 55.01 1.00 0.61 0.50 0.000107 2.11 75 180 175 0 175 5 70 237 11 24.44 0.49 0.27 0.41 0.000237 1.17 70 160 155 0 155 25 20 297 14 60.75 0.49 0.68 0.53 0.000297 1.70 67 60 90 0 165 20 20 531 16.5 1.10 50.48 0.56 0.73 0.000531 2.39 66 70 160 0 240 15 60 561 15.5 44.06 1.04 0.49 0.55 0.000561 2.08 91 160 130 0 140 2 10 4032 126.5 0.91 0.48 0.01 0.58 0.004032 1.08 Bliss Landing (ǰɛǰɩšʔsiʔəm)2 37 260 50 0 70 2 195 4000 60 1.91 0.23 0.02 0.25 0.004 0.51 Okeover Inlet (toχʷnač)2 59 50 60 0 90 8 40 1442 41 11.04 0.82 0.12 0.67 0.001442 1.61 63 70 70 0 210 8 50 692 10 38.66 1.50 0.43 0.63 0.000692 2.56 39 140 40 0 40 4 40 3913 52.5 4.36 0.22 0.05 0.64 0.00.3913 0.91 Portage Cove (qɛgiyɩn)2 19 210 50 0 120 4 50 16500 98 2.34 0.46 0.03 0.54 0.0165 1.05 76 180 88 0 160 8 88 185 14 29.74 0.60 0.33 0.46 0.000185 1.39 Prideaux Haven (mačɛnay)2 53 60 45 0 40 3 70 660 36 4.76 0.38 0.05 0.52 0.00066 0.96 54 40 30 0 25 3 90 600 41 4.18 0.36 0.05 0.26 0.0006 0.66 48 60 55 0 45 3 90 2022 84.5 2.03 0.39 0.02 0.45 0.002022 0.86 60 30 200 0 200 18 10 844 21 40.60 0.87 0.45 0.66 0.000844 1.98 56 30 250 0 290 20 20 5109 60.5 18.29 1.04 0.20 0.58 0.005109 1.82 68 40 70 0 150 20 40 522 24 39.81 1.36 0.44 0.67 0.000522 2.47 96 0 60 0 70 10 5 860 24 22.62 1.17 0.25 0.88 0.00086 2.29 97 20 45 0 50 8 15 370 12.5 32.62 0.77 0.36 0.79 0.00037 1.93 98 10 30 0 50 8 15 276 9.5 40.10 1.25 0.45 0.76 0.000276 2.45 Note: 1Use IDs to find additional information in Tables 10-12, 14; Figures 22-29.; 2See Appendix A.

116

Figure 24: Combined viewshed of all sites. Map on the left generated from ESRI basemap catalogue using ArcMap 10.3. The ID numbers provide additional information in Tables 10-12, 14-15. Viewshed map on right generated from mosaicked digital elevation model (DEM) using viewshed tool in ArcMap 10.3.

117 4.4.1. Open Shoreline Site Complexes

Emmonds

Overall, the Emmonds site complex is among the most highly defensible in our sample. In particular, the settlement location along an open shoreline results in it having among the lowest DI score (Tables 13 and 15), but its association with the five lookouts and a trench embankment feature compensates for this. The maximum viewing distance from Emmonds is roughly in line with the other settlements, but it has by far the widest area covered of all the site complexes (Tables 13-14; Figure 25A;). The wide composite viewshed means that canoes coming from any direction are visible from the Emmonds lookouts well before any component of the settlement is visible from the water (Figure 25B). In addition, the trail to Okeover could have been used both as an avenue of escape or for runners to transmit information between settlements (e.g., Suttles 1951:322; Tollefson 1996:155). Finally, the various lookouts and settlements on Savary Island to the west of Emmonds could have also served to warn people at Emmonds of attackers (Figure 25, IDs 12, 27, 34, 61, 69, 80) via a signal fire (e.g., Angelbeck 2009:116) or by dispatching a messenger in a canoe.

118

Figure 25: Map A shows combined viewsheds from the Emmonds trench embankment (white), settlement (light grey), and lookouts (medium grey). Map B shows viewsheds from canoes at the points that they are visible from the various components of the Emmonds settlement. The ID numbers provide additional information in Tables 10- 12, 14-15. The viewshed maps were generated from a mosaicked DEM using the viewshed tool in ArcMap 10.3.

119 Bliss Landing

Compared with Emmonds, the Bliss Landing settlement is not highly defensible (Table 15), given its location on the open shoreline and lack of constructed defensive features and lookouts. Although its viewshed distance and area are greater than that of a canoe (Tables 13 and 14), the islands and islets fronting Turner Bay act as a “double-edged sword” for the occupants of the settlement (Figure 26A). That is, while these landforms do provide some level of protection by partially hiding the settlement from observers on the water, they also allow for canoes coming from the north or south to get quite close before they are visible (Figure 26A). Only canoes coming straight at the settlement would be visible in time for the inhabitants to launch an effective defense.

The low-level defensibility of Bliss Landing may be associated with its relatively older occupation history. The site was established by at least 3500 years BP (Tables 12 and 13), this is not known to have been a time of major conflict in the Coast Salish region (Angelbeck 2016; Angelbeck and Grier 2012); other factors seem to have taken priority in selecting the settlement location. In later history, the trail connecting the settlement with the nearby site at Cochrane Bay may have been used as an escape route to Malaspina Inlet or as an early warning system on the movements of potential belligerents on the west side of Malaspina Peninsula.

4.4.2. Inlet Site Complexes

Okeover

As predicted, the extensive view down Okeover Inlet meant the occupants of the Okeover settlement were relatively well protected from invaders (Figure 27A; Table 13). Although limited in areal extent due to the linear geography and relatively steep sides of the

120

Figure 26: Map A shows the viewshed from the Bliss settlement. Map B shows the viewsheds from canoes at the point that they are visible from the Bliss settlement. The ID numbers provide additional information in Tables 10-12, 14- 15. The viewshed maps were generated from a mosaicked DEM using the viewshed tool in ArcMap 10.3.

121 inlet, these views allow observers at the settlement or lookouts to see a canoe approximately 7.5 km up the inlet, and approximately 5 km before the settlement is visible from the water (Figure 27). The defensibility of the Okeover complex is also indicated by its maximum defense index associated with one of the lookouts.

Not reflected in the quantitative measures is how the Okeover site complex situates within a larger defensive network. In particular, a canoe or group of canoes heading for the head of Okeover Inlet would pass by four settlements and three lookouts in Malaspina Inlet that could potentially provide early warning. Of particular importance is the settlement located on the isthmus joining Coode Island and Malaspina Peninsula (Figure 27; Table 10, ID 13). Most of Malaspina Inlet is visible from this settlement and its associated lookout (Table 11, ID 58). The settlement, in turn, is visible from the lookout on the east side of the Okeover settlement, which means warning could be given by fire or messenger well before an attacking force would be visible to the sites at the head of the inlet.

Portage Cove

Quantitatively, the Portage Cove settlement is not highly defensible (Tables 13 and 15). However, this is in part made up for by its geographical context, which potentially situates it in a larger defensive network. In particular, although the viewshed from the settlement is restricted to the northwest, the lookout located northeast of the settlement significantly augments what is visible out toward Homfray Channel and Desolation Sound (Figure 28A). By contrast, from the point of view of a canoe, the settlement is only visible right near the entrance to Portage Cove, ~350 m from the settlement. On the southeast side of the settlement the situation is quite different. Visibility down Lancelot Inlet is restricted, allowing for canoes coming up the inlet on the west side to get within 1 km before they are visible. However, similar to the Okeover complex, accessing the inlet necessitates passing by the defensive network in Malaspina Inlet making a surprise attack from this side highly unlikely.

122

Figure 27: Map A shows the combined viewsheds from the Okeover settlement (white) and lookouts (light grey). Map B shows the viewshed from the canoe at the point it is visible from the Okeover settlement and lookouts. The ID numbers provide additional information in Tables 10-12, 14-15. The viewshed maps were generated from a mosaicked DEM using the viewshed tool in ArcMap 10.3.

123

Figure 28: Map A shows the combined viewsheds from the Portage Cove settlement and associated lookout. Map B shows the viewsheds from the canoes at the point that they are visible from the settlement or lookout. The ID numbers provide additional information in Tables 10-12, 14-15. The viewshed maps were generated from a mosaicked DEM using the viewshed tool in ArcMap 10.3.

124 4.4.3. Cove Site Complexes

Prideaux Heaven

Our analyses support the observations of Vancouver (Lamb 1984:604) and Menzies (1923:66-68) that the Prideaux Haven settlements and associated redoubts are highly defensible locations. All are located in coves with entrances fronted by numerous islands and islets that both partially conceal the settlements and force access along certain routes that could be monitored. In addition, the associated lookouts significantly extend the otherwise restricted viewsheds from the settlements (Figure 29A; Tables 13 and 14). From the lookouts, canoes are visible almost 13 km up Homfray Channel and 6 km up Waddington Channel. At these distances, the three settlements are not visible from observers on the water (Figure 29B). The benefits and/or necessity of the lookouts to the three Prideaux settlements is highlighted by the medium to low DI scores for two of the three settlements (Tables 12 and 15).

125

Figure 29 Map A shows the combined viewsheds from the Cove settlements (white) and lookouts (light grey). Map B shows the viewsheds from the canoes at the point they are visible from the Cove settlements and lookouts. The ID numbers provide additional information in Supplemental Tables 10-12, 14-16. The viewshed maps were generated from a mosaicked DEM using the viewshed tool in ArcMap 10.3.

126 4.5. Discussion

The threat of conflict was interwoven into the daily lives of the descendant and ancestral Tla’amin people, especially in the recent past. The ethnohistoric accounts at two of the site complexes in our subsample attest to the prevalence of conflict in recent history, coinciding with the coast-wide period of conflict noted for this time. Radiocarbon dates from lookouts and other features suggest that conflict was present in Tla’amin territory from approximately 900 years BP, but primarily between 600 to 100 years BP (Table 12). The presence of conflict in Tla’amin territory during the period of relative calm between the early (1600-500 years BP) and more recent (250-100 years BP) periods of conflict documented for the Salish Sea suggest a more complicated sociopolitical climate across the region. We propose that the continuation of conflict at the northern end of the Salish Sea reflects an earlier start for Laich-kwil-tach expansionism. Abutting the cultural boundary between the Southern Kwakwaka’wakw and the Coast Salish world, the Tla’amin, and other northern Coast Salish, would have felt the initial impacts of Laich- kwil-tach territorializing well before other Salishan groups to the south. The hiddenness of the Prideaux settlements, some of the most northern Tla’amin settlements, was likely a response to this expansionist aggression.

Prior to Laich-kwil-tach movement south, the violent conflict pervading the Salish Sea generally and Tla’amin territory specifically may have had a more localized genesis— internecine violence associated with a larger sociopolitical struggle fueled by regional population rise and conflict between centralizing and decentralizing power interests inherent in Coast Salish social structure (Angelbeck and Grier 2012; Ritchie et al. 2012). Using radiocarbon data in combination with dated site types for the lower Fraser Valley, Ritchie et al. (2012) show a general population rise beginning around 6000 years ago, with a significant jump between 2250 and 2000 cal. BP. We suggest that this population increase shown for the lower Fraser Valley was part of a regional social phenomenon as aggregation into large, multi-family plankhouses at approximately 2300 years BP occurred among both coastal and riverine populations in the Salish Sea.

This move into large houses, among other things, increased both the individual families’ and the aggregated household’s visibility, thereby symbolically communicating and reifying territorial and tenurial claims that were increasingly under stress due to population rise and its attendant territoriality. Subsequently, these large houses became social fields that fueled otherwise latent tendencies toward centralized power. Concomitant with this, the houses, along

127 with other forms of symbolic communication (e.g., cranial modification, [Angelbeck and Grier 2012]), became new media for differentiating status. Status that had long been symbolically expressed among the ancestral Coast Salish through various means such as mortuary practices (Carlson and Hobler 1993; Coupland et al. 2016) and the use of personal adornment (Coupland et al. 2016; La Salle 2008). By 1600 years BP the ramped-up signalling behaviour led to an increase in violent conflict between ancestral Coast Salish communities, which has been argued was a form of leveling mechanism in response to centralizing power interests (Angelbeck and Grier 2012:564-566). We contend that this conflicted sociopolitical climate motivated the earlier period of violence in ancestral Tla’amin territory with the more recent local pressure from the Laich-kwil-tach adding additional fuel to an already adversarial climate.

In response to this violent atmosphere, our results demonstrate the ways in which the ancestral Tla’amin created defensive networks using both natural and modified landscapes. Such networks increased the defensibility of settlement complexes by providing a variety of lookouts, escape routes, and hideouts. The importance of the network system is highlighted in the Prideaux case where, employing the complex geography of the area to their advantage, the ancestral Tla’amin used other locations to function as the ‘eyes’ of the settlements sited in the hidden coves. In other cases, adjacent and nearby settlement/lookout configurations connected through social networks further enhanced the defensiveness of some locations.

More generally, these factors of Tla’amin defensiveness bring to the fore the singular nature of conflict in coastal environments. Combining the medium of water with built environments and a convoluted shoreline fronted by myriad islands and islets, and crenulated and incised with bays, coves, and long straight and winding inlets, provides strategic opportunities not necessarily transferable to purely terrestrial contexts. A lake environment may be analogous to a certain degree but, ultimately, it is a closed circuit that lacks the open- endedness of a saltwater shoreline. The complex vertical and horizontal geographies of many coastal strips afford possibilities for multi-directional surprise attack or defense, early warning viewsheds, varied refugia, and the construction of defensive systems combining modified and natural structures. Coastal strips also have spatial and temporal constraints that influence the decision making of the attacker. For example, when moving through space by canoe, the inherent directionality of coastal landforms constrains movement to a limited number of choices (e.g., Mackie 2001:12); thereby increasing the strategic advantage of the defender. The temporal constraints of tidal water also fetter movement through marine space—the ebb and

128 flow of tides are limiting conditions that can affect speed and access, especially through narrow passages only navigable at slack.

While conflict seems to have been part of the social landscape of the Tla’amin for generations, there was also variation in the degree to which it influenced site settlement. For instance, in places like Emmonds and Prideaux, defense seems to have been a primary consideration. In contrast, our quantitative analyses and the lack of defensive features suggests that Bliss Landing was settled some three millennia ago with relatively little concern for defense. That it continued to be used throughout the period of heightened conflict in the late Holocene may reflect the existence of social nets that connected this settlement to other defensive networks on the east side of Malaspina Peninsula. That the Bliss Landing occupants chose not to abandon this place, however, despite the heightened conflict around them, also suggests that their powerful, multi-generational connection to place overrode other considerations.

The history of conflict in Tla’amin territory has revealed the long use of defensive networks, a tactic for increasing strategic advantage that has been noted in other areas of the Salish Sea region (Angelbeck 2016; Schaepe 2006). Our study also revealed that defensive networks mirror the social networks that connected descendant Tla’amin communities and other Coast Salish communities throughout the region. This is highlighted in the trail connections between different site complexes (e.g., Emmonds and Okeover), shared lookouts (e.g., Prideaux Haven), and adjacent and nearby site complexes acting as early warning systems by extending viewsheds to areas otherwise invisible (e.g., Emmonds and Savary). This connection between the landscape and social structure highlights the important role of built environments in the expression of territorial and tenurial claims among descendent and ancestral Tla’amin and Coast Salish populations; especially during periods of conflict.

Defense of territorial and tenurial interests is arguably the most overt form of territoriality among non-state societies. Offensive action can result in the expansion of territory, but it can also be for the purposes of seizing resources, avenging a real or perceived wrong, or simply to show strength, none of which leave a significant or unambiguous archaeological signature. By contrast, defense of home or community is an unequivocal statement of claimed space that can be seen archaeologically in landscape modifications and settlement patterning. The latter is highlighted through visibility analyses that move beyond the single site perspectives to consider their positions in the larger landscape. Taking this broader view gives substance to both the

129 interconnectedness of ancestral built environments of non-state societies and their territorial and tenurial interests through expressions of territoriality.

130 Chapter 5. Conclusion

The built environments and social structures of non-state societies are the physical and abstract geographies that reflect and frame interaction across various spatial and temporal scales. As both the media and outcomes of social practices, these constructs shape the mundane and ceremonial practices of the human experience and, as fields of symbolic communication, are occasionally reconfigured and redefined to validate or challenge changes in practice. The contrasting roles of physical and social structures as extensions of social continuity and change is highlighted among non-state societies through expressions of territory, tenure, and territoriality. For populations that record, transmit, and validate their history through the oral record, territorial and tenurial claims are given additional substance and legitimacy through the use of built environments and social relations as physical and abstract representations of those claims. Linking claims to built environments and social relations is a form of territoriality that helps to create the physical and abstract cultural landscapes that give people a sense of place and history.

House remains are among the best material proxies of territoriality among non- state societies. This is attested to by the importance given the house in much of the anthropological, sociological, and archaeological literature that examines the role of built environments in expressions of economic and sociopolitical interests (Blanton 1994; Bourdieu 1977; Coupland et al. 2009; Dovey 2005; Rapoport 1969; Madella et al. 2013). An integral assumption of much of this research holds that the fundamental attributes of culturally defined social institutions or groups (e.g., kin groups, social classes, households, and extended families) are reflected in, and expressed through, the constructed places associated with those institutions or groups, thereby affecting interactions within, or in relation to, those places across space and through time (Blanton 1994:9-20, Coupland et al. 2009:85-92; Dovey 2005:291-292; Hillier 1996:371-371). Houses are seen to objectify associated social institutions, groups, or roles and, in an archaeological context, become the material correlates of those cultural units (Coupland 2013; Coupland et al. 2009; Lepofsky et al. 2009; Sobel et al. 2006; Springer and Lepofsky 2011).

