Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An archaeology of food and settlement on the Northwest Coast McKechnie, Iain Mitchell Patrick 2013

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

Item Metadata


24-ubc_2014_spring_mckechnie_iain.pdf [ 72.87MB ]
JSON: 24-1.0072131.json
JSON-LD: 24-1.0072131-ld.json
RDF/XML (Pretty): 24-1.0072131-rdf.xml
RDF/JSON: 24-1.0072131-rdf.json
Turtle: 24-1.0072131-turtle.txt
N-Triples: 24-1.0072131-rdf-ntriples.txt
Original Record: 24-1.0072131-source.json
Full Text

Full Text

AN ARCHAEOLOGY OF FOOD AND SETTLEMENT  ON THE NORTHWEST COAST  by Iain Mitchell Patrick McKechnie  B.A., University of California, Santa Cruz, 1999 M.A., Simon Fraser University, 2005  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Anthropology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)   December 2013  © Iain Mitchell Patrick McKechnie, 2013 – ii –  Abstract This dissertation examines multiple scales of Indigenous history on the Northwest Coast from the disciplinary perspective of archaeology. I focus on cultural lifeways archaeologically represented in two key domains of human existence: food and settlement. The dissertation consists of six individual case studies that demonstrate the utility of applying multiple spatial and temporal scales to refine archaeological understanding of cultural and historical variability on the Northwest Coast over the Mid-to-Late Holocene (ca. 5,000-200 BP). The first of three regionally scaled analyses presents a coast-wide examination of fisheries data indicating that Pacific herring (Clupea pallasii) exhibit a pervasive and previously under-recognized importance in Northwest Coast Indigenous subsistence practices. Next, I use zooarchaeological data from the southern British Columbia coast to identify a pattern of regional coherence in Coast Salish and Nuu-chah-nulth hunting traditions reflecting the scale of intergenerational cultural practice. The third study re-calibrates the settlement history of a small and historically significant locality in Coast Tsimshian territory (Prince Rupert Harbour) to clarify the temporal resolution of existing radiocarbon datasets and test inferences about social and political change. Following this regional exploration of scale, I document site-specific temporal variability in archaeological fisheries data from a Nuu-chah-nulth ‘big-house’ reflecting climatic and socio-economic change. I examine Indigenous oral histories and archaeological datasets to evaluate these parallel records of settlement in the neighbouring territory of an autonomous Nuu-chah-nulth polity before and during the occupation of a large defensive fortress. Finally, I demonstrate how everyday foodways are archaeologically expressed and reflect ecological differences and active management strategies within several spatially associated sites over millennial – iii –  timescales. These linked case studies offer new clarity into long-standing debates concerning archaeologically relevant scales of cultural-historical variability on the NWC. They collectively demonstrate an enduring regional and temporal coherence for key aspects of indigenous resource use and settlement and a historical dynamism at finer scales. I argue this has cultural, historical, and archaeological significance as well as relevance for contemporary understandings of the Northwest Coast environment. I conclude that a focus on the pervasive aspects of the everyday over millennia offers insight into individual actions across broader patterns of history.   – iv –  Preface Chapter 2 is a multi-authored manuscript that was submitted for publication on September 16, 2013 and accepted for publication on December 12, 2013. I am the principal author and analyst on this manuscript, which has been co-written primarily with Dana Lepofsky. Other authors include Madonna L. Moss, Virginia L. Butler, Trevor J. Orchard, Gary Coupland, Fredrick Foster, Megan Caldwell, and Ken Lertzman. Aubrey Cannon has generously provided unpublished data and commented on multiple versions of this manuscript. Andrew Martindale provided column and auger sample fauna for osteological identification by the author.  Chapter 3 has been previously published as: McKechnie, Iain and Rebecca J. Wigen (2011) In Human Impacts on Seals, Sea Lions, and Sea Otters: Integrating Archaeology and Ecology in the Northeast Pacific, edited by Todd J. Braje and Torben C. Rick, pp. 129–166. University of California Press, Berkeley. © 2011 by the Regents of the University of California. I designed the study, compiled data, completed quantitative and cartographic analyses, wrote much of the chapter and worked with co-author Rebecca Wigen on developing concepts, revisions and edits.  Chapter 4 is a manuscript co-authored with Morley Eldridge. I conducted background research and designed and conducted calibration analyses, and co-wrote the paper and conducted the bulk of the revisions on multiple drafts over a period of four years. An initial version of this chapter was originally presented in 2008 at the Society for American Archaeology meetings in Vancouver and subsequent drafts have been heavily revised in correspondence with Andrew Martindale and Susan Marsden.  