131 Given the intrinsic relationship between houses and their occupants, inferences can be drawn on how these built environments were used to communicate the various local and extra-local concerns of the occupants. The study presented here used this fundamental relationship to show that large above-ground houses became widely adopted in the Salish Sea region circa 2300 years BP, in part, to facilitate communication of territorial and tenurial interests following a regional population rise (Ritchie et al. 2016) and its attendant territoriality. The modular nature of the above- ground house was better suited than in-ground houses to the structured fluidity of ancestral Coast Salish kinship relations and social networks. Above-ground houses were more adaptable than the in-ground house to shifts in the sociopolitical climate of the region. The highly visible nature of above-ground architecture was also better suited to signaling territorial and tenurial interests in the surrounding area. The importance of the above-ground house as a signaling device post-2300 years BP is underscored by its adoption among lower Fraser Valley communities where the in-ground house was better suited to the colder winter climate.

The social networks linking families and households among the ancestral and descendent Coast Salish populations were also a means through which to express territorial and tenurial interests. In Chapter 3, the role of these networks in expressing abstract territorial and tenurial claims is explored through the spatial and temporal distributions of obsidian toolstone throughout the Salish Sea region. The concepts of territory and tenure are expanded to include the development and continuing maintenance of the networks that moved the toolstone. Thus, the relationships that facilitated access to toolstone from distant locales and the toolstone itself were, respectively, forms of social and economic capital for the families and households participating in obsidian trade. Through legitimation during witnessed events where these forms of capital would have been expressed and displayed, they were transformed into symbolic capital (Bourdieu 1989) that represented the abstract territorial and tenurial claims of access to the extra-local resource.

The final study moves back into the realm of built environments as physical representations of territorial and tenurial claims. Focusing in on the Northern Coast Salish-Tla’amin Nation, defensive strategies are examined as a form of territoriality. Among the Tla’amin, and other Coast Salish communities, defensive territoriality played out in defensive networks of settlements, lookouts, and other tactical features. These

132 defensive networks were both localized and interconnected with other defensive networks, thereby extending the expression of territoriality out from one complex of sites into adjacent and other nearby complexes of sites. What linked defensive networks were the social networks formed out of group exogamy and bilateral kinship that connected families, multifamily households, and settlements at local and regional scales. Through these links, the social networks were made manifest in the modified places that comprised the defensive networks and, along with major settlements and other features on the landscape, were symbolic and physical expressions of territorial and tenurial claims to the area.

The intersection of social networks and defensive networks highlights the importance of the network strategy among ancestral and descendent Coast Salish people. As socioeconomic conduits, these networks were fundamental to Coast Salish interaction both locally and regionally. As the three case studies show, the networking strategy was critical to expressing claims of proprietary/communal interest, rights of access, and defense of place. Despite the intangible nature of intention, the distributions of constructed places and artifacts are the material correlates of past actions and give some indication of the socioeconomic and political interests of the associated ancestral groups.

The archaeological record of the Salish Sea provided an opportunity to examine the nature of territory, tenure, and territoriality among non-state societies at both regional and local scales. The Salish Sea is by no means a unique geography for such an undertaking. The built environments and artifact distributions that define the archaeological records of other non-state societies would be equally amenable to this kind of investigation. It is generally accepted that interaction between ancestral communities in many parts of the world extended beyond the boundaries of specific sites or even larger geographic areas (Bolnick and Smith 2007; Braun and Plog 1982); especially in the context of trade and exchange (Oka and Kusimba 2008). Given this, by considering these kinds of extra-local interactions as forms of territoriality that serve to extend territorial and tenurial claims grounded in affinal and kinship relations, allows for a more nuanced examination of how non-state societies participated socially, politically, and economically at regional and interregional scales.

133 Considering the wider context of ancient interaction also allows for analyses that move beyond current political boundaries that arbitrarily divide traditional territories, turning previously connected communities into two or more border communities. The issue of modern borders redefining previously unbounded space is highlighted by Starks et al. (2011) in their study of the effects of borders on Indigenous communities in the United States:

For them, [Indigenous communities] the two sides of an international boundary may compose a single contiguous space: a homeland, or a network of relationships reaching far back to a distant past, or a set of natural and cultural resources that are used in common and need to be protected and sustained, or perhaps the piece of earth out of which they— the people themselves—originally came (Starks et al. 2011:1).

Among researchers focusing on the ancestral Coast Salish, the International boundary that follows the 49th parallel has often been used, either intentionally or unintentionally, as a boundary defining study areas both north (e.g., Bocinsky 2014) and south (e.g., Stein et al. 2003) of the border. Although accessing certain kinds of data such as unpublished technical reports from state and provincial regulatory has its difficulties, it is not insurmountable. To examine the ancestral Coast Salish within a broader context of socioeconomic connections it is necessary to look beyond modern impediments to uncover the ancestral cultural landscape. As shown Chapter 3 with the distribution of obsidian toolstone in the Salish Sea region, it is clear the descendant and ancestral Coast Salish did not use the 49th parallel as a point of reference for regional and inter-regional interaction. Rather, their social networks were the means through which access to their wider world was achieved. The Coast Salish cognitive map would have been defined by these networks, and by their physical and symbolic modifications of the environment.

The extent to which the descendant and ancestral Coast Salish used their physical and social environments to communicate a myriad of interests underscores both the importance of networks and the deep connection to place that existed, and continues to exist, in Coast Salish communities. The archaeology and ethnohistory of the Coast Salish generally and the Tla’amin specifically, provided an opportunity to examine the roles of territory, tenure, and territoriality among non-state societies. Using a landscape perspective, the three case studies show that past claims and access to

134 land and resources was intertwined with social structure and the practices and places associated with creating, maintaining, and challenging that social structure. In addition, the importance of a landscape perspective for understanding both the physical and social contexts of territorial and tenurial claims is highlighted in the research. Although ancestral non-state societies were connected to specific places, they were also woven into the intra and interregional fabric that can only be examined by looking beyond individual places to consider interactions across the broader cultural landscapes.

135 References

Abbott, Donald N. (1972). The Utility of the Concept of Phase in the Archaeology of the Southern Northwest Coast. Syesis 5:267-278.

Acheson, Steven, S. Cassidy, and S. Claxton. (1975). Report of the Archaeological Survey of the Southwestern Gulf of Georgia. Conducted under Heritage Inspection Permit 1975-0005. Report on file with the Archaeology Branch of British Columbia, Victoria.

Acheson, Steven, and Sid Riley. (1976). Gulf of Georgia Archaeological Survey: Powell River and Sechelt Regional Districts. Conducted under Heritage Inspection Permit 1976-0007. Report on file with the Archaeology Branch of British Columbia, Victoria.

Alden, James M. (1858). “Village, Nanaimo Bay, Vancouver’s Island” 1858. Original image on file with the British Columbia Archives at the Royal British Columbia Museum, Victoria.

Alexander, W. (1798). Deserted Indian Village, Georgia Strait. Edward E. Ayer Art Collection. VAULT oversize Ayer Art Alexander, drawing #17. Newberry Library, Chicago.

Ames, Kenneth M. (1996). Life in the Big House: Household Labor and Dwelling Size on the Northwest Coast. In People Who Lived in Big Houses: Archaeological Perspectives on Large Domestic Structures, edited by G. Coupland and E. G. Banning, pp. 131-150. Prehistory Press, Madison, Wisconsin.

Ames, Kenneth M. (2005). The Place of Ozette in Northwest Coast Archaeology. In Ozette Archaeological Project Research Reports Volume III: Ethnobotany and Wood Technology, edited by David L. Welchel, pp. 9-24. Washington State University Department of Anthropology Reports of Research 68, National Park Service, Northwest Regional Office, Seattle.

Andrefsky, William Jr. (2005). Lithics: Macroscopic Approaches to Analysis. 2nd ed. Cambridge Manuals in Archaeology. Cambridge University Press, Cambridge.

Angelbeck, Bill. (2007). Conceptions of Coast Salish Warfare, or Coast Salish Pacifism Reconsidered: Archaeology, Ethnohistory, and Ethnography. In Be of Good Mind: Essays on the Coast Salish, edited by Bruce G. Miller, pg. 260-283. University of British Columbia Press, Vancouver.

Angelbeck, Bill. (2009). “They Recognize no Superior Chief”: Power, Practice, Anarchism and Warfare in the Coast Salish Past. Unpublished Doctoral dissertation, Department of Anthropology, University of British Columbia, Vancouver.

Angelbeck, Bill. (2016). The Balance of Autonomy and Alliance in Anarchic Societies: The Organization of Defences in the Coast Salish Past. World Archaeology 48(1):51-69.

136 Angelbeck, Bill, and Colin Grier. (2012). Anarchism and the Archaeology of Anarchic Societies: Resistance to Centralization in the Coast Salish Region of the Pacific Northwest Coast. Current Anthropology 53:547-587. Archaeology Branch of British Columbia.

Angelbeck, B. and E. McLay. (2011). The Battle at Maple Bay: The Dynamics of Coast Salish Political Organization through Oral Histories. Ethnohistory 58(3):359-392.

Aoyama, Kazuo. (2014) Symbolic and Ritual Dimension of Exchange, Production, Use, and Deposition of Ancient Maya Obsidian Artifacts. In Obsidian Reflections: Symbolic Dimensions of Obsidian in Mesoamerica, edited by Marc N. Levine and David M. Carballo, pp. 127-158. University Press of Colorado, Boulder.

Archaeology Branch. (2015a). British Columbia Archaeology Branch Site Form – EdSn- 35. Form on file with the British Columbia Archaeology Branch, Victoria.

Archaeology Branch. (2015b). British Columbia Archaeology Branch Site Form – DkSf-2. Form on file with the British Columbia Archaeology Branch, Victoria.

Archaeology Branch. (2015c). British Columbia Archaeology Branch Site Form – DhRs- 1. Form on file with the British Columbia Archaeology Branch, Victoria.

Archaeology Branch. (2015d). British Columbia Archaeology Branch Site Form – DgRw- 4. Form on file with the British Columbia Archaeology Branch, Victoria.

Archaeology Branch. (2016a). British Columbia Archaeology Branch Site Form – DgRs- 14. Form on file with the British Columbia Archaeology Branch, Victoria, British Columbia.

Archaeology Branch. (2016b). British Columbia Archaeology Branch Site Form – DhRo- 10. Form on file with the British Columbia Archaeology Branch, Victoria, British Columbia.

Archaeology Branch. (2016c). British Columbia Archaeology Branch Site Form – DhRo- 28. Form on file with the British Columbia Archaeology Branch, Victoria, British Columbia.

Archaeology Branch. (2018). British Columbia Archaeology Branch Site Form – EaSe- 23. Form on file with the British Columbia Archaeology Branch, Victoria, British Columbia.

Baker, M. J. (2003). An Equilibrium Model of Land Tenure in Hunter-Gatherer Societies. Journal of Political Economy 111(1):124-173.

Bar, Moshe. (2004). Visual Objects in Context. Nature Reviews Neuroscience 5:617- 629.

Barnett, Homer G. (1935-1936). Unpublished field notes on file at the University of British Columbia Library Special Collections Division, Vancouver.

137 Barnett, Homer G. (1938). The Coast Salish of Canada. American Anthropologist, New Series 40:118-141.

Barnett, Homer. (1944). Underground Houses on the British Columbian Coast. American Antiquity 9:265-270.

Barnett, Homer. (1955) The Coast Salish of British Columbia. The University of Oregon Press, Eugene.

Beattie, Owen. (1972). Salvage archaeology at Bliss Landing. In Salvage ’71: Reports on Salvage Archaeology Undertaken in British Columbia in 1971, edited by Roy Carlson, pp. 23-39. Department of Archaeology, Simon Fraser University, Burnaby.

Bettinger, Robert L. (1982). Aboriginal Exchange and Territoriality in Owens Valley, California. In Contexts for Prehistoric Exchange, edited by J. E. Ericson and T. K. Earle, pp. 103-127. Academic Press, New York.

Bilton, David Harrison. (2014). Northern, Central, Diversified, Specialized: The Archaeology of Fishing Adaptations in the Gulf of Georgia (Salish Sea), British Columbia. Unpublished Doctoral dissertation, Department of Anthropology, University of Toronto, Toronto.

Blackburn, Roderick H. (1986). Okiek Resource Tenure and Territoriality as Mechanisms for Social Control and Allocation of Resources. Sprache und Geschichte in Afrika 7.1:61-82

Blake, Michael. (2004). Fraser Valley Trade and Prestige as Seen from Scowlitz. In Complex Hunter-Gatherers: Evolution and Organization of Prehistoric Communities on the Plateau of Northwestern North America, edited by William C. Prentiss and Ian Kuit, pp. 103-112. The University of Utah Press, Salt Lake City.

Blanton, Richard E. (1994). Houses and Households: A Comparative Study. Plenum Press, New York.

Bloomfield, Gladys. (2005). Magnetic Isle: Gladys Bloomfield’s Savary. Savary Island Heritage Society, Whonnock.

Blukis Onat, Astrida R., M. E. Morgenstein, P. LeTourneau, R. P. Stone, J. Kosta, and P. Johnson. (2000). Archaeological Investigations at Stuwe’yuq – Site 45KI464, Tolt River, King County, Washington. BOAS Inc., Seattle.

Boas, Franz. (1890). The Lku’ngen: Houses and Boats. In Sixth Report on the North- Western Tribes of Canada, pp. 11-14. British Association for the Advancement of Science, Leeds Meeting, 1890. Office of the Association Burlington House, London.

Bocinsky, Kyle R. (2014). Extrinsic Site Defensibility and Landscape-Based Archaeological Inference: An Example from the Northwest Coast. Journal of Anthropological Archaeology 35:164-176.

138

Bolnick, Deborah A., and David Glenn Smith. (2007). Migration and Social Structure among the Hopewell: Evidence from Ancinet DNA. American Antiquity 72(4):627- 644.

Borden, Charles E. (1961). Archaeological Excavations at Site DjRi 3, Fraser Canyon, B.C. – Summer 1959: A Brief Progress Report. Report on file with the British Columbia Archaeology Branch, Victoria.

Borden, Charles E. (1983). Prehistoric Art of the Lower Fraser Region. In Indian Art Traditions of the Northwest Coast, edited by Roy L. Carlson, pp. 131-165. Archaeology Press, Department of Archaeology, Simon Fraser University, Burnaby.

Bourdieu, Pierre. (1977). Outline of a Theory of Practice. Translated by Richard Nice. Cambridge University Press, Cambridge.

Bourdieu, Pierre. (1989). Social Space and Symbolic Power. Sociological Theory 7:14- 25.

Bourdieu, Pierre. (1990). The Logic of Practice. Translated by Richard Nice. Stanford University Press, Stanford.

Bourdieu, Pierre. (2005). Habitus. In Habitus: A Sense of Place, edited by Jean Hillier and Emma Rooksby, pp. 43-50. Ashgate, Aldershot, UK

Braun, David P., and Stephen Plog. (1982). Evolution of Tribal Social Networks: Theory and Prehistoric North American Evidence. American Antiquity 47(3):504-525.

Brolly, Richard, Guy Cross, Jackie Holecka, Karla Kusmer, Dana Lepofsky, Heather Pratt, June Ryder, and Shannon Wood. (1996). Archaeological Investigations at the Beach Grove Site (DgRs 1), Tsawwassen, B.C. Conducted under Heritage Conservation Act Permit 1995-016. Report on file with the British Columbia Archaeology Branch, Victoria.

Brolly, Richard, and Bob Muir. (1993). 1993 Archaeological Investigations at the Locarno Beach Site (DhRt 6), Vancouver, B.C. Conducted under Permit 1993- 104. Report on file with the British Columbia Archaeology Branch, Victoria.

Brown, Doug, and Nicole Oakes. (2011). 2009/2010 Coquitlam Lake Archaeological Inventory and Site Assessment. Conducted under Heritage Conservation Act Permit 2008-488. File on report with the British Columbia Archaeology Branch, Victoria.

Bryan, Alan Lyle. (1955). An Intensive Archeological Reconnaissance in the Northern Puget Sound Region. Unpublished Master’s thesis. Department of Anthropology, University of Washington, Seattle.

Bryan, Alan Lyle. (1963). An Archaeological Survey of Northern Puget Sound. Occasional papers of the Idaho University Museum Volume 11, Pocatello.

139 Burley, David. (1989). Senewélets: Culture History of the Nanaimo Coast Salish and the False Narrows Midden. Royal British Columbia Museum, Memoir No. 2. Royal British Columbia Museum, Victoria.

CARD (Canadian Archaeological Radiocarbon Database). (2016). DfRu-8. Available at: http://www.canadianarchaeology.ca/. Accessed on: February 2016.

Carlson, Keith Thor. (2001). Intercommunity Conflict. In A Sto:lō Coast Salish Historical Atlas, edited by K.T. Carlson, pp. 48-49. Douglas & McIntyre, Vancouver.

Carlson, Roy. (1986). The 1986 Excavations at the Canal Site (DeRt 1 and 2). Simon Fraser University Pender Project. Conducted under Heritage Investigation Permit 1986-0007. Report on file with the Archaeology Branch of British Columbia, Victoria.

Carlson, Roy. (1994). Trade and Exchange in Prehistoric British Columbia. In Prehistoric Exchange Systems in North America, edited by Timothy G. Baugh and Jonathon E. Ericson, pp. 307-361. Plenum Press, New York.

Carlson, Roy. (1996). Early Namu. In Early Human Occupation in British Columbia, edited by Roy L. Carlson and Luke Dalla Bona, pp. 83-102. UBC Press, University of British Columbia, Vancouver.

Carlson, Roy, and Philip M. Hobler. (1993). The Pender Canal Excavations and the Development of Coast Salish Culture. BC Studies 99:25-52.

Carmichael, Colin, Lisa Codd, Paul Ewonus, Simon Kaltenrieder, Mike Rousseau, Lisa Seip, and Mike Will. (2001). Archaeological Monitoring for Interfor’s Port Hammond Cedar Dry Kiln Replacement Project at the Port Hammond Site (DhRp 17), Lower Fraser River, B.C., 2000 – 2001. Conducted under Heritage Conservation Act Permit 2000-292. Report on file with the British Columbia Archaeology Branch, Victoria.

Chapman, Margo. (1974). A Brief Report on Excavations at the O’Connor Site (EeSu 5), Near Port Hardy, B.C. Conducted under permit 1973-18. Report on file with the British Columbia Archaeology Branch, Victoria.

Charles, Douglas K. (2010). Riverworld: Life and Meaning in the Illinois Valley. In Hopewell Settlement Patterns, Subsistence, and Symbolic Landscapes, edited by A. Martin Byers and DeeAnne Wymer, pp. 19-36. University Press of Florida, Gainesville.

Chatters, James C. (with contributions by Virginia Butler, Karin A. Hoover, David F. Rhode, and Elizabeth Vance. (1987). Tualdad Altu (45KI59): A 4th Century Village on the Black River, King County, Washington. First City Equities, Seattle.

Chatters, James C. (1989). The Antiquity of Economic Differentiation within Households in the Puget Sound Region, Northwest Coast. In Households and Communities: Proceedings of the Twenty-first Annual Conference of the Archaeological Association of the University of Calgary, edited by Scott MacEachern, David J. W. Archer, and Richard D. Garvin, pp. 168-178. Archaeological Association,

140 Department of Archaeology, University of Calgary, Calgary.

Chisholm, Brian. (1986). Reconstruction of Prehistoric Diet in British Columbia Using Stable Carbon Isotopic Analysis. Unpublished Doctoral Dissertation, Simon Fraser University, Burnaby.

Chou, C. 1997. Contesting the Tenure of Territoriality: The Orang Suku Laut. Bijdragen tot de Taal-, Land- en Volkenkunde 153:605-629.

Clague, John J., and Ron W. Mathewes. (1996). Neoglaciation, Glacier-Damned Lakes, and Vegetation Change in Northwestern British Columbia, Canada. Arctic and Alpine Research 28:10-24.

Clark, Terence Norman. (2010). Rewriting Marpole: The Path to Cultural Complexity in the Gulf of Georgia. Unpublished Doctoral dissertation, Graduate Department of Anthropology, University of Toronto, Toronto.

Codere, Helen. (1990). Kwakiutl: Traditional Culture. In Northwest Coast, edited by Wayne Suttles, pp. 359-377. Smithsonian Institution, Handbook of North American Indians, vol. 7. Washington, D.C.

Cole, Douglas, and David Darling. (1990). History of the Early Period. In Handbook of North American Indians, Vol. 7: Northwest Coast, edited by Wayne Suttles, pp. 119-134. Smithsonian Institution, Washington, D.C.

Collard, Nicole, Hartley Odwak, Paul Ewonus, and Rudy Reimer. (2006). Proposed Swimming Pool and Geothermal Energy System Developments within the Site Boundaries of DhRt-6 (The “Locarno” Site), 4458 NW Marine Drive, Vancouver.

Collins, June McCormick. (1979). Multilineal Descent: A Coast Salish Strategy. In Currents in Anthropology: Essays in Honor of Sol Tax, edited by Robert Hinshaw, pp. 243-254. Mouton, The Hague.

Commisso, Rob, Andrew W. Hickok, Doug Ross, George Kaufmann, Sean P. Connaughton, Kathleen Young, Robyn G. Ewing, Jordan Handley, Emily Wilkerson, Gay Frederick, Curtis Dean, Nicholas Chesworth, Rena Varsakis, David Le Beau, Mélissa Burns, Nikki Chunick, David Luzi, Zoe MacDougall, Stephanie Rhodin, and Nyssa Temmel. (2015). Archaeological Investigations at the St. Mungo Cannery (DgRr- 2) and Glenrose Cannery (DgRr-6) Sites, South Fraser Perimeter Road, Delta, British Columbia. Conducted under Heritage Investigation Permits 2011-0252 and 2011-0280. Report on file with the British Columbia Archaeology Branch, Victoria.

Conaty, Gerald T., and A. Joanne Curtin. (1984). Crescent Beach Monitoring Programme: Final Report. Conducted under Permit 1983-45. Report on file with the British Columbia Archaeology Branch, Victoria.

Connolly, Thomas J., Craig E. Skinner, and Paul W. Baxter. (2015). Ancient Trade Routes for Obsidian Cliffs and Newberry Volcano Toolstone in the Pacific Northwest. In Toolstone Geography of the Pacific Northwest, edited by Terry L. Ozbun and Ron L. Adams, pp. 180-192. Archaeology Press, Simon Fraser

141 University, Burnaby.

Cotterell, Brain, and Johan Kamminga. (1987). The Formation of Flakes. American Antiquity 52(4):675-708.

Cotterell, Brian., Johan Kaminga, and F. P. Dickson. (1985). The Essential Mechanics of Conchoidal Flaking. International Journal of Fracture 29:205-221.

Cookson, Corey, (2013). An Analysis of Site Selection Behaviours and Landscape in the Prince Rupert Harbour Area. Unpublished Master’s thesis, Department of Anthropology, University of Alberta, Edmonton.

Coupland, Gary. (1985). Household Variability and Status Differentiation at Kitselas Canyon. Canadian Journal of Archaeology 9:39-56.

Coupland, Gary. (1996a). This Old House: Cultural Complexity and Household Stability on the Northern Northwest Coast of North America. In Emergent Complexity: The Evolution of Intermediate Societies, edited by J. E. Arnold, pp. 74-90. International Monographs in Prehistory, Ann Arbor.

Coupland, Gary. (1996b). The Evolution of Multi-Family Households on the Northwest Coast of North America. In People Who Lived in Big Houses: Archaeological Perspectives on Large Domestic Structures, edited by G. Coupland and E. B. Banning, pp. 121-130. Monographs in World Prehistory No. 27. Prehistory Press, Madison.

Coupland, Gary. (2006). A Chief’s House Speaks: Communicating Power on the Northern Northwest Coast. In Household Archaeology on the Northwest Coast, edited by Elizabeth A. Sobel, D. Ann Trieu Gahr, and Kenneth M. Ames, pp. 80- 96. Archaeological Series 16. International Monographs in Prehistory, Ann Arbor.

Coupland, Gary. (2013). Household Archaeology of Complex Hunter-Gatherers on the North-west Coast of North America. In The Archaeology of Household, edited by Marco Madella, Gabriella Kovacs, Brigitta Kulcsarne-Berzsenyi, and Ivan Briz I Godino, pp. 45-66. Oxbow Books, Oxford.

Coupland, Gary, David Bilton, Terence Clark, Jerome S. Cybulski, Gay Frederick, Alyson Holland, Bryn Letham, and Gretchen Williams. (2016). A Wealth of Beads: Evidence for Material Wealth-Based Inequality in the Salish Sea Region, 4000- 3500 Cal. B.P. American Antiquity 81(2):294-315.

Coupland, Gary, Terrence Clark, and Amanda Palmer. (2009). Hierarchy, Communalism, and the Spatial Order of Northwest Coast Plank- Houses: A Comparative Study. American Antiquity 74:77-106.

Croes, Dale. (1995). The Hoko River Archaeological Site Complex: The Wet/Dry Site (45CA213), 3,000-1,700 B.P. Washington State University Press, Pullman.

Cybulski, Jerome S. (2014). Updating the Warrior Cache: Timing the Evidence for Warfare at Prince Rupert Harbour. In Violence and Warfare among Hunter- Gatherers, edited by M. W. Allen and T. L. Jones, pp. 333-350. Taylor & Francis

142 Group, New York.

Dalager, B. ( 2005). Distance to the horizon. Available at: http://www.pvv.ntnu.no/~bcd/shadowworld/info/horizon.html. Accessed on: March 15, 2015.

Daugherty, Richard D., J. Jeffrey Flenniken, and Jeanne M. Welch. (1987). A Data Recovery Study of Judd Peak Rockshelters (45-LE-222) in Lewis County, Washington, Report on file with the Washington Department of Archaeology and Historic Preservation, Olympia, WA.

David, Bruno. (2002). Landscapes, Rock-Art and the Dreaming. Leicester University Press, London.

Derr, Kelly M., Kaitlyn Scott, and Colin Grier. (2012). The Old House and the Sea: Current Excavations at a Late Period Plankhouse in the Southern Gulf Islands, Coastal British Columbia. Poster presented at the 65th Annual Northwest Anthropological Conference “The Past is Present”, in Pendleton, Oregon, March 15th – 17th, 2012.

Dixon, J. E., J. R. Cann, and Colin Renfrew. (1968). Obsidian and the Origins of Trade. Scientific American 218(3):38-46.

Dodd, Chris, and Doris Zibauer. (2011). Archaeological Impact Assessment Permit Report, Cloudworks Energy Inc.’s Statlu Creek, Tretheway Creek, Shovel Creek, and Big Silver Creek Waterpower Projects, Harrison Lake, BC. Conducted under Heritage Conservation Act Permit 2008-0279, In-SHUCK-Ch Heritage Permit 2008-002, Seyem’ Qwantlen Heritage Investigation Permit 2009-003, Stó:lō Heritage Investigation Permit, and Sts’ailes Cultural Heritage Investigations Permit 2008-007. Report on file with the British Columbia Archaeology Branch, Victoria.

Dovey, Kim. (2005). The Silent Complicity of Architecture. In Habitus: A Sense of Place, 2nd addition, edited by Jean Hillier and Emma Rooksby, pp. 267-280. Ashgate, Aldershot.

Duff, Wilson. (1952). The Upper Stalo Indians of the Fraser Valley, British Columbia. Anthropology in British Columbia, Memoirs No. 1. British Columbia Provincial Museum Department of Education, Victoria.

Duff, Wilson. (1964). The Indian History of British Columbia: The Impact of the White Man. Royal British Columbia Museum, Victoria.

Earle, Timothy. (2000). Archaeology, Property, and Prehistory. Annual Review of Anthropology 29:39-60.

Eerkens, Jelmer W. (1999). Common Pool Resources, Buffer Zones, and Jointly Owned Territories: Hunter-Gatherer Land and Resource Tenure in Fort Irwin, Southeastern California. Human Ecology 27(2):297-318.

143 Elmendorf, William W. (1971). Coast Salish Status, Ranking, and Intergroup Ties. Southwestern Journal of Anthropology 27:353-380.

Elmendorf, William W. (1992). The Structure of Twana Culture with Comparative Notes on the Structure of Yurok Culture. Research Studies, a Quarterly Publication of Washington State University, Monographic Supplement No. 2. Washington State University, Pullman.

Feinman, Gary M., Kent G. Lightfoot, and Steadman Upham. (2000). Political Hierarchies and Organizational Strategies in the Puebloan Southwest. American Antiquity 65:449-470.

Ferguson, Aaron, Ryan Dickie, Nova Pierson, and Peter Vigneault. (2011). BC Hydro Southern Communities Grid Connection Project: Archaeological Investigations at DkRm-1 and DkRn-1, Harrison – Lillooet Valley, B.C. Conducted under Heritage Conservation Act Permit 2009-338. Report on file with the British Columbia Archaeology Branch, Victoria.

Field, Julia S. (2005). Land Tenure, Competition and Ecology in Fijian Prehistory. Antiquity 79(305):586-600.

Finnamore, Bert, Randy Timothy, Devin Pielle, Drew Blaney, Tanner Timothy, and Betty Wilson. (2016). Ahms gije. Tla’amin Place Names Project, Tla’amin First Nation and the Powell River Museum, Powell River.

Fitzhugh, Ben. (2002). Residential and Logistical Strategies in the Evolution of Complex Hunter-Gatherer on the Kodiak Archipelago. In Beyond Foraging and Collecting: Evolutionary Change in Hunter-Gatherer Settlement Systems, edited by Ben Fitzhugh and Junko Habu, pp. 257-304. Plenum, New York.

Forgeng, Eric, John Maxwell, and Tina Christensen. (2007). Final Report on Systematic Data Recovery at Site DiSh-17 Elsie Reservoir, Vancouver, Island, B.C. Conducted under Heritage Conservation Act Permit 2005-401 and BC Hydro Contract 00022736-001. Report on file with the British Columbia Archaeology Branch, Victoria.

Galm, Jerry R. (1994). Prehistoric Trade and Exchange in the Interior Plateau of Northwestern North America. In Prehistoric Exchange Systems in North America, edited by Timothy G. Baugh and Jonathon E. Ericson, pp. 275-305. Plenum Press, New York.

Gaston, Jeanette L. (1975). The Extension of the Fraser Delta Cultural Sequence into Northwest Washington. Unpublished Master’s thesis, Department of Anthropology, Western Washington State College, Bellingham.

GeoGratis. (2015). GeoGratis: Search, Discover and Download Free Maps, Data and Publications. Available at: http://geogratis.gc.ca/geogratis/search?lang=en. Accessed on March 3, 2015.

144 Gillis, Nichole. (2007). An Investigation of Charles Culture Housepit Deposits at the Ferndale Site (45WH34). Unpublished Master’s thesis, Department of Anthropology, Western Washington University, Bellingham.

Grabert, Garland F. (1983). Ferndale in Prehistory: Archaeological Investigations in the Lower and Middle Nooksack Valley, Occasional Paper No. 19. Center for Pacific Northwest Studies Western Washington University, Bellingham.

Graesch, Anthony P. (2006). Archaeological and Ethnoarchaeological Investigations of Households and Perspectives on a Coast Salish Historic Village in British Columbia. Unpublished Doctoral dissertation, Department of Anthropology, University of California, Los Angeles.

Grier, Colin. (2003). Dimensions of Regional Interaction in the Prehistoric Gulf of Georgia. In Emerging from the Mist: Studies in Northwest Coast Culture History, edited by R. G. Matson, Gary Coupland, and Quentin Mackie, pp. 170-187. UBC Press, University of British Columbia, Vancouver.

Grier, Colin. (2006) Political Dimensions of Monumental Residences on the Northwest Coast of North America. In Palaces of Power in the Americas: From Peru to the Northwest Coast, edited by Jessica Joyce Christie and Patricia Joan Sarro, pp. 141-165. University of Texas Press, Austin.

Grier, Colin. (2014). Landscape Construction, Ownership and Social Change in the Southern Gulf Islands of British Columbia. Canadian Journal of Archaeology 38:211-249.

Grier, Colin, and Bill Angelbeck. (2007). Village Economics at Dionisio Point (DgRv- 003): Report on the 2002 Excavations at the Dionisio Point Site (DgRv-003), Galiano Island, BC. Conducted under Heritage Conservation Act Permit 2002- 115. Report on file with the British Columbia Archaeology Branch, Victoria.

Grier, Colin, and Kim Jangsuk. (2012). Resource Control and the Development of Political Economies in Small-scale Societies: Contrasting Prehistoric Southwestern Korea and the Coast Salish Region of Northwestern North America. Journal of Anthropological Research 68:1-34.

Grier, Colin, Lilian Alessa, and Andrew Kliskey. (2017). Looking to the Past to Shape the Future: Addressing Social-Ecological Change and Adaptive Trade-Offs. Regional Environmental Change 17:1205-1215.

Gunn, R. G. (2011). Eastern Arrernte Rock Art and Land Tenure. Rock Art Research 28(2):225-240.

Gunther, Erna. (1927). Klallam Ethnography. University of Washington Publications in Anthropology 1(5):173-314.

Hall, Dave, Sue Woods, and Rudy Reimer. (2007). An Archaeological Impact Assessment of Pacific Place Development Corporation (Concord Pacific) and the Squamish Nation’s Proposed Master Planned Community at Porteau Cove near Squamish, B.C., Final Permit Report. Conducted under Heritage Conservation

145 Act Permit 2006-233 and Squamish Nation Archaeological Investigation Permit 06-0199. Report on file with the British Columbia Archaeology Branch, Victoria.

Harris, Cole. (1994). Voices of disaster: Smallpox around the Strait of Georgia in 1782. Ethnohistory 41(4):591-626.

Ham, Leonard C., Arlene J. Yip, and Leila V. Kullar. (1986 ). The 1982/83 Archaeological Excavations at the St. Mungo Site (DgRr-2), North Delta, British Columbia. Conducted under Heritage Conservation Act Permit 1982-21. Report on file with the British Columbia Archaeology Branch, Victoria.

Hillier, Bill. (1996). Space is the Machine: A Configurational Theory of Architecture. Cambridge University Press, Cambridge.

Hillier, Jean. (2005). Mind the Gap. In Habitus: A Sense of Place, 2nd edition, edited by Jean Hillier and Emma Rooksby, pp. 144-171. Ashgate, Aldershot.

Hillier, Jean, and Emma Rooksby (eds.). (2005). Introduction. In Habitus: A Sense of Place, 2nd edition, pp. 3-25. Ashgate, Aldershot.

Hill-Tout, Charles. (1978a). The Salish People. The Local Contribution of Charles Hill- Tout, Volume III: The Mainland Halkomelem, edited with an introduction by Ralph Maud. Talonbooks, Vancouver.