Chapter 5 has been previously published as: McKechnie, Iain (2012) Zooarchaeological Analysis of the Indigenous Fishery at the Huu7ii Big House and Back Terrace, Huu-ay-aht Territory, Southwestern Vancouver Island. In Huu7ii: Household Archaeology at a         Nuu-chah-nulth Village Site in Barkley Sound, by Alan D. McMillan and Denis E. St. Claire, pp. 154–186. © 2012 Archaeology Press, Simon Fraser University, Burnaby, BC. This manuscript was conducted in the service of the Huu7ii Archaeological Project with the support of the Huu-ay-aht First Nation and the 2006 University of Victoria archaeological field school. I took part in the primary fieldwork, field based sampling, coordinated sorting of samples and conducted osteological identification and am the sole author of this manuscript, which was written in 2011 and completed in the Spring of 2012. – v –  Table of Contents Abstract .................................................................................................................................... ii	  Preface ..................................................................................................................................... iv	  Table of Contents .....................................................................................................................v	  List of Tables ........................................................................................................................ viii	  List of Figures ......................................................................................................................... ix	  List of Abbreviations ........................................................................................................... xiv	  Glossary of Technical Terms ................................................................................................ xv	  Acknowledgements .............................................................................................................. xvi	  Dedication ........................................................................................................................... xviii	  Chapter 1.	   Archaeological Scales of Analysis on the Northwest Coast ........................ 1	  Introduction ....................................................................................................................... 1	  Theoretical Framework ..................................................................................................... 2	  Food and Place as Multi-scalar ......................................................................................... 6	  Dissertation Structure ...................................................................................................... 11	  Conclusion ...................................................................................................................... 18	  Chapter 2.	   Archaeological Data Provide Alternative Hypotheses on Pacific Herring (Clupea pallasii) Distribution, Abundance, and Variability on the Northwest Coast 19	  Introduction ..................................................................................................................... 19	  Material and Methods ..................................................................................................... 26	  Results ............................................................................................................................. 31	  Discussion ....................................................................................................................... 41	  Conclusion ...................................................................................................................... 49	  Chapter 3.	   Towards a Historical Ecology of Pinniped and Sea Otter Hunting Traditions on the Coast of Southern British Columbia ................................................ 51	  – vi –  Introduction ..................................................................................................................... 51	  Context ............................................................................................................................ 52	  Methods........................................................................................................................... 59	  Results ............................................................................................................................. 64	  Discussion ....................................................................................................................... 80	  Conclusions ..................................................................................................................... 87	  Chapter 4.	   Re-Calibrating Archaeological Chronologies on the Northern Northwest Coast: Radiocarbon Data from Prince Rupert Harbour .............................................. 91	  The Importance of Calibration in Radiocarbon Dating .................................................. 92	  Calibrating Radiocarbon Age Estimates for Marine Samples ........................................ 99	  Variable Reporting of Radiocarbon Dates and Calibration Results ............................. 106	  Summed Probability as an Analytical Method ............................................................. 110	  Recalibrating the Radiocarbon Record for Prince Rupert Harbour .............................. 113	  Calibration Issues in Village Abandonment Dates from Shell ..................................... 