Hill-Tout, Charles. (1978b). The Salish People. The Local Contribution of Charles Hill- Tout, Volume II: The Squamish and the Lillooet, edited with an introduction by Ralph Maud. Talonbooks, Vancouver.

Hill-Tout, Charles. (1978c). The Salish People. The Local Contribution of Charles Hill- Tout, Volume IV: The Sechelt and South-Eastern Tribes of Vancouver Island, edited with an introduction by Ralph Maud. Talonbooks, Vancouver.

Hodder, Ian. (1974). Regression Analysis of some Trade and Marketing Patterns. World Archaeology 6:172-189.

Hodgetts, Lisa. (2010). Subsistence Diversity in the Younger Stone Age Landscape of Varangerfjord, northern Norway. Antiquity 84(323):41-54.

Hoffmann, Tanja, Anna Barran, Amy Homan, Stephanie Huddlestan, Emily Wilkerson, Alejandra Diaz, Sean P. Connaughton, and Celia Nord. (2010). Archaeological Excavations at DhRp-52, Final Report. Conducted under Heritage Investigation Permit 2007-097. Report on file with the Archaeology Branch of British Columbia, Victoria.

Howe, D. Geordie. (2006). Proposed Recreational/Residential Development, Block 18, District Lot 2469, Plan 2323, at Potts Point, Port Graves, Gambier Island Archaeological Impact Assessment. Conducted under Heritage Conservation Act Permit 2006-204 and Squamish Nation Archaeological Investigation Permit 06- 0117. Report on file with the British Columbia Archaeology Branch, Victoria.

Hughes, Richard E. (1990). Letter to Grady H. Caulk of x-ray fluorescence results,

146 appended to Cultural Resource Inventory Record Form – Big Rock Creek Obsidian Point Isolate, Wind River Ranger District. Form on file with the Washington Department of Archaeology and Historic Preservation, Olympia.

Hutchings, Richard M. (2004). Mid-Holocene River Development and South-Central Pacific Northwest Coast Prehistory: Geoarchaeology of the Ferndale Site (45WH34), Nooksack River, Washington. Unpublished Master’s thesis, Department of Anthropology, Western Washington University, Bellingham.

Ingold, Timothy. (1986). The Appropriation of Nature: Essays on Human Ecology and Social Relations. University of Iowa Press, Iowa City.

Jackley, Julia. (2012). Weaving the Histories of Klehkwahnnohm: A Tla’amin Community in Southwest British Columbia. Unpublished Master’s thesis, Department of Archaeology, Simon Fraser University, Burnaby.

James, Malcolm A., Jeff Bailey, and John M. D’Auria. (1996). A Volcanic Glass Library for the Pacific Northwest: Problems and Prospects. Canadian Journal of Archaeology 20:93-122.

Jenness, Diamond. (1934). The Saanich Indians of Vancouver Island. Unpublished manuscript. University of British Columbia: Special collections.

Johnson, Sarah Elizabeth, and Dana Lepofsky. (2009). Report of Survey and Test Excavations in Grace Harbour, Malaspina Inlet, and Lancelot Inlet. Conducted under Heritage Conservation Act Permit 2008-0299. Report on file with the Archaeology Branch of British Columbia, Victoria.

Johnstone, Dave. (1991). The Function(s) of a Shellmidden Site from the Southern Strait of Georgia. Unpublished Master’s thesis, Department of Archaeology, Simon Fraser University, Burnaby.

Jones, Siân. (1999). Historical Categories and the Praxis of Identity: The Interpretation of Ethnicity in Historical Archaeology. In Historical Archaeology: Back from the Edge, edited by Pedro Paulo A. Funari, Martin Hall and Siân Jones, pp. 219-232. Routledge London.

Keeley, L. H. (1996). War before Civilization. Oxford University Press, Oxford.

Keddie, Grant. (1984). Fortified Defensive Sites and Burial Cairns of the Songhees Indians. The Midden 16(4):7–9.

Keddie, Grant. (1996). Aboriginal Defensive Sites. Discovery. Available at: http://www.royalbcmuseum.bc.ca/Content_Files/Files/AboriginalDefensiveSitesar ticleOct2006.pdf]. Royal British Columbia Museum, Victoria.

Kennedy, Dorothy. (2007). Quantifying “Two Sides of a Coin”: A Statistical Examination of The Central Coast Salish Social Network. BC Studies 153:3-34.

Kennedy, Dorothy, and Randy Bouchard. (1983). Sliammon Life, Sliammon Lands. Talonbooks, Vancouver.

147

Kennedy, D. and R. Bouchard. (1990). Northern Coast Salish. In Northwest Coast, edited by William Suttles, pp. 441-452. Handbook of North American Indians, vol. 7. Smithsonian Institution, Washington, D.C.

Klokler, Daniela. (2017). Shelly Coast: Constructed Seascapes in Southern Brazil. Hunter Gatherer Research 3:87-105.

Krisciunas, Kevin, and Don Carona. (2015). At What Distance can the Human Eye Detect a Candle Flame? Poster presented at the 225th American Astronomical Meeting, held in Seattle, WA, January 4-8, 2015.

Kristensen, Kira, Hayley E. Kanipe, Jonathon B. Hall, Ryan Spady, Kristina Bowie, Becky Wigen, Andrew Hickok, Beth Weathers, Shane D. Bond, Ian R. Wilson, Tracy Cardinal, and Simon P. Kaltenrieder. (2009). Archaeological Excavation Park Farm Site (DhRq 22) Proposed Shopping Centre Pitt Meadows, B.C. Conducted under Heritage Conservation Act Permit 2007-0126. Report on file with the British Columbia Archaeology Branch, Victoria.

La Salle, Marina J. (2008). Beyond Lip Service: An Analysis of Labrets and Their Social Context on the Pacific Northwest Coast of British Columbia. Unpublished Master’s thesis. Department of Anthropology, University of British Columbia, Vancouver.

Lamb, William Kaye (ed.). (1960). The Letters and Journals of Simon Fraser, 1806-1808. Macmillan, Toronto.

Lamb, William Kaye. (1984). A Voyage of Discovery to the North Pacific Ocean and Round the World, 1791-1795, Vol. II. The Hakluyt Society, London.

Larson, Lynn L., and Dennis E. Lewarch. (1995). The Archaeology of West Point, Seattle, Washington 4,000 years of Hunter-Fisher-Gatherer Land Use in Southern Puget Sound, Volume 1, Part 1. Report on file with the Washington Department of Archaeology and Historic Preservation, Olympia.

Leach, Neil. (2005). Belonging: Toward a Theory of Identification with Space. In Habitus: A Sense of Place, 2nd edition, edited by Jean Hillier and Emma Rooksby. Ashgate, Aldershot.

Lepofsky, Dana, Michael Blake, Doug Brown, Sue Morrison, Nicole Oakes, and Natasha Lyons. (2000a). The Archaeology of the Scowlitz site, Southwestern British Columbia. Journal of Field Archaeology. 27:391-416.

Lepofsky, Dana, Douglas Brown, Nicole Oakes, Sandra Morrison, Michael Blake, and Tracy Rogers. (2000b). Report on the 1999 Excavations at the Scowlitz Site (DhRl 16), Fraser Valley, B.C. Conducted under Heritage Conservation Act Permit 1999-112 and Stó:lō Permit 1999-16. Report on file with the British Columbia Archaeology Branch, Victoria.

148 Lepofsky, Dana, Ken Lertzman, Douglas Hallett, Rolf Mathewes. (2005). Climate Change and Culture Change on the Southern Coast of British Columbia 2400- 1200 Cal. B.P.: An Hypothesis. American Antiquity 70(2)267-293.

Lepofsky, Dana, David M. Schaepe, Anthony P. Graesch, Michael Lenert, Patricia Ormerod, Keith Thor Carlson, Jeanne E. Arnold, Michael Blake, Patrick Moore, and John Clague. (2009). Exploring Stó:lō-Coast Salish Interaction and Identity in Ancient Houses and Settlements in the Fraser Valley, British Columbia. American Antiquity 74:595-626.

Lugrin, N. De Bertrand. (1932). Indian Saga: Heroic Tales from the Golden Age of the Indian’s Supremacy on the West Coast. Maclean’s Magazine December 15, p. 38.

MacDonald, Richard. (2000). Body Heat as a Strategy for Winter Survival in Housepits. In The Ancient Past of Keatley Creek, Volume II: Socioeconomy, edited by Brian Hayden, pp. 239-252. Archaeology Press, Department of Archaeology, Simon Fraser University, Burnaby.

Madella, Marco, Gabriella Kovacs, Brigitta Kulcsarne-Berzsenyi, and Ivan Briz I Godino (eds.). (2013). The Archaeology of Household. Oxbow Books, Oxford.

Mack, C., and V. Baucom. (1992). Washington Archaeological Site Inventory Form – 45- SA-303. Form on file with the Washington Department of Archaeology and Historic Preservation, Olympia.

Mack, Cheryl A., James C. Chatters, and Anna M. Prentiss. (2010). Archaeological Data Recovery at the Beech Creek Site (45LE415), Gifford Pinchot National Forest, Lewis County, Washington. Report on file with the Washington Department of Archaeology and Historic Preservation, Olympia.

Mackie, Quentin. (2001). Settlement Archaeology in A Fjordland Archipelago: Network Analysis, Social Practice, and The Built Environment of Western Vancouver Island, British Columbia, Canada since 2,000 BP. Bar International Series 926. British Archaeology Reports, Oxford.

Maclachlan, Morag. (1998). The Fort Langley Journals, 1827-1830. UBC Press, Vancouver.

Martindale, Andrew, and Kisha Supernant. (2009). Quantifying the Defensiveness of Defended Sites on the Northwest Coast of North America. Journal of Anthropological Archaeology 28(2):191-204.

Maschner, Herbert G. (1997). Settlement and subsistence in the later prehistory of Tebenkof Bay, Kuiu Island, Southeast Alaska. Arctic Anthropology 34(2):74-99.

Mason, Andrew, and Tanja Hoffmann. (1998). Archaeological Mitigation of DjSf-13 Buckley Bay Main Road and Interchange Vancouver Island Highway Project Heritage Conservation Act Permit 1997-345. Report on file with the British Columbia Archaeology Branch, Victoria.

149 Mason, Andrew, Tanja Hoffmann, Jeff Bailey, David Pokotylo, and Erika Laanela. (1998). Archaeological Mitigation of DiSh-6 Oyster River Crossing, Oyster River, B.C. Vancouver Island Highway Project. Conducted under Heritage Conservation Act Permit 1997-227. Report on file with the British Columbia Archaeology Branch, Victoria.

Mathews, Darcy. (2003). Excavations at EaSf-36: Mitigative Data Recovery of Strata Lot B, Plan VIS3115, Cortes Island, BC. Conducted under Heritage Inspection Permit 2002-397. Report on file with the Archaeology Branch of British Columbia, Victoria.

Mathews, Darcy. (2014). Funerary Ritual, Ancestral Presence, and the Rocky Point Ways of Death. Unpublished Doctoral dissertation. Department of Anthropology, University of Victoria, Victoria.

Matson, R. G. (2003). The Coast Salish House: Lessons from Shingle Point, Valdes Island, British Columbia. In Emerging from the Mist: Studies in Northwest Coast Culture History, edited by R. G. Matson, Gary Coupland, and Quentin Mackie, pp. 76-104. UBC Press, Vancouver.

Matson, R. G., Deanna Ludowicz, and William Boyd. (1980). Excavations at Beach Grove in 1980. Conducted under Heritage Inspection Permit 1980-0014. Report on file with the Archaeology Branch of British Columbia, Victoria.

Matson, R.G., and Gary Coupland. (1995). The Prehistory of the Northwest Coast. Academic Press, San Diego, California.

Matson, R. G., Heather Pratt, and Lisa Rankin (with contributions by Susan Crockford, Jerome Cybulski, Brian Thom, and Rebecca Wigen. (1991). 1989 and 1990 Crescent Beach Excavations, Final Report. The Origins of the Northwest Coast Ethnographic Pattern: The Place of the Locarno Beach Phase. Conducted under Heritage Investigation Permits 1989-0027 and 1990-0067. Report on file with the Archaeology Branch of British Columbia, Victoria.

Matter, S. (2010). Clashing Claims: Neopatrimonial Governance, Land Tenure Transformation, and Violence at Enoosupukia, Kenya. Political and Legal Anthropology Review 33(1):67-88.

Mattson, John L. (1971). A Contribution to Skagit Prehistory. Unpublished Master’s thesis, Department of Anthropology, Washington State University, Bellingham.

Mauger, Jeffrey Adams. (1978). Shed Roof Houses at the Ozette Archaeological Site: A Protohistoric Architectural System. Unpublished Doctoral dissertation, Department of Anthropology, Washington State University, Bellingham.

McLay, Eric, Doug Brown, Nicole Oakes, and Lindsay Oliver. (2009). Archaeological Investigations of Ground Penetrating Radar Mapping at Somenos Creek (DeRw- 018), B.C. Conducted under Heritage Conservation Act Permit 2006-393. Report on file with the British Columbia Archaeology Branch, Victoria.

McMillan, Alan D., and Denis St. Claire. (1982). Alberni Prehistory: Archaeological and

150 Ethnographic Investigations on Western Vancouver Island. Theytus Books, Penticton.

Mearns, Robin. (1993). Territoriality and Land Tenure among Mongolian Pastoralists: Variation, Continuity and Change. Nomadic Peoples 33:73-103.

Menzies, A. (1923). Menzies journal of Vancouver’s voyage, April to October, 1792 (C. F. Newcombe, ed.). Archives of British Columbia Memoirs 5, Victoria.

Merchant, Peter, Monty Mitchell, Mike Rousseau, David Schaepe, and Jim Spafford. (1999). Preliminary Archaeological Investigation of DgRk-10: A Ts’élweyeqw Pithouse Village Settlement the Chilliwack River Valley. Conducted under Heritage Investigation Permit 1998-352. Report on file with the Archaeology Branch of British Columbia, Victoria.

Miller, Bruce G., and Daniel L. Boxberger. (1994). Creating Chiefdoms: The Puget Sound Case. Ethnohistory 41:267-293.

Mitchell, Donald H. (1971). Archaeology of the Gulf of Georgia Area, A Natural Region and its Culture Types. The British Columbia Provincial Museum, Syesis 4(1):1-228.

Mitchell, Donald H. (1974a). Salvage Archaeology 1974, DjSf 13 and DjSf 14, Site Report. Opportunities for Youth Project. Conducted under Permit 1974-18. Report on file with the British Columbia Archaeology Branch, Victoria.

Mitchell, Donald H. (1974b). Salvage Excavations at the Buckley Bay Site (DjSf 13), Baynes Sound, Vancouver Island, British Columbia. Conducted under Permit 1974-18. Report on file with the British Columbia Archaeology Branch, Victoria.

Mitchell, Donald H. (1990). Prehistory of the Coasts of Southern British Columbia and Northern Washington. In Northwest Coast, edited by Wayne Suttles pp. 340-358. Handbook of North American Indians, Vol. 7, Smithsonian Institution, Washington, D.C.

Mitchell, Donald, and David L. Pokotylo. (1996). Early Period Components at the Milliken Site. In Early Human Occupation in British Columbia, edited by Roy L. Carlson and Luke Dalla Bona, pp. 65-82. UBC Press, University of British Columbia, Vancouver.

Morgan, Vera A., Susan Axton, Suzanne Bell, Kathy Calaway, Fred Crisson, Judy Donald, Stan Gough, Glenn D. Hartmann, Craig Holstine, Ryan Ives, Sharon Larsen, R. Lee Lyman, Michele Parvey, Rebecca Stevens, and Sara Walker. (1999). The SR-101 Sequim Bypass Archaeological Project: Mid- to Late-Holocene Occupations on the Northern Olympic Peninsula, Clallam County, Washington, Volume I. Eastern Washington University Reports in Archaeology and History 100-108 Archaeological and Historical Services, Cheney.

Morin, Jesse. (2012). The Political Economy of Stone Celt Exchange in Pre-Contact British Columbia: The Salish Nephrite/Jade Industry. Unpublished Doctoral dissertation, The Faculty of Graduate Studies (Anthropology), University of

151 British Columbia, Vancouver.

Moss, Madonna L., and Jon M. Erlandson. (1992). Forts, Refuge Rocks, and Defensive Sites: The Antiquity of Warfare along the North Pacific Coast of North America. Arctic Anthropology 29(2):73–90.

Netting, Robert McC., Richard R. Wilke, and Eric J. Arnould. (1984). Households: Comparative and Historical Studies of the Domestic Group. University of California Press, Berkeley.

Nokes, Randolph David. (2004). Mid-Holocene Terrestrial Animal Use in the Gulf of Georgia Region: A Case Study from the Ferndale Site, Lower Nooksack River, Washington. Unpublished Master’s thesis, Department of Anthropology, Western Washington University, Bellingham.

Nakonechny, Lyle Daniel. (2015). Archaeological Survey and Testing in the Willapa River Valley of Southwest Washington. Unpublished Doctoral Dissertation, Department of Anthropology, Washington State University, Pullman.

Northwest Research Obsidian Studies Laboratory. (2015). Results of X-Ray Fluorescence Analysis (DhRl-16). Report on file with the Laboratory of Archaeology, University of British Columbia, Vancouver.

Northwest Research Obsidian Studies Laboratory. (2016). Results of X-Ray Fluorescence Analysis of Obsidian Artifacts from Lasqueti Island and Hare Point, British Columbia, Canada. Report on file with the authors.

Northwest Research Obsidian Studies Laboratory. (2017). Results of X-Ray Fluorescence Analysis of Obsidian Artifacts from DhRq-21, British Columbia, Canada. Report on file with Brown & Oakes Archaeology, New Westminster.

NASA. (2013). Earth fact sheet. Available at: http://nssdc.gsfc.nasa.gov/planetary/factsheet/earthfact.html. Accessed on: March 19, 2015.

Odell, George H. (2003). Lithic Analysis. Manuals in Archaeological Methods, Theory, and Technique. Springer Science and Business Media, New York.

Oka, Rahul Chandrashekhar, and Chap Kusimba. (2008). The Archaeology of Trading Systems, Part 1: Towards a New Trade Synthesis. Journal of Archaeological Research 16:339-395.

Oliva, Aude, and Antonio Torralba. (2007). The Role of Context in Object Recognition. Trends in Cognitive Sciences 11(12):520-527.

Palmer, Stephen E. 1975. The Effects of Contextual Scenes on the Identification of Objects. Memory & Cognition 3(5):519-526.

Percy, Richard Charles William. (1974). The Prehistoric Cultural Sequence at Crescent Beach, British Columbia. Unpublished Master’s thesis, Department of Archaeology, Simon Fraser University, Burnaby.

152

Polana, Ramprasad, and Randal Nelson. 1994. Low Level Recognition of Human Motion (Or How to Get Your Man Without Finding His Body Parts). 77 - 82. 10.1109/MNRAO.1994.346251.

Pratt, Heather, D. Geordie Howe, and Dave Hall. (1998). West Bay Archaeological Mitigation, Gambier Island, B.C. Conducted under Heritage Conservation Act Permit 1997-216 and Squamish First Nation Permit 1997-0018. Report on file with the British Columbia Archaeology Branch, Victoria.

Rapoport, Amos. (1969). House Form and Culture. Prentice Hall, Englewood Cliffs, New Jersey.

Rapoport, Amos. (1986). Culture and Built Form – A Reconsideration. In Architecture in Culture Change: Essays in Built Form & Culture Research, edited By David G. Saile, pp. 157-175. Built Form & Culture Studies, School of Architecture and Urban Design. The University of Kansas, Lawrence.

Reed, Patrick, Thomas Snowden, and Sarah Campbell. (2010). State of Washington Archaeological Site Inventory Form – 45WH47. Form on file with the Washington Department of Archaeology and Historic Preservation, Olympia.

Reimer, Rudy (Yumks). (2006). The 2006 Squamish Nation Trust Archaeological Research Project Final Report. Conducted under Heritage Conservation Act Permit 2006-278. Report on file with the British Columbia Archaeology Branch, Victoria.

Reimer, Rudy (Yumks). (2012a). The Mountains and Rocks are forever: Lithics and Landscapes of Skwxwú7mesh Uxwumixw. Unpublished Doctoral dissertation, School of Graduate Studies (Anthropology), McMaster University, Hamilton.

Reimer, Rudy (Yumks). (2012b). X-Ray Fluorescent Report on Materials Recovered from DhRl-29 in the lower Fraser Valley. Report in possession of author.

Reimer, Rudy (Yumks). (2014). The Nch’kay (Mount Garibaldi) Obsidian Source and its Local to Regional Archaeological Distribution. International Association for Obsidian Studies Bulletin 50:10-17.

Renfrew, Colin. (1977). Alternative Models for Exchange and Spatial Distribution. In Exchange Systems in Prehistory, edited by Timothy K. Earle and Jonathon Ericson, pp. 71-90. Academic Press, New York.

Ritchie, Morgan. (2010). From Watershed to House: The Cultural Landscapes of the Sts’ailes People. Unpublished Master’s thesis. Department of Archaeology, Simon Fraser University, Burnaby.

Ritchie, Morgan, Dana Lepofsky, Sue Formosa, Marko Porcic, and Kevan Edinborough. (2016). Beyond Culture History: Coast Settlement Patterning and Demography in the Fraser Valley, BC. Journal of Anthropological Archaeology 43:140-154.

Ritchie, Morgan, and Ian Sellers. (2016). Eagle Mountain Pipeline Project

153 Archaeological Impact Assessment, Final Report. Report on file with Inlailawatash, North Vancouver.

Rooke, Lara C., Jason B. Cooper, and James C. Chatters. (2011). Centralia-Chehalis Flood Reduction Project, Volume I: Intensive Archaeological Survey of the Airport, Salzer Creek, and Skookumchuck Levees, Lewis County, Washington. Report on file with the Washington Department of Archaeology and Historic Preservation, Olympia.

Rorabaugh, Adam N., and Caitlyn Y. McNabb. (2014). A Geospatial Analysis of Toolstone Acquisition and Use: A Preliminary Investigation of Material Quality and Access Over 4,000 Years in the Salish Sea. Canadian Journal of Archaeology 38:371-393.

Sakaguchi, Takashi, Jesse Morin, and Ryan Dickie. (2010). Defensibility of Large Prehistoric Sites in the mid-Fraser Region on the Canadian Plateau. Journal of Archaeological Science 37:1171-1185.

Samuels, Stephan R. (2006). Ozette Household Production. In Household Archaeology on the Northwest Coast, edited by Elizabeth A. Sobel, D Ann Trieu Gahr, and Kenneth M. Ames, pp. 200-232. International Monographs in Prehistory, Ann Arbor.

Schaepe, David, Albert (Sonny) McHalsie, Keith Thor Carlson, Patricia Ormerod. (2001). Changing Households, Changing Houses. In A Stó:lō-Coast Salish Historical Atlas, edited by Keith Thor Carlson, pp. 40-47. Douglas & McIntyre, Vancouver.

Schaepe, David M. (2006). Rock Fortifications: Archaeological Insights into Precontact Warfare and Sociopolitical Organization among the Stó:lō of the Lower Fraser River Canyon, BC. American Antiquity 71(4):671-705.

Sepúlveda, Marcela, Thibault Saintenoy, Luis Carnejo, Carole Dudognon, Francisco Espinoza, Zaray Guerrero-Bueno, and Enrique Cerillo-Cuenca. Rock Art Painting and Territoriality in the Precordillera of Northernmost Chile (South-Central Andes). (2017). Archaeological and Spatial Approaches to the Naturalistic Tradition. Quaternary International, in press.

Shackley, Steven M. (1998). Geochemical Differentiation and Prehistoric Procurement of Obsidian in the Mount Taylor Volcanic Field, Northwest New Mexico. Journal of Archaeological Science 25:1073-1082.

Smith, Dan J., and Joseph R. Desloges. (2000). Little Ice Age History of Tzeetsaytsul Glacier, Tweedsmuir Provincial Park, British Columbia. Géographie Physique et Quaternaire 54(2):135-141.

Sobel, Elizabeth A. (2006). Household Prestige and Exchange in Northwest Coast Societies: A Case Study from the Lower Columbia River Valley. In Household Archaeology on the Northwest Coast, edited by Elizabeth A. Sobel, D Ann Trieu Gahr, and Kenneth M. Ames, pp. 159-199. International Monographs in Prehistory, Ann Arbor.

154 Sobel, Elizabeth A. (2012). An Archaeological Test of the “Exchange Expansion Model” of Contact Era Change on the Northwest Coast. Journal of Anthropological Archaeology 31:1-21.

Sobel, Elizabeth A., D Ann Trieu Gahr, and Kenneth M. Ames (eds.). (2006). Household Archaeology on the Northwest Coast. Archaeological Series 16. International Monographs in Prehistory, Ann Arbor.

Springer, Chris, Megan Caldwell, Nyra Chalmer, Julia Jackley, and Dana Lepofsky (with contributions from Craig Rust, John Welch, Shannon Wood, and Kasia Zimmerman). (2013). Final Report on Archaeological Investigations (2009-0132) by the Simon Fraser University-Tla’amin First Nation Archaeology and Heritage Stewardship Program, Powell River Regional District, British Columbia. Conducted under Heritage Inspection Permit 2009-0132. Report on file with the Archaeology Branch of British Columbia, Victoria.

Springer, Chris, and Dana Lepofsky. (2011). Pithouses and People: Social Identity and Pithouses in the Harrison River Valley of Southwestern British Columbia. Canadian Journal of Archaeology 35:18-54.

Springer, Chris, Dana Lepofsky, and Michael Blake. (2018). Obsidian in the Salish Sea: An archaeological examination of ancestral Coast Salish social networks in SW British Columbia and NW Washington state. Journal of Anthropological Archaeology 51:45-66.

Spurgeon, Terry. (1994). The Park Farm Site: DhRq 22. Unpublished Bachelor of Art’s Honours thesis, Department of Archaeology, Simon Fraser University, Burnaby.

Starks, Rachel Rose, Jen McCormack, and Stephen Cornell. (2011). Native Nations and U.S. Borders: Challenges to Indigenous Culture, Citizenship, and Security. The Udall Center for Studies in Public Policy, The University of Arizona, Tucson.

Stanley, Golden. (1954). Notes from Noel George, Sliammon. Typescript on file at the Powell River Museum, Powell River.

Stein, Julie K., Jennie N. Deo, and Laura S. Phillips. (2003). Big Sites – Short Time: Accumulation Rates in Archaeological Sites. Journal of Archaeological Science 30:297-316.

Stern, Theodore. (1998). Columbia River Trade Network. In Handbook of North American Indians, Vol. 12: Plateau, edited by Deward E. Walker, Jr., pp. 641- 652. Smithsonian Institution, Washington, D.C.

Stryd, Arnoud H., Geordie Howe, A. Joanne Curtin, Randy Bouchard, Dorothy Kennedy, Karla Kusmer, Dana Lepofsky, Karen Preckel, June Ryder, Susan Crockford, Rebecca Wigen, Brian Chisolm, Richard Brolly, and Beth Bedard. (1991). Archaeological Investigations at Tsawwassen, B.C. Conducted under Heritage Investigation Permits 1989-0041 and 1990-0002. Report on file with the Archaeology Branch of British Columbia, Victoria.

155 Stuiver, Minze, Paula Reimer, and Ron Reimer. (2017). CALIB 7.1 Radiocarbon Calibration. Available at: www.calib.qub.ac.uk/calib/index.html

Supernant, Kisha. (2014). Intervisibility and Intravisibility of Rock Feature Sites: A Method for Testing Viewshed Within and Outside the Socio-spatial System of the Lower Fraser River Canyon, British Columbia. Journal of Archaeological Science 50:497-551.

Suttles, Wayne P. (1951). Economic Life of the Coast Salish of Haro and Rosario Straits. Unpublished Doctoral dissertation, Department of Anthropology, University of Washington, Seattle.

Suttles, Wayne P. (1960). Affinal Ties, Subsistence, and Prestige among the Coast Salish. American Anthropologist 62:296-305.

Suttles, Wayne P. (1963). The Persistence of Inter-Village Ties among the Coast Salish. Ethnology 2:512-525.

Suttles, Wayne P. (1987). Coast Salish Essays. Talonbooks, Vancouver.

Suttles, Wayne P. (ed.). (1990). Northwest Coast. Handbook of North American Indians Vol. 7. Smithsonian Institution, Washington, D.C.

Suttles, Wayne P. (1991). The Shed Roof House. In A Time of Gathering: Native Heritage in Washington State, edited by Robin K. Wright, pp. 212-222. University of Washington Press, Seattle.

Suttles, Wayne P. (2000). The Ethnographic Significance of the Fort Langley Journals. 2nd ed. In The Fort Langley Journals, edited by Morag Maclachlan, pp. 163-210. UBC Press, University of British Columbia, Vancouver.

Taçon, Paul. (2008). Marks of Possession: The Archaeology of Territory and Cross- Cultural Encounter in Australia and South Africa. In Handbook of Landscape Archaeology, edited by Bruno David and Julian Thomas, pp. 218-225. Left Coast Press, Walnut Creek.

Taylor, H. C. and W. Duff. (1956). A Post-Contact Southward Movement of the Kwakiutl. Research Studies of the State College of Washington 24(1):56–66.

Teit, James. (1900). The Thompson Indians of British Columbia. American Museum of Natural History Memoirs Vol. 1 Part 2. AMS Press, New York.

Teit, James. (1930). The Salishan Tribes of the Western Plateaus. Extract from the Forty-Fifth Annual Report of the Bureau of American Ethnology, edited by Franz Boas. United States Government Printing Office, Washington, D.C.

Tla’amin Place Names. (2002). Tla’amin Place Names. On file with the Tla’amin First Nation, Powell River.

Tla’amin Traditional Use Study. (1996). Tla’amin Treaty Society Traditional Use Study. On file with the Tla’amin First Nation, Powell River.

156

Toft, Monica Duffy. (2003). The Geography of Ethnic Violence: Identity, Interests, and the Indivisibility of Territory. Princeton University Press, New Jersey.

Tolfree, J. (1989). Washington Archaeological Site Inventory Form – 45-SA-117. Form on file with the Washington Department of Archaeology and Historic Preservation, Olympia.

Tollefson, K. (1996). In Defense of A Snoqualmie Political Chiefdom Model. Ethnohistory 43:145-171.

Toniello, Ginevra, Dana Lepofsky, Nicole Smith, Christina Neudorf, and Travis Crowell. (2016). Quadra Island Clam Garden Project, 2014-2015 in Kanish and Waiatt Bays, Quadra Island, BC. Conducted under Heritage Conservation Act Permit 2014- 0095. Report on file with the British Columbia Archaeology Branch, Victoria.

Trosper, R. L. (2002). Northwest Coast Indigenous Institutions that Supported Resilience and Sustainability. Ecological Economics 41: 329-344.

Turner, Nancy J., and James T. Jones. (2000). Occupying the Land: Traditional Patterns of Land and Resource Ownership among First Peoples of British Columbia. Paper presented at the International Association for the Study of Common Property Conference.

Turner, Nancy J., Robin Smith, and James T. Jones. (2005). “A Fine Line between Two Nations”: Ownership Patterns for Plant Resources among Northwest Coast Indigenous Peoples. In Keeping it Living: Traditions of Plant Use and Cultivation on the Northwest Coast of North America, edited by Douglas Deur and Nancy J. Turner, pp. 151-178. University of Washington Press, Seattle.

Tykot, Robert H. (2002). Chemical Fingerprinting and Source Tracing of Obsidian: The Central Mediterranean Trade in Black Gold. Accounts of Chemical Research 35(8):618-627.

Urban, Patricia, and Edward Schortman. (2013). Monumentality, Territoriality, and Networks during the Middle Preclassic in Northwest Honduras. Archaeological Papers of the American Anthropological Association 22:87-106.

Vaughan, Kevin Allan. (2010). A Comparison of Lithic Technology and Function across Environmental Zones in the Southern Washington Cascades. Unpublished Master’s thesis, Department of Cultural and Environmental Resource Management, Central Washington University, Ellensburg.

Von Krogh, Georg Henning. (1980). Archaeological Investigations at the Flood and Pipeline Sites, Near Hope, British Columbia. Occasional Papers 4, Heritage Conservation Branch, Victoria.

Walker, Sara L., Ryan S. Ives, Charles T. Luttrell, R. Lee Lyman, Rebecca J. Wigen, Julie K. Stein, Phoebe Anderson, Michael A. O’Neal, and Richard E. Hughes. (2003). Archaeological Investigations at Sites 45SJ165 and 45SJ169, Decatur Island, San Juan County, Washington. Eastern Washington University Reports in

157 Archaeology and History 100118 Archaeological and Historical Services, Cheney.

Washington, Michelle (Siemthlut). (2004). Bringing Traditional Teachings to Leadership. American Indian Quarterly 28:583-603.

Wessen, Gary C. (2000). Archaeological Testing and Evaluation Studies Associated with the Federal Energy Regulatory Commission Relicensing of the Cowlitz River Hydroelectric Project (FERC No. 2016), Lewis County, Washington. Report on file with the Washington Department of Archaeology and Historic Preservation, Olympia.

Widmer, Randolph J. (2004). Explaining Sociopolitical Complexity in the Foraging Adaptations of the Southeastern United States: The Roles of Demography, Kinship, and Ecology in Sociocultural Evolution. In Signs of Power: The Rise of Cultural Complexity in the Southeast, edited by Jon L. Gibson and Philip J. Carr, pp. 234-253. The University of Alabama Press, Tuscaloosa.

Williams, Louise Eileen. (2013). Revisiting the Locarno Beach Site (DhRt-6), Vancouver, BC. Unpublished Master’s thesis, Department of Archaeology, Simon Fraser University, Burnaby.

Wilmeth, Roscoe. (1978). Anahim Lake Archaeology and the Early Historic Chilcotin Indians. National Museum of Man Mercury Series Archaeological Survey of Canada Paper No. 82, National Museum of Canada, Ottawa.

Witt, Sandra, Rudy Reimer, Robert Muir, and Cameron M. Smith. (2001). Archaeological Investigations at EcRq 1 on the Birkenhead River, Mount Currie, BC. Conducted under Heritage Conservation Act Permit 2000-242. Report on file with the British Columbia Archaeology Branch, Victoria.

Zedeño, María Nieves. (1997). Landscapes, Land Use, and the History of Territory Formation: An Example from the Puebloan Southwest. Journal of Archaeological Method and Theory 4(1):67-103.

Zedeño, María Nieves. (2008). The Archaeology of Territory and Territoriality. In Handbook of Landscape Archaeology, edited by Bruno David and Julian Thomas, pp. 210-217. Left Coast Press, Walnut Creek.

Zedeño, María Nieves and Brenda J. Bowser. (2009). The Archaeology of Meaningful Places. In The Archaeology of Meaningful Places, edited by Brenda J. Bowser and Maria Nieves Zedeño, pp. 1-14. The University of Utah Press, Salt Lake City.

158 Appendix A.

Tla’amin Place Names

159 Table A1: Tla’amin place names mentioned in the text Location Orthography Phonetic Meaning Powell River settlement tiskʷat Teeskwat big river Paukeanum (Tla’amin reserve on p̓ aq̓ iʔaǰɩm Pah Qee ahjim Maple tree place Cortes Island) Gorge Harbour (Cortes Island) saʔyiɬit Sah yeetl salt water lake Blubber Bay (Texada Island) t̓ atlaχʷnač TatlaXw nach water swirls around Mary Point gi: t̓ aχʷ Xah Xah jemin to quarrel Pocahontas Bay (Texada Island) šɛtɛqʷən Sheht ay kwahn steep or high Vananda Cove (Texada Island) lɛχʷamɛn Leh Xwa men unknown Grief Point χakʷum Xah Kwoom having Indian Rhubarb Myrtle Rock/Point kʷʊθaysqɛn Kwoo'thays qen island at the head or island in a bay Smelt Bay (Cortes Island) kʊmaχən Koom ahXun protection for bay Marina Island šɛtqaǰɛ Shet qah jeh to tie a rope around a tree Keefer Bay (Savary Island) qɛyɛ qʷən Qaye qwun freshwater spring Coode Island (Okeover Inlet) ƛaqəmayɩn Klah quh my yin pointing down the inlet Entrance to Gorge Harbour (Cortes yipikʷʊ Yip'i'kwu Break ice Island) Tla’amin Reserve t̓ išosəm t'ishosum milky waters from herring spawn Stillwater Bay qʷoqʷnəs Qwoqwness little whale Portage Cove qɛgiyɩn Qeh ghee yin short cross over Emmonds Beach šɛʔaystən Sheh ays tun steep bluff Settlement at head of Theodosia toqʷanan Toqwanun unknown Harwood Island ʔaʔgayqsən Ah gyk sun pointed nose Stag Bay (Hernando Island) ɬay ta ƛač Tly ta Klach towards the shore Cochrane Bay (Malaspina Inlet) qɛqɛgɩš Qe'Qe'Gish a place to walk over Grace Harbour (Malaspina Inlet) q̓ a qɛy q̓ ay Qah Qeh qay camp overnight Hernando Island kʷʊp ƛač Koop Klach hump on stomach Scuttle Bay ƛɛkʷanəm Kleh Kwahn num water coming in fast Blind Creek (Cortes Island) jɩmoθɛn Jim o then hard to see, hard to get at Indian Point/Head (Savary Island) θatɛq Thah teq broken off Texada Island sayayɩn Sah yah yin end of the island Bliss Landing ǰɛǰɩšʔsiʔəm Jeh Jish shee um packing on your back

160 Location Orthography Phonetic Meaning Hinder Creek čɛn Chen dog Settlement at head of Okeover toχʷnač Toxw nach strung out rear end Klahanie čuχʷo θɛn Choo Xwo then between any two points Tenedos Bay kʷʊmqɛn Kwoom qen facing towards the head Galley Bay kʷišitəm Kwish ee tum naming Prideaux Haven mačɛnay Mah Chen I louse Lund ƛaʔamɛn Klah ah men a place to head towards, a refuge Manson Bay (Cortes Island) ɬaytoθɛn Tlay toh then the mouth of the bay is facing inwards Keays Bay χʷoχʷ ǰusɛm XwoXw joosem covering head with hands Laura Cove (Prideaux Haven) χɛpǰɛʔqɛn Xahp jeh qen to go back Melanie Cove (Prideaux Haven) muθkʷamɛn Muth kwa me unknown Willingdon Beach ʔahʔǰumɩχʷ Ah joo miexw clear ground Campsite/lookout near Indian t̓ i: t̓ i: may T'eet'ee may many wild cherry trees Point/Head (Savary Island) Gibsons Beach qʷɛqʷiqʷɛy Qweh qwee qway little sandy beach Dinner Rock qʷaqʷtɛm qwaqw tehm any little stream McRae Cove Islets qʷoqʷnəsəm KwuKwnessum unknown Homfray Channel θiyčɛmayič Thee chum mi further back inside yich Waddington Channel sayqɛn Sayqen area inside the mouth Toba Inlet yɛkʷamɛn Yekwamen unknown Jervis Inlet yɛkʷamɛn Yekwamen unknown Cortes Bay (Cortes Island) kʷakʷamaws Kwah kwah maws red rocky area Bay past Cortes Bay (Cortes Island) χaχajɩmɩn XaXa jimin to quarrel Copeland Islands kʷʊkʷakʷθays Kwoo Kwahk thys a bunch of islands Scott Point (Malaspina Inlet) t̓ it̓ agayitᶿa Teeta GuyeeTHa picking lice from someone else’s hair Settlement on Harwood Island maloʔhom Malo hom waters bubbling up from the ground Bunster Range ʔɛyičɩn Aye yih chin top of mountain Forbes Bay ʔap̓ ʊkʷəm Ah pooh kwom having maggots

161 Location Orthography Phonetic Meaning Brem Bay (Toba Inlet) qʷeqʷti:čɩnəm Qweqw'tee getting humpback salmon chinum Klahoose settlement ~24 km up Toba nat θuwom nat thoowom unknown River Lasqueti Island xʷɛʔɛtay Xweh et tay Yew tree

162 Appendix B

Statistical Comparisons of House Forms

163 Table B1: Descriptive Statistics and Mann-Whitney U Results. Descriptive Statistics Mann-Whitney U Results Test # Groups Mean Sum of n Mean Median s Min. Max. MWU p rank ranks Pre-2300 cal. 13 35.6 36.0 20.0 12.2 80.0 8.3 108.0 1) All Coast houses BP 17.0 0.001 livable space (m2) Post-2300 cal. 14 185.9 123.5 146.8 26.0 500.0 19.3 270.0 BP Pre-2300 cal. 2) All Lower Fraser 12 83.5 76.1 59.9 15.7 200.0 21.4 257.0 BP houses livable space 179.0 0.458 Post-2300 cal. (m2) 35 102.5 76.3 75.9 14.0 407.3 24.9 871.0 BP 3) Post-2300 above- Coast 12 208.6 128.0 146.5 26.0 500.0 10.7 129.0 ground house livable 45.0 0.836 Lower Fraser 8 181.3 181.0 113.7 14.0 407.3 10.1 81.0 space (m2) 4) In-ground house Coast 12 36.7 34.0 21.4 12.2 80.0 14.4 172.5 94.5 0.001 livable space (m2) Lower Fraser 39 80.5 73.4 46.0 15.7 200.0 29.6 1153.5 5) Lower Fraser in- Pre-2300 12 83.5 76.1 59.9 15.7 200.0 19.7 237.0 ground house livable 159.0 0.934 Post-2300 27 79.1 68.6 39.6 16.0 178.8 20.1 543.0 space (m2)

164 Appendix C

Results of Obsidian Elemental Analyses by Northwest Research Obsidian Studies Laboratory (NWROSL) and the Department of Archaeology pXRF Laboratory at SFU

165 Results of Obsidian Elemental Analyses by Northwest Research Obsidian Studies Laboratory (NWROSL) in Corvallis, Oregon, and the Department of Archaeology pXRF Laboratory at SFU

A total of 71 obsidian artifacts were run by NWROSL (n=44) and SFU (n=27). Twelve artifacts came from unknown sources, three artifacts were analyzed by both labs, the lab report for one artifact analyzed by NWROSL has gone missing, and one artifact run by SFU was mis- assigned during the analysis presented in this dissertation. Of the twelve unknowns, one (NWROSL 2009b, EaSe-19 sample 2) was added to the Kingcome sample from site EaSe-76 (Table 4, ID 39). It was correlated to Unknown Type A and is likely associated with the Mount Silverthrone volcanic event that produced the Kingcome source. (It should be noted that site EaSe-19 was amalgamated with EaSe-76 by the B.C. Archaeology Branch in 2009). The three artifacts that were analyzed by both labs include one from Lasqueti Island and two from site EaSe-18. The Lasqueti Island artifact was collected from a private garden near Tucker Bay (Table 4, ID 64) and was assigned to Glass Buttes by SFU and Whitewater Ridge by NWROSL. The NWROSL result was used for this study. Two artifacts collected from site EaSe-18 (Table 4, ID 42) were both sourced to Anahim by both labs. The single obsidian artifact recovered from site DlSd-17 (Table 4, ID 45) was sourced to Glass Buttes 1 by NWROSL. However, both the hard and digital copies of the results have gone missing. The artifact that was mis-assigned was correlated to CCUB by the SFU lab (SFU 2017). For the analysis presented in this dissertation, the artifact was mistakenly assigned to the Kingcome source. This does not affect the results as CCUB and Kingcome are likely both associated with Mount Silverthrone, north of the Salish Sea region. In summary, 57 of the 71 artifacts analyzed by NWROSL and SFU were used in this study (see Table 4, IDs 39-42, 44-46, 49-50, 63-64, and 149-150). The reports are organized below by year beginning with the most recent.

166 Simon Fraser University Department of Archaeology X-Ray Fluorescence Report 2017-002

Dr. Rudy Reimer Simon Fraser University Department of Archaeology

Abstract: Eleven obsidian artifacts submitted for energy dispersive X-Ray fluorescence (XRF) analysis attribute to three known sources. The samples were prepared and analyzed at the Department of Archaeology, Simon Fraser University, Burnaby BC. The obsidian artifacts match the elemental composition of flows at Anahim Peak, Unknown Central Coast B of British Columbia and Obsidian Cliff in Oregon.

Analytical Methods

X-Ray Fluorescence (XRF) is a nondestructive technique that is well suited to the trace element analysis of almost any material. This analysis used a Bruker Tracer III-V+ portable XRF spectrometer. The system is equipped with a Peltier cooled Ag-free SiPIN, resolution ~175eV @ 5.9 KeV in an area of 12 mm. The tube’s power supply is driven by a 40 kV 1mA with a range of 4 to 40 kV. For various types of materials, occurring in the archaeological record the Bruker Tracer III-V+ can adjust its power settings, use of filters to narrow the x-ray beam to focus analysis on specific elements and or use of a vacuum system to allow for the detection of light elements of the periodic table.

This analysis used a filter made of 0.003’’ Cu, 0.001’’ Ti and 0.012 Al with instrument power settings at 40 kV and 15 micro amps with no vacuum system (i.e. not analyzing light elements). These settings allow X-rays from 17-40 keV to reach the sample, thus efficiently exciting elements from Fe to Mo. These elements are the most useful to identify the origins of igneous rock particularly obsidian (Ferguson 2012; Glascock and Ferguson 2012; Reimer 2015; Speakman 2012). All samples ran for a total of 200 seconds and on flat surfaces to ensure accurate and precise calculation of elemental peak data.

Results

Diagnostic trace element values most common to characterize obsidian are compared directly to those directly known or established through published literature. This provides the means for accurate source designation. Included in this analysis are the known obsidian in the Simon Fraser University, Department of Archaeology reference collection and other known obsidian source values reported in the literature. When needed, unpublished elemental data collected through the analysis of other labs is used to confirm SFU lab results (Godfrey-Smith 1986; James et al. 1996; Moss and Erlandson 1992; Reimer 2012, 2015, n.d.; Reimer and Hamilton 2015).

The results of analysis are presented in Figure 1 and Table 1 with all values in parts per million (ppm). Results derived by the utilization of Bruker’s S1CalProces that includes the calibration of results to over 45 known obsidian/rhyolite sources, tested and known through a variety of analysis (Ferguson 2012; Glascock and Ferguson 2012; Reimer

167 2012, 2015; n.d.; Reimer and Hamilton 2015; Speakman 2012). As such, calculated values for artifacts submitted for analysis correlate to three distinct obsidian sources with trace elemental values falling within two standard deviations of analytical uncertainty (95%), marking the known upper and lower limits of elemental variability of known sources and flows within sources. The samples originate from three known sources Anahim Peak, Unknown Central Coast B in British Columbia and Obsidian Cliffs in Oregon USA (Figure 2).

Figure 1. Results of XRF analysis biplot of Zr and Zr to match results to source. Submitted samples are red dots. All values in parts per million (ppm). Note, samples that fall outside the ellipses are assigned to source by review of spectra for samples and sources.

168 Table 1. Results of XRF analysis. Source Artifact Mn Fe Zn Ga Th Rb Sr Y Zr Nb

Anahim EaSe18 bg 17 1 338 7747 259 25 36 375 11 103 136 109

Anahim EaSe18 bg15 1 460 10371 344 9 54 493 19 137 163 129

Anahim EaSe18 bg15 2 354 7903 247 18 47 423 16 126 162 132

Anahim EaSe18 bg15 3 542 10670 376 12 53 492 22 131 173 128

Anahim EaSe18 bg15 4 505 10682 339 8 60 512 21 137 186 141

Anahim EaSe18 bg15 5 743 11630 378 8 62 511 20 137 172 135

Anahim EaSe18 bg15 6 636 12033 424 13 61 518 14 141 185 140

Anahim EaSe18 bg15 7 236 5858 169 20 37 343 13 101 135 122

Obsidian Cliffs EaSe18 bg16 1 323 5982 23 15 1 59 160 12 102 10

Unknown Central Coast B EaSe18 bg24 1 487 17626 107 18 9 139 15 45 378 52

Anahim EaSe18 wf2 521 8242 272 8 40 385 18 103 134 115

169

Figure 2. Known sources in the SFU Archaeology obsidian source catalogue.

170 References

Carlson, Roy, L. 1994 Trade and Exchange in Prehistoric British Columbia, in Prehistoric Exchange Systems in North America (timothy G. Baugh and Jonathan E. Ericson eds.). Springer Press, 307- 365.

Ferguson, Jeffery, R. 2012 X-Ray Fluorescence of obsidian: approaches to the calibration and the analysis of small samples. In Studies in Archaeological Sciences: Handheld XRF for Art and Archaeology, ed. By Aaron N. Shugar, and Jennifer L. Mass. Leuven University Press.

Glascock, Michael, D. and Jeffrey, R. Freguson 2012 Report Analysis of Obsidian Source Samples by Multiple Analytical Methods. Report by the Archaeometry Laboratory, University of Missouri Research Reactor. Columbia Missouri.

James, Malcom, A. Jeff Bailey and John D’Auria 1996 A Volcanic Glass Library for the Pacific Northwest: Problems and Prospects. Canadian Journal of Archaeology, 20: 93-123.

Reimer, Rudy 2012 The Mountains and Rocks are Forever: Lithics and Landscapes of Skwxwú7mesh Uxwumixw Territory. Unpublished PhD dissertation, Department of Anthropology, McMaster University, Hamilton, Ontario, Canada.

2015 Reassessing the role of Mount Edziza obsidian in northwestern North America. Journal of Archaeological Science: Reports, 2: 418-426. n.d. Obsidian and Basaltic Rock Characterization for sources in the Pacific Northwest. Report on file in the Department of Archaeology, Simon Fraser University, Burnaby, B.C.

Reimer, Rudy, and Tyrone Hamilton 2015 Implications Between Technological Organization and Portable X-Ray Fluorescence Analysis on Lithic Material Use at Two Rockshelter Sites on the Southern Northwest Coast. In Tool Stone Geography of the Pacific Northwest, Archaeology Press, Simon Fraser University, Burnaby BC.

Speakman, Robert, J. 2012 Evaluation of Bruker’s Tracer Family Factory Obsidian Calibration for Handheld Portable XRF Studies of Obsidian. Centre for Applied Isotope Studies, University of Georgia, Athens Georgia.

171 Northwest Research Obsidian Studies Laboratory Report 2016-67

X-Ray Fluorescence Analysis of Obsidian Artifacts from Lasqueti Island and Hare Point, British Columbia, Canada

Alex J. Nyers Northwest Research Obsidian Studies Laboratory

Two obsidian artifacts from Lasqueti Island and the Hare Point archaeological site (EaSe-71), British Columbia, Canada, were submitted for energy dispersive X-ray fluorescence trace element provenance analysis. The samples were prepared and analyzed at the Northwest Research Obsidian Studies Laboratory under the accession number 2016-67.

Analytical Methods

X-Ray Fluorescence Analysis. Nondestructive trace element analysis of the samples was completed using a Thermo NORAN QuanX-EC energy dispersive X-ray fluorescence (EDXRF) spectrometer. The analyzer uses an X-ray tube excitation source and a solid-state detector to provide spectroscopic analysis of elements ranging from sodium to uranium (atomic numbers 11 to 92) and in concentrations ranging from a few parts per million to 100 percent. The system is equipped with a Peltier-cooled Si(Li) detector and an air-cooled X-ray tube with a rhodium target and a 76 micron Be window. The tube is driven by a 50 kV 2mA high voltage power supply, providing a voltage range of 4 to 50 kV. During operation, the tube current is automatically adjusted to an optimal 50% dead time, a variable that is significantly influenced by the varying physical sizes of the different analyzed samples. Small specimens are mounted in 32 mm-diameter sample cups with mylar windows on a 20-position sample tray while larger samples are fastened directly to the surface of the tray.

For the elements that are reported in Table A-1, we analyzed the collection with a 3.5 mm as well as an 8.8mm beam collimator installed with tube voltage and count times adjusted for optimum results. Instrument control and data analysis are performed using WinTrace software (version 7) running under the Windows 7 operating system.

The diagnostic trace element values used to characterize the samples are compared directly to those for known obsidian and fine-grained volcanic (FGV) sources reported in the literature and with unpublished trace element data collected through analysis of geologic source samples (Northwest Research 2016a). Artifacts are correlated to a parent obsidian, FGV, or basalt source (or geochemical source group) if diagnostic trace element values fall within about two standard deviations of the analytical uncertainty of the known upper and lower limits of chemical variability recorded for the source. Occasionally, visual attributes are used to corroborate the source assignments although sources are never assigned solely on the basis of megascopic characteristics.

Results of Analysis

X-Ray Fluorescence Analysis. The obsidian artifacts analyzed by X-ray fluorescence methods were correlated with two known obsidian sources, Timber Butte, Idaho and Whitewater Ridge, Oregon. The locations of the sites and the identified sources are shown in Figure 1. Analytical results are presented in Table A-1 in the Appendix and are summarized in Table 1 and Figure 2.

172 Northwest Research Obsidian Studies Laboratory Report 2016-67

Figure 1. Locations of the project sites and sources of the obsidian artifacts. The black triangles in the map above designate the location of the identified sources.

173 Northwest Research Obsidian Studies Laboratory Report 2016-67

Table 1. Summary of results of trace element analysis of the project specimens.

SITES ANALYZED –

BRITISH COLUMBIA, CANADA GEOCHEMICAL Lasqueti Island Hare Point TOTAL SOURCE Timber Butte, ID 1 - 1

Whitewater Ridge, OR - 1 1

Total 1 1 2

Figure 2 - Scatterplot of zirconium (Zr) plotted versus rubidium (Rb) for analyzed artifacts. Ellipses represent the range of geochemical variability for analyzed source reference samples.

174 Northwest Research Obsidian Studies Laboratory Report 2016-67

Information concerning the location, geologic setting, and prehistoric use of obsidian sources identified in the current investigation may be found at www.sourcecatalog.com (Northwest Research 2016b).

References Cited Northwest Research Obsidian Studies Laboratory 2016a Northwest Research Obsidian Studies Laboratory World Wide Web Site (www.obsidianlab.com).

2016b Northwest Research U. S. Obsidian Source Catalog (www.sourcecatalog.com).

175

Appendix

Results of X-Ray Fluorescence Analysis

176 Northwest Research Obsidian Studies Laboratory Table A-1. Results of XRF Studies: Lasqueti Island and Hare Point, British Columbia, Canada Specimen Trace Element Concentrations Ratios Site No. Catalog No. Rb Sr Y Zr Nb Ti Mn Ba Fe2 O 3 T Fe:Mn Fe:Ti Geochemical Source

Lasqueti Island 1 NA 216 19 45 60 37 NM NMND NM NMNM Timber Butte ± 2 1 2 2 2 NM NMNM NM Hare Point 2 EaSe71:007 140 103 27 137 10 NM NM385 NM NM NM Whitewater Ridge * ± 2 2 1 2 1 NM NMNM NM NA RGM-1 RGM-1 150 105 27 224 10 NM NM803 NM NMNM RGM-1 Reference Standard ± 3 2 2 3 2 NM NMNM NM

All trace element values reported in parts per million; ± = analytical uncertainty estimate (in ppm). Iron content reported as weight percent oxide. NA = Not available; ND = Not detected; NM = Not measured; * = Small sample; FGV = Fine-grained volcanic specimen.

177 Simon Fraser University Department of Archaeology X-Ray Fluorescence Report 2013-007

Dr. Rudy Reimer- Simon Fraser University Department of Archaeology

Abstract: Sixteen artifacts from two recorded sites, EaSe 2, EaSe 18 and an unknown site on Lasqueti Island, submitted for energy dispersive X-Ray fluorescence analysis attribute to multiple known and an unknown source. The samples were prepared and analyzed at the Department of Archaeology, Simon Fraser University, Burnaby, BC under the accession number 2013-007. Samples that assigned to known sources include Anahim Peak, Kingcome Inlet, and Glass Buttes. ______

Analytical Methods

X-Ray Fluorescence (XRF) is a nondestructive technique that is well suited to the trace element analysis of obsidian and other igneous materials. This analysis used a Bruker Tracer III-V+ portable XRF spectrometer. The system is equipped with a Peltier cooled Ag-free SiPIN, resolution ~175eV @ 5.9 KeV in an area of 12 mm. The tube’s power supply is driven by a 40 kV 1mA with a range of 4 to 40 kV. For major and trace elements including Rb, Sr, Y, Zr, and Nb that reported in Table 1. This analysis used a filter made of 0.003’’ Cu, 0.001’’ Ti and 0.012 Al with instrument power settings at 40 kV and 13 micro amps with no vacuum system (i.e. not analyzing light elements). These settings allow X-rays from 17-40 keV to reach the sample, thus efficiently exciting elements from Fe to Mo. These elements are the most useful to identify the origins of igneous rock particularly as obsidian (Ferguson 2012; Glascock and Ferguson 2012; Speakman 2012). All samples ran for a total of 180 seconds, ensured accurate and precise calculation of elemental peak data.

Results of XRF Analysis

A diagnostic trace element value most common to characterize obsidian compared directly to those directly known or established through published literature, allows for source designation. This includes known obsidian sources in the Simon Fraser University, Department of Archaeology reference collection and other known obsidian source values reported in the literature and unpublished elemental data collected through the analysis of other labs (Godfrey-Smith 1986; James et al. 1996; Moss and Erlandson 1992; Reimer n.d.). The results of the analysis of all 16 samples are in Table 1, with all values in part per million (ppm). Results derived by the utilization of Bruker’s S1CalProces that includes the calibration of results to over 45 known obsidian sources, tested and known through a variety of analysis (Ferguson 2012; Glascock and Ferguson 2012; Reimer n.d.; Speakman 2012).

As such, calculated values for artifacts submitted for analysis, subject to statistical analysis. Either correlate or do not correlate to a parent obsidian source or geochemical group, as known in the reference library. Results of this analysis are plotted with the sense of best results pertaining to diagnostic elements pertain to individual sources and

178 since elemental values fall within two standard deviations of analytical uncertainty, in this case two (95%), of the known upper and lower limits of chemical variability of known sources (Figure 1). Further statistical testing of characterization testing used principle component analysis (Figure 2) of all ten elements examined in this study that confirms characterization results in Figure 1.

Table 1. XRF data for all samples, all numbers are in parts per million (ppm). Site Source Mn Fe Zn Ga Th Rb Sr Y Zr Nb EaSe- Kingcome 338 14166 72 17 10 120 29 43 372 51 18 EaSe- Kingcome 491 19595 120 18 11 113 2 59 632 48 18 EaSe- Anahim 409 9639 274 21 44 423 32 111 144 117 18 EaSe- Anahim 504 10944 360 13 46 495 19 127 166 130 18 EaSe- Anahim 246 7525 236 23 39 392 18 111 149 119 18 EaSe- Kingcome 355 20465 116 18 8 120 4 59 642 47 18 EaSe-2 Anahim 233 7360 96 19 16 233 28 72 114 81 EaSe-2 Kingcome 345 14223 74 17 8 126 18 48 357 51 Lasqueti Unknown 277 6411 25 16 18 185 87 11 108 14 Lasqueti Unknown 299 6913 23 15 19 191 89 11 112 17 Lasqueti Unknown 367 6012 10 14 20 186 77 13 106 17 Lasqueti Unknown 367 6342 30 16 18 187 87 10 103 15 Lasqueti Unknown 408 6480 38 16 16 192 85 13 108 17 Lasqueti Unknown 284 5914 11 15 18 175 73 15 109 16 Lasqueti Unknown 362 6862 12 13 18 187 86 12 131 18 Lasqueti Glass 264 7323 26 14 3 84 36 25 89 7 Buttes

179

Figure 1. Bi-plot of Yttrium (Y) and Niobium (Nb). All numbers are in parts per million (ppm). Note samples submitted are colored red and the single red sample classified as Anahim Peak is small and raw spectral comparison confirms this characterization.

180

Figure 2. Principle Component Analysis of all ten elements for all samples plus source materials. All values in parts per million (ppm).

181 References

Carlson, Roy, L. 1994 Trade and Exchange in Prehistoric British Columbia, in Prehistoric Exchange Systems in North America (timothy G. Baugh and Jonathan E. Ericson eds.). Springer Press, 307- 365.

Erlandson, Jon, M. Madonna Moss and Richard, E. Hughes 1996 Archaeological Distribution and Trace element Geochemistry of Volcanic Glass from Obsidian Cove, Suemez Island, South East Alaska. Canadian Journal of Archaeology, 16, 89-95.

Fladmark, Knut 1984 Mountain of glass: archaeology of the Mount Edziza obsidian source, British Columbia, Canada. World Archaeology 16(2), 139-156.

1985 Glass and Ice: A Report on the Archaeology of the Mt. Edziza and Spectrum Ranges, Northwest British Columbia. Publication No. 14, Archaeology Press, Simon Fraser University, Burnaby.

Glascock, Michael, D. and Jeffrey, R. Ferguson 2012 Report Analysis of Obsidian Source Samples by Multiple Analytical Methods. Report by the Archaeometry Laboratory, University of Missouri Research Reactor. Columbia Missouri.

Godfrey-Smith, Dorothy, I. 1985 X-Ray Fluorescence Characterization of the Obsidian Flows from the Mount Edziza volcanic complex of British Columbia, Canada. Unpublished MA Thesis, Department of Archaeology, Simon Fraser University, Burnaby, BC.

James, Malcom, A. Jeff Bailey and John D’Auria 1996 A Volcanic Glass Library for the Pacific Northwest: Problems and Prospects. Canadian Journal of Archaeology, 20: 93-123.

Reimer, Rudy n.d. Obsidian Characterization for sources in the Pacific Northwest. Report on file in the Department of Archaeology, Simon Fraser University, Burnaby, B.C.

182

Appendix

Results of X-Ray Fluorescence Analysis

183 Northwest Research Obsidian Studies Laboratory Table A-1. Results of XRF Studies: Several Sites in British Columbia, Canada Specimen Trace Element Concentrations Ratios SiteNo. Catalog No. Rb Sr Y Zr Nb Ti Mn Ba Fe 23 OT Fe:Mn Fe:Ti Geochemical Source

DlSd-6 1 1 113 7 63 617 47 NM NM137 NM NMNM Kingcome (Central Coast Unknown A) * ± 21232NM NM36 NM EaSe-18 2 2 118 32 48 352 53 NM NM671 NM NMNM Unknown Vitrophyre 1 ± 22232NM NM41 NM EaSe-18 3 3 526 16 128 160 144 NM NMNM NM NMNM Anahim * ± 42323NM NMNM NM EaSe-18 4 4 91 3 60 593 46 NM NMNM NM NMNM Kingcome (Central Coast Unknown A) * ± 21242NM NMNM NM EaSe-18 5 5 461 16 113 139 126 NM NMNM NM NMNM Anahim * ± 53434NM NMNM NM EaSe-18 6 6 502 16 124 154 140 NM NMNM NM NMNM Anahim * ± 42333NM NMNM NM EaSe-18 7 7 122 0 70 652 50 NM NMNM NM NMNM Kingcome (Central Coast Unknown A) * ± 21232NM NMNM NM EaSe-18 8 8 438 15 115 148 140 NM NMNM NM NMNM Anahim * ± 52334NM NMNM NM EaSe-18 9 9 543 17 126 160 145 NM NMNM NM NMNM Anahim * ± 42333NM NMNM NM EaSe-18 10 10 122 0 70 660 52 NM NMNM NM NMNM Kingcome (Central Coast Unknown A) * ± 21242NM NMNM NM DlSd-11 11 11 145 15 54 374 60 NM NMNM NM NMNM Central Coast Unknown B * ± 21232NM NMNM NM EaSe-2 13 13 130 15 51 371 58 NM NMNM NM NMNM Central Coast Unknown B * ± 32233NM NMNM NM Lasqueti Island 14 14 78 36 26 85 5 NM NM913 NM NM NM Unknown Vitrophyre 2 ± 22222NM NM35 NM EaSe-76 15 15 128 6 70 668 50 NM NMNM NM NMNM Kingcome (Central Coast Unknown A) * ± 21232NM NMNM NM

184 Northwest Research Obsidian Studies Laboratory Table A-1. Results of XRF Studies: Several Sites in British Columbia, Canada Specimen Trace Element Concentrations Ratios SiteNo. Catalog No. Rb Sr Y Zr Nb Ti Mn Ba Fe 23 OT Fe:Mn Fe:Ti Geochemical Source

EaSe-76 16 16 136 4 75 669 50 NM NMNM NM NMNM Kingcome (Central Coast Unknown A) * ± 32243NM NMNM NM EaSe-76 17 17 132 8 74 670 55 NM NMNM NM NMNM Kingcome (Central Coast Unknown A) * ± 21232NM NMNM NM EaSe-76 18 18 121 6 72 650 51 NM NMNM NM NMNM Kingcome (Central Coast Unknown A) * ± 42354NM NMNM NM NA RGM-1 RGM-1 154 111 26 226 8 NM NM758 NM NMNM RGM-1 Reference Standard ± 22232NM NM38 NM

185 Table 1. Summary of results of trace element analysis of obsidian and vitrophyre artifacts.

PROJECT SITES

Lasqueti OBSIDIAN SOURCE DlSd-6 DlSd-11 EaSe-2 Ea-Se-18 EaSe-76 Island TOTAL

Anahim ––– 5 – – 5

Central Coast Unknown B – 1 1 – – – 2

Kingcome (Central Coast Unknown A) 1 – – 3 4 – 8

Unknown Vitrophyre 1 – – – 1 – – 1

Unknown Vitrophyre 2 – – – – – 1 1

TOTAL 1 1 1 9 4 1 17

Figure 1. Scatterplot of zirconium (Zr) plotted versus rubidium (Rb) for all analyzed artifacts.

186

Appendix

Results of X-Ray Fluorescence Analysis

187 Northwest Research Obsidian Studies Laboratory Table A-1. Results of XRF Studies: EaSe-76, Cochrane Bay, British Columbia, Canada Specimen Trace Element Concentrations Ratios SiteNo. Catalog No. Zn Pb Rb Sr Y Zr Nb Ti Mn Ba Fe 23 OT Fe:Mn Fe:Ti Geochemical Source

EaSe-76 1 1 136 19 132 13 68 650 51 NM NMNM NM 38.3104.3 Kingcome (Central Coast Unknown A) * ± 1664947297 NMNM NM

EaSe-76 2 2 131 19 111 11 60 582 46 NM NMNM NM 45.0117.1 Kingcome (Central Coast Unknown A) * ± 1764947297 NMNM NM

EaSe-76 3 3 185 53 438 20 116 151 141 NM NMNM NM 21.5137.0 Anahim * ± 1655947297 NMNM NM

EaSe-76 4 4 120 13 115 7 61 588 42 969 39977 2.08 42.568.4 Kingcome (Central Coast Unknown A) ± 1654947299 3325 0.14

EaSe-76 5 5 166 19 138 7 69 667 48 NM NMNM NM 50.3103.9 Kingcome (Central Coast Unknown A) * ± 1654947298 NMNM NM

EaSe-76 6 6 125 20 121 29 67 621 49 NM NMNM NM 35.573.9 Kingcome (Central Coast Unknown A) * ± 1654947297 NMNM NM

NA RGM-1 RGM-1 23 28 159 105 23 221 10 1608 287723 1.77 50.335.8 RGM-1 Reference Standard ± 18549472101 3225 0.14

188 189

Appendix

Results of X-Ray Fluorescence Analysis

190 Northwest Research Obsidian Studies Laboratory Table A-1. Results of XRF Studies: Grief Point (DkSd-1), Kleh Kwa Num (DlSd-6), and Lang Bay Sites, British Columbia, Canada Specimen Trace Element Concentrations Ratios SiteNo. Catalog No. Zn Pb Rb Sr Y Zr Nb Ti Mn Ba Fe 23 OT Fe:Mn Fe:Ti Geochemical Source

DkSd-1 1 1 122 17 120 6 65 621 44 NM NMNM NM 43.2104.1 Kingcome (Central Coast Unknown A) * ± 18 7 5 10 4 8 2 NM NMNM NM

DkSd-1 2 2 143 19 124 5 64 639 48 733 43593 2.59 48.1110.3 Kingcome (Central Coast Unknown A) ± 16 6 4 10 4 7 2 99 3324 0.14

DkSd-1 3 3 45 14 36 278 10 99 9 690 619741 0.71 10.435.9 Garibaldi ± 1654947299 3324 0.14

DlSd-6 4 4 118 23 122 11 66 612 45 NM NMNM NM 34.843.4 Kingcome (Central Coast Unknown A) * ± 16549472NM NMNM NM

Lang Bay 1 1 41 23 135 69 23 120 11 NM NMNM NM 30.548.0 Whitewater Ridge * ± 16549472NM NMNM NM

Lang Bay 2 2 51 9 35 294 11 94 6 NM NMNM NM 16.039.2 Garibaldi * ± 16549472NM NMNM NM

Lang Bay 3 3 60 24 95 171 27 84 14 NM NMNM NM 14.2157.8 Gregory Creek * ± 16649472NM NMNM NM

191 192 Northwest Research Obsidian Studies Laboratory Report 2009-07

X-Ray Fluorescence Analysis of Artifact Obsidian from EaSe-19 and EaSe-67, British Columbia, Canada

Craig E. Skinner and Jennifer J. Thatcher Northwest Research Obsidian Studies Laboratory

Four obsidian artifacts from EaSe-19 (N=3) and EaSe-67 (N=1), British Columbia, Canada, were submitted for energy dispersive X-ray fluorescence trace element provenance analysis. The samples were prepared and analyzed at the Northwest Research Obsidian Studies Laboratory under the accession number 2009-07.

Analytical Methods

X-Ray Fluorescence Analysis. Nondestructive trace element analysis of the samples was completed using a Spectrace 5000 energy dispersive X-ray fluorescence spectrometer. The system is equipped with a Si(Li) detector with a resolution of 155 eV FHWM for 5.9 keV X-rays (at 1000 counts per second) in an area 30 mm2. Signals from the spectrometer are amplified and filtered by a time variant pulse processor and sent to a 100 MHZ Wilkinson type analog-to-digital converter. The X-ray tube employed is a Bremsstrahlung type, with a rhodium target, and 5 mil Be window. The tube is driven by a 50 kV 1 mA high voltage power supply, providing a voltage range of 4 to 50 kV. For the elements Zn, Rb, Sr, Y, Zr, Nb, and Pb that are reported in Table A-1, we analyzed the collection with a collimator installed and used a 45 kV tube voltage setting and 0.60 mA tube current setting.

The diagnostic trace element values used to characterize the samples are compared directly to those for known obsidian sources reported in the literature and with unpublished trace element data collected through analysis of geologic source samples (Northwest Research 2009a). Artifacts are correlated to a parent obsidian source (or geochemical source group) if diagnostic trace element values fall within about two standard deviations of the analytical uncertainty of the known upper and lower limits of chemical variability recorded for the source. Occasionally, visual attributes are used to corroborate the source assignments although sources are never assigned solely on the basis of megascopic characteristics.

Additional details about specific analytical methods and procedures used for the analysis of the elements reported in Table A-1 are available at the Northwest Research Obsidian Studies Laboratory World Wide Web site at www.obsidianlab.com.

Results of Analysis

X-Ray Fluorescence Analysis. Two geochemical obsidian sources or source groups were identified among the four obsidian artifacts that were characterized by X-ray fluorescence analysis. The locations of the sites and the obsidian sources are shown in Figure 1. Analytical results are presented in Table A-1 in the Appendix and are summarized in Table 1 and Figure 2.

193 Northwest Research Obsidian Studies Laboratory Report 2009-07

Table 1. Summary of results of trace element studies of artifacts from the project sites.

Project Sites

Obsidian Source EaSe-19 EaSe-67 Total

Kingcome (Central Coast Unknown A/B) 2 1 3

Unknown Variety A 1 – 1

Total 3 1 4

Figure 1. Locations of the project sites and the identified obsidian source.

194 Northwest Research Obsidian Studies Laboratory Report 2009-07

Figure 2. Scatterplot of zirconium (Zr) plotted versus rubidium (Rb) for the analyzed artifacts.

The Kingcome obsidian source has only been recently located (in 2008) and is thought to be the same one that Carlson (1994) formerly identified as either the Central Coast Unknown A or Central Coast Unknown B sources. At the present time, we are uncertain as to which of these two unknown source groups correspond to the newly-discovered Kingcome source.

The Unknown Type A source also almost certainly corresponds to either the Central Coast Unknown A or B source but belongs to a different unknown source than the Kingcome glass. Again, we are unable at this time to distinguish which of the two Central Coast unknowns is equivalent to Unknown Type A.

Additional information about the obsidian sources identified in the current investigation may also be found at www.sourcecatalog.com (Northwest Research 2009b).

References Cited

Carlson, Roy L. 1994 Exchange in British Columbia. In Prehistoric Exchange Systems in North America, edited by Timothy G. Baugh and Jonathon E. Ericson, pp. 307–361.

195 Northwest Research Obsidian Studies Laboratory Report 2009-07

Northwest Research Obsidian Studies Laboratory 2009a Northwest Research Obsidian Studies Laboratory World Wide Web Site (www.obsidianlab.com).

2009b Northwest Research United States and Canada Obsidian Source Catalog (www.sourcecatalog.com).

196

Appendix

Results of X-Ray Fluorescence Analysis

197 Northwest Research Obsidian Studies Laboratory Table A-1. Results of XRF Studies: EaSe-19 and EaSe-67, British Columbia, Canada Specimen Trace Element Concentrations Ratios Site No. Catalog No. Zn Pb Rb Sr Y Zr Nb Ti Mn Ba Fe 23 OT Fe:Mn Fe:Ti Geochemical Source

EaSe-19 1 2008-01 142 16 130 5 71 657 51 837 45299 2.88 51.1107.2 Kingcome (Central Coast Unknown A/B) ± 16 6 4 11 4 7 2 99 3329 0.14 EaSe-19 2 2008-12 88 22 141 34 51 348 59 626 356633 1.43 33.373.1 Unknown Type A ± 1654947298 3231 0.14 EaSe-19 3 2008-13 140 20 124 5 66 633 53 NM NMNM NM 51.1115.9 Kingcome (Central Coast Unknown A/B) * ± 16 5 4 11 4 7 2 NM NMNM NM EaSe-67 4 2008-39 NM NM 137 9 72 651 52 NM NMNM NM 49.856.0 Kingcome (Central Coast Unknown A/B) * ± NMNM59482NM NMNM NM

NA RGM-1 RGM-1 30 22 163 111 25 232 13 1794 288793 2.03 56.736.4 RGM-1 Reference Standard ± 17549472101 3230 0.14

198 Northwest Research Obsidian Studies Laboratory Report 2009-46

X-Ray Fluorescence Analysis and Obsidian Hydration Measurement of Artifact Obsidian from Lasqueti Island and EaSe-76, British Columbia, Canada

Craig E. Skinner and Jennifer J. Thatcher Northwest Research Obsidian Studies Laboratory

Four obsidian artifacts from Lasqueti Island (N=1) and EaSe-76 (N=3), British Columbia, Canada, were submitted for energy dispersive X-ray fluorescence trace element provenance analysis. The specimen from Lasqueti Island was also processed for hydration rim measurements. The samples were prepared and analyzed at the Northwest Research Obsidian Studies Laboratory under the accession number 2009-46.

Analytical Methods

Obsidian Hydration Analysis. An appropriate section of each artifact is selected for hydration slide preparation. Two parallel cuts are made into the edge of the artifact using a lapidary saw equipped with 4-inch diameter diamond-impregnated .004" thick blades. The resultant cross-section of the artifact (approximately one millimeter thick) is removed and mounted on a petrographic microscope slide with Lakeside thermoplastic cement and is then ground to a final thickness of 30-50 microns.

The prepared slide is measured using an Olympus BHT petrographic microscope fitted with a video micrometer unit and a digital imaging video camera. When a clearly defined hydration layer is identified, the section is centered in the field of view to minimize parallax effects. Four rim measurements are

199 Northwest Research Obsidian Studies Laboratory Report 2009-46 typically recorded for each artifact or examined surface. Hydration rinds smaller than one micron often cannot be resolved by optical microscopy. Hydration thicknesses are reported to the nearest 0.1 μm and represent the mean value for all readings. Standard deviation values for each measured surface indicate the variability for hydration thickness measurements recorded for each specimen. It is important to note that these values reflect only the reading uncertainty of the rim values and do not take into account the resolution limitations of the microscope or other sources of uncertainty that enter into the formation of hydration rims.

Additional details about specific analytical methods and procedures used for the analysis of the elements reported in Table A-1 and the preparation and measurement of hydration rims are available at the Northwest Research Obsidian Studies Laboratory World Wide Web site at www.obsidianlab.com (Northwest Research 2009a).

Results of Analysis

X-Ray Fluorescence Analysis. Three geochemical obsidian sources or source groups were identified among the four obsidian artifacts that were characterized by X-ray fluorescence analysis. The locations of the sites and the obsidian sources are shown in Figure 1. Analytical results are presented in Table A-1 in the Appendix and are summarized in Table 1 and Figure 2.

Figure 1. Locations of the sites and the sources of the analyzed artifacts.

200 Northwest Research Obsidian Studies Laboratory Report 2009-46

Table 1. Summary of results of trace element studies of artifacts from the project sites.

PROJECT SITES

OBSIDIAN SOURCE Lasqueti Island EaSe-76 TOTAL

Kingcome (Central Coast Unknown A) – 2 2

Unknown 1 1 – 1

Whitewater Ridge – 1 1

TOTAL 1 3 4

Figure 2. Scatterplot of zirconium (Zr) plotted versus rubidium (Rb) for the analyzed artifacts.

201 Northwest Research Obsidian Studies Laboratory Report 2009-46

The Kingcome obsidian source has only been recently located (in 2008) and is the same one that Carlson (1994) formerly identified as the Central Coast Unknown A source. The Whitewater Ridge source is located in northcentral Oregon and nodules of raw material are found at many geographically-scattered localities in the Bear Valley region of the state (Ambroz 1997; Skinner 1995a, 1995b, Skinner and Thatcher 2009).

The trace element profile of the artifact from Lasqueti Island failed to match any of the United States or Canadian obsidian sources currently contained in our source reference database.

Additional information about the obsidian sources identified in the current investigation may also be found at www.sourcecatalog.com (Northwest Research 2009b).

Obsidian Hydration Analysis. The obsidian artifact from Lasqueti Island that was characterized by X-ray fluorescence analysis was prepared for obsidian hydration analysis and yielded a hydration rim of 2.7 microns. The specimen slide is curated at the Northwest Research Obsidian Studies Laboratory under accession number 2009-46. The results are reported in Table B-1 in the Appendix.

References Cited

Ambroz, Jessica A. 1997 Characterization of Archaeologically Significant Obsidian Sources in Oregon by Neutron Activation Analysis. Unpublished Master's Thesis, Department of Chemistry, University of Missouri, Columbia, Missouri.

Carlson, Roy L. 1994 Exchange in British Columbia. In Prehistoric Exchange Systems in North America, edited by Timothy G. Baugh and Jonathon E. Ericson, pp. 307–361.

Northwest Research Obsidian Studies Laboratory 2009a Northwest Research Obsidian Studies Laboratory World Wide Web Site (www.obsidianlab.com).

2009b Northwest Research United States and Canada Obsidian Source Catalog (www.sourcecatalog.com).

Skinner, Craig E. 1995a Obsidian Characterization Studies. In Archaeological Investigations, PGT-PG&E Pipeline Expansion Project, Idaho, Washington, Oregon, and California, Volume V: Technical Studies, by Robert U. Bryson, Craig E. Skinner, and Richard M. Pettigrew, pp. 4.1–4.54. Report prepared for Pacific Gas Transmission Company, Portland, Oregon, by INFOTEC Research, Inc., Fresno, California, and Far Western Anthropological Research Group, Davis, California.

1995b Obsidian Hydration Studies. In Archaeological Investigations, PGT-PG&E Pipeline Expansion Project, Idaho, Washington, Oregon, and California, Volume V: Technical Studies, by Robert U. Bryson, Craig E. Skinner, and Richard M. Pettigrew, pp. 5.1-5.51. Report prepared for Pacific Gas Transmission Company, Portland, Oregon, by INFOTEC Research, Inc., Fresno, California, and Far Western Anthropological Research Group, Davis, California.

Skinner, Craig E. and Jennifer J. Thatcher 2009 Obsidian Studies in the Bear Valley Area, Malheur Forest, Oregon - Part One: The Sources. Report 2009-01 prepared for the Malheur National Forest, John Day, Oregon, by Northwest Research Obsidian Studies Laboratory, Corvallis, Oregon.

202

Appendix

Results of X-Ray Fluorescence and Obsidian Hydration Analysis

203 Northwest Research Obsidian Studies Laboratory Table A-1. Results of XRF Studies: Lasqueti Island Isolate and EaSe-76, British Columbia, Canada Specimen Trace Element Concentrations Ratios Site No. Catalog No. Zn Pb Rb Sr Y Zr Nb Ti Mn Ba Fe 23 OT Fe:Mn Fe:Ti Geochemical Source

Lasqueti Island 1 Lasqueti:001 59 25 168 14 28 123 14 447 26193 1.05 33.975.9 Unknown 1 ± 1654947298 3224 0.14 EaSe-76 2 EaSe76:009 116 17 127 ND 66 640 49 NM NMNM NM 46.2108.2 Kingcome (Central Coast Unknown A) * ± 16 6 4 ND 4 7 2 NM NMNM NM

EaSe-76 3 EaSe76:029 38 15 130 95 27 138 8 NM NMNM NM 38.938.6 Whitewater Ridge ± 17549472NM NMNM NM EaSe-76 4 EaSe76:039 141 12 129 7 64 638 48 NM NMNM NM 44.3105.9 Kingcome (Central Coast Unknown A) * ± 16649472NM NMNM NM

NA RGM-1 RGM-1 41 24 163 110 23 226 12 1651 374713 1.92 41.837.5 RGM-1 Reference Standard ± 16549472101 3325 0.14

204 Northwest Research Obsidian Studies Laboratory Table B-1. Obsidian Hydration Results and Sample Provenience: Lasqueti Island Isolate and EaSe-76, British Columbia, Canada Specimen Artifact Hydration Rims B Site No. Catalog No. Unit Depth Type A Artifact Source Rim 1 Rim 2 Comments

Lasqueti Island 1 Lasqueti:001 -- -- EMP Unknown 1 2.7± NM0.1NM± --

EaSe-76 2 EaSe76:009 39 cmN/38cmE Level 1 EMP Kingcome (Central Coast Unknown A) * NM± NMNMNM± --

EaSe-76 3 EaSe76:029 Unit 1 Level 2 DEB Whitewater Ridge NM± NMNMNM± --

EaSe-76 4 EaSe76:039 Unit 1 Level 3 DEB Kingcome (Central Coast Unknown A) * NM± NMNMNM± --

A DEB = Debitage; EMP = Edge-Modified Piece B See text for explanation of comment abbreviations NA = Not Available; NM = Not Measured; * = Small sample 205B Northwest Research Obsidian Studies Laboratory Report 2009-46

Abbreviations and Definitions Used in the Comments Column

All hydration rim measurements are recorded in microns.

BEV - (Beveled). Artifact morphology or cut configuration resulted in a beveled thin section edge.

BRE - (BREak). The thin section cut was made across a broken edge of the artifact. Resulting hydration measurements may reveal when the artifact was broken, relative to its time of manufacture.

DES - (DEStroyed). The artifact or flake was destroyed in the process of thin section preparation. This sometimes occurs during the preparation of extremely small items, such as pressure flakes.

DFV - (Diffusion Front Vague). The diffusion front, or the visual boundary between hydrated and unhydrated portions of the specimen, are poorly defined. This can result in less precise measurements than can be obtained from sharply demarcated diffusion fronts. The technician must often estimate the hydration boundary because a vague diffusion front often appears as a relatively thick, dark line or a gradation in color or brightness between hydrated and unhydrated layers.

DIS - (DIScontinuous). A discontinuous or interrupted hydration rind was observed on the thin section.

HV - (Highly Variable). The hydration rind exhibits variable thickness along continuous surfaces. This variability can occur with very well- defined bands as well as those with irregular or vague diffusion fronts.

IRR - (IRRegular). The surfaces of the thin section (the outer surfaces of the artifact) are uneven and measurement is difficult.

1SO - (1 Surface Only). Hydration was observed on only one surface or side of the thin section.

NOT - (NOT obsidian). Petrographic characteristics of the artifact or obsidian specimen indicate that the specimen is not obsidian.

NVH - (No Visible Hydration). No hydration rind was observed on one or more surfaces of the specimen. This does not mean that hydration is absent, only that hydration was not observed. Hydration rinds smaller than one micron often are not birefringent and thus cannot be seen by optical microscopy. "NVH" may be reported for the manufacture surface of a tool while a hydration measurement is reported for another surface, e.g. a remnant ventral flake surface.

OPA - (OPAque). The specimen is too opaque for measurement and cannot be further reduced in thickness.

PAT - (PATinated). This description is usually noted when there is a problem in measuring the thickness of the hydration rind, and refers to the unmagnified surface characteristics of the artifact, possibly indicating the source of the measurement problem. Only extreme patination is normally noted.

REC - (RECut). More than one thin section was prepared from an archaeological specimen. Multiple thin sections are made if preparation quality on the initial specimen is suspect or obviously poor. Additional thin sections may also be prepared if it is perceived that more information concerning an artifact's manufacture or use can be obtained.

UNR - (UNReadable). The optical quality of the hydration rind is so poor that accurate measurement is not possible. Poor thin section preparation is not a cause.

WEA - (WEAthered). The artifact surface appears to be damaged by wind erosion or other mechanical action.

206

Appendix

Results of X-Ray Fluorescence Analysis

207 Northwest Research Obsidian Studies Laboratory Table A-1. Results of XRF Studies: Lasqueti Island Obsidian Artifacts, British Columbia, Canada Specimen Trace Element Concentrations Ratios SiteNo. Catalog No. Zn Pb Rb Sr Y Zr Nb Ti Mn Ba Fe 23 OT Fe:Mn Fe:Ti Geochemical Source

Lasqueti Island 1 1 44 12 115 86 24 132 9 904 2661472 0.86 28.132.7 Whitewater Ridge ± 16549472101 3225 0.14

Lasqueti Island 2 2 26 11 109 79 18 136 19 1178 334521 0.97 25.028.0 Lasqueti Unknown Obsidian 3 ± 17549472100 3224 0.14

Lasqueti Island 3 3 29 14 35 305 12 98 8 1012 447845 0.91 17.830.7 Garibaldi ± 16549472100 3324 0.14

NA RGM-1 RGM-1 52 31 161 106 26 224 9 1612 495761 1.85 30.737.1 RGM-1 Reference Standard ± 16549472101 3325 0.14

208 2009-61-1 2009-61-2 2009-61-3

209