122	  Recalibrating Radiocarbon Dates on Human Remains from Prince Rupert Harbour .. 134	  Interpretive Implications of Recalibrations on Individual Dates .................................. 138	  Conclusions ................................................................................................................... 141	  Chapter 5.	   Zooarchaeological Analysis of the Indigenous Fishery at the Huu7ii Big House & Back Terrace, Huu-ay-aht Territory, Southwestern Vancouver Island .... 144	  Methods......................................................................................................................... 145	  Results ........................................................................................................................... 157	  Discussion and Interpretation ....................................................................................... 192	  Conclusions ................................................................................................................... 201	  Chapter 6.	   Indigenous Oral History & Settlement Archaeology in the Broken Group Islands, Western Vancouver Island .............................................................................. 203	  Indigenous Oral History and the Archaeological Record ............................................. 203	  Translating Between Archaeological and Oral Historical Information ........................ 205	  Objectives ..................................................................................................................... 206	  – vii –  Analysis of Archaeological and Oral Historical Sequences ......................................... 207	  Historical Background .................................................................................................. 208	  The Broken Group Study Area ..................................................................................... 211	  Settlement History of Ts’ishaa, The Ts’ishaa7ath Origin Village ................................ 221	  Oral Historical Accounts of the Maktl7ii7ath Local Group Territory .......................... 224	  Archaeological Case Studies in Maktl7ii7ath Local Group Territory .......................... 228	  Discussion ..................................................................................................................... 248	  Conclusion .................................................................................................................... 253	  Chapter 7.	   Zooarchaeological Scales of Variability on the Northwest Coast: A Barkley Sound Case Study ............................................................................................. 257	  Introduction ................................................................................................................... 257	  Theoretical and Methodological Context ...................................................................... 258	  Analysis of Scale and Variability ................................................................................. 262	  Methods......................................................................................................................... 266	  Results ........................................................................................................................... 267	  Discussion ..................................................................................................................... 278	  Conclusion .................................................................................................................... 284	  Chapter 8.	   Conclusion: Significance, Limitations, and Future Directions .............. 287	  Limitations .................................................................................................................... 298	  Variation and Scale into the Future .............................................................................. 300	  References .............................................................................................................................303	  Appendices For Chapter 2 ..................................................................................................365	    – viii –  List of Tables Table 2.1 Archaeological data for Pacific herring for all sites (N = 171)…………………..366 Table 2.2 Individual assemblage data for the 50 sites with within site…………………......376 Table 2.3 Distance between archaeological sites and herring spawning locations…………379 Table 3.1 Ecological characteristics for pinnipeds and sea otters in British Columbia. ......... 61	  Table 3.2 Site Names, Locations, References, and Number of Identified Specimens. .......... 73	  Table 3.3 Sex Distribution of Steller Sea Lion Remains from Selected Sites. ....................... 79	  Table 4.1 Paired marine and terrestrial radiocarbon dates from Prince Rupert Harbour. .... 119	  Table 4.2 Recalibrated radiocarbon dates on shell from uppermost shell midden layers ..... 119	  Table 4.3 Carbon isotope values. .......................................................................................... 131	  Table 4.4 Recalibrated radiocarbon dates on human bone collagen. .................................... 132	  Table 5.1 Column samples containing identified fish remains screened .............................. 155	  Table 5.2. List of identified taxa (NISP) in the column sample assemblage. ....................... 155	  Table 6.1. The 15 largest shell midden sites in the Broken Group Islands ........................... 219	  Table 6.2. Lineage household names (ushtakimilh) names at the village of Maktl7ii .......... 226	  Table 6.3 Radiocarbon dates from archaeological sites in the Maktl7ii7ath study area. ...... 255	  Table 7.1 Examined vertebrate remains from the seven study sites ..................................... 269	  Table 7.2 Examined invertebrate remains from the seven study sites .................................. 270	  Table 7.3. Mean lengths and standard deviations in cm for all measured fish ..................... 273	   – ix –  List of Figures Figure 1.1 Map of the Northwest Coast .................................................................................. 12	  Figure 1.2 Temporal scales of analysis. .................................................................................. 13	  Figure 2.1. Percent abundance of herring bones in archaeological sites ................................ 30 Figure 2.3. Boxplots showing percent abundance of herring bones ....................................... 32	  Figure 2.4. Percent abundance of herring bones in archaeological sites by sub-regions. ....... 32	  Figure 2.5. Percent abundance of herring bones over time in sites ........................................ 35	  Figure 2.6. Mean rank order and abundance of archaeological herring bones ....................... 38	  Figure 2.7. Boxplot showing variability in the archaeological abundance of herring bones. . 39	  Figure 2.8. Schematic figure representing three alternative hypotheses regarding the relationship between modern and archaeological abundance and variability of herring populations. ............................................................................................................................. 45	  Figure 3.1. Map of the study area showing. ............................................................................ 64	  Figure 3.2. Relative composition of marine and terrestrial mammals (%NISP) for all sites. 69 Figure 3.4. Relative abundance of marine versus terrestrial mammals for sites dating to within the four examined periods. .......................................................................................... 95 Figure 4.2. Example of calibration effects on a radiocarbon date with a large error range. ... 97	  Figure 4.3. Map of the northern Northwest Coast showing .................................................. 102	  Figure 4.4. Schematic depicting the steps required for radiocarbon calibration. ................. 105	  Figure 4.5. An example of three typical radiocarbon dates on terrestrial charcoal .............. 111	  Figure 4.6. Local marine reservoir (Delta-R) estimates over the Holocene ......................... 118	  Figure 4.7. Village abandonment data obtained from Archer 2001 and Archer 1992 .......... 125	  – x –  Figure 4.8 Comparison of different calibration criteria applied to radiocarbon dates on human remains from Prince Rupert Harbour .................................................................................... 135	  Figure 5.1 Excavating column samples from the sidewall of excavation units .................... 146	  Figure 5.2. Plan view showing column samples locations recovered from House 1. ........... 147	  Figure 5.3. Perspective view of the Huu7ii village looking west. ........................................ 148	  Figure 5.4. Photo of column sample taken from the north wall of the back terrace unit . .... 149 Figure 5.6. Comparison between the relative abundance of fish remains in the column sample assemblage ............................................................................................................................ 159	  Figure 5.7. Taxonomic composition of the entire column sample fish assemblage. ............ 161	  Figure 5.8. Relative abundance over time for the ten most numerous fish taxa. .................. 168	  Figure 5.9. The number of specimens (NISP) and individuals (MNI) per cubic metre ....... 168	  Figure 5.10. Temporal trends (left to right) in the abundance of four important fish taxa i . 173	  Figure 5.11. Herring and anchovy abundance in individual column sample levels ............. 176	  Figure 5.12. NISP and MNI data for the entire column sample assemblage. ....................... 177	  Figure 5.13. Histograms showing fish length estimates. ...................................................... 180	  Figure 5.14. Measured salmon vertebrae from the column sample assemblage .................. 181	  Figure 5.15. The identification rate for fish, mammals and birds in the column sample assemblage. ........................................................................................................................... 183	  Figure 5.16. Identification rate for fish remains in the 12 examined column sample assemblages. .......................................................................................................................... 183	  Figure 5.17. Estimated number of individual fish per litre in three deposits at Huu7ii.. ...... 184	  Figure 5.18. Collectors curve for taxonomic richness in the column sample assemblage. .. 185	  Figure 5.19. Taxonomic richness of the House 1 versus the back terrace assemblages. ...... 186	  – xi –  Figure 5.20. Scatterplot showing the non-linear relationship between the weight of bone and the weight of shell in individual column sample levels. ....................................................... 187	  Figure 5.21. Burning and other modifications to bone specimens in the column sample assemblage by fish, mammal, and bird categories. ............................................................... 190	  Figure 5.22. Relative percent of cranial vertebral and caudal elements for the 10-most numerous fish in the column sample assemblage. ................................................................ 191	  Figure 6.1 Territorially bounded local group areas in the Broken Group Islands during the 18th century as described in oral historical accounts synthesized by St. Claire (1991). ...... 213	  Figure 6.2. Alex Thomas (left), Frank Williams and his daughter (centre) in Port Alberni in 1914. Edward Sapir (right) pictured in New Haven in 1934 ................................................ 215	  Figure 6.3. Tseshaht place names as detailed in oral historical. ........................................... 217	  Figure 6.4 Recorded archaeological sites in the Broken Group and Sechart Channel. ........ 218	  Figure 6.5. The size and number of shell midden sites in different quadrants of the Broken Group Islands.. ...................................................................................................................... 220	  Figure 6.6. Site map of the village of Ts'ishaa (204 and 205T) showing oral historically named locations of household lineages (ushtakimilh) .......................................................... 224	  Figure 6.7. Overview of the Maktl7ii7ath study area with individual study sites ................ 225	  Figure 6.8. Oblique perspective view of Maktl7ii looking north. ......................................... 230	  Figure 6.9. Calibrated Radiocarbon dates from Maktl7ii (site 206T) at Wouwer Island. .... 231	  Figure 6.10. Terraced midden with house platforms at site 206T, Outer Wouwer Island. ... 232	  Figure 6.11. Oblique perspective view of Huumuuwaa. ...................................................... 234	  Figure 6.12. Calibrated radiocarbon dates from Huumuuwaa (site 304T). .......................... 234	  Figure 6.13. Oblique view of the archaeological deposits at the site of Shiwitis (82T).. ..... 237	  – xii –  Figure 6.14. Calibrated radiocarbon dates for Shiwitis, 82T. ............................................... 237	  Figure 6.15. Southwest facing view of Ch'ituukwachisht. .................................................... 238	  Figure 6.16. Calibrated radiocarbon dates on charcoal from deposits at upper Huts’atswilh (site 129T, top), N. Cree Island (site 132T, middle), and S. Cree (site 131T, bottom).. ...... 239	  Figure 6.17. Perspective view of a topographic model of Huts’atswilh s ............................ 241	  Figure 6.18. View from upper Huts’atswilh (site 129T) looking west. ................................ 242	  Figure 6.19. Dates from lower Huts’atswilh showing calibrated ages from core samples. .. 244	  Figure 6.20 Map of the lower village (83T) and upper defensive site (129T). ..................... 247	  Figure 6.21 Dates on cultural deposits on upper Dicebox .................................................... 248	  Figure 6.22. Settlement history in Maktl7ii7ath Local Group territory ................................ 250	  Figure 7.1. Map of the Maktl7ii7ath study area in the Broken Group Islands. .................... 261	  Figure 7.2. Arial overview of the Maktl7ii7ath study area with individual study sites. ....... 261	  Figure 7.3 Proportion of fish, birds and mammals in the examined assemblages. ............... 269	  Figure 7.4 Proportion of fish taxa by number of identified speciemens (NISP) .................. 271	  Figure 7.5. Proportion of shellfish by weight from the five study sites ................................ 271	  Figure 7.6. Distribution of fish lengths for three ubiquitous and abundant fish taxa ........... 273	  Figure 7.7. Identified fish remains from the five study sites. ............................................... 275	  Figure 7.8. Proportion of shellfish from the five study sites. ............................................... 276	  Figure 7.9. Four measures of abundance applied across the five study sites for anchovy (left) and clam (right).. ................................................................................................................... 277	  Figure 7.10. Ancient DNA identifications of rockfish in the study area (from Rodrigues, McKechnie, and Yang 2013). Size of pie chart is scaled to sample size. ............................. 280	  Figure 7.11. Anthropogenically altered beach at Shiwitis (Site 82T).. ................................. 282	  – xiii –  Figure 8.1 Illustration by John Webber depicting the inside a portion of the house of Mowachaht Muchalaht Chief Maquinna. ............................................................................. 291	  Figure 8.2 Recorded shell midden sites on the British Columbia Coast. ............................. 302	   – xiv –  List of Abbreviations  NWC – Northwest Coast NISP – Number of Identified Specimens  MNI – Minimum Number of Individuals  DFO – Department of Fisheries and Oceans Canada  TEK – Traditional Ecological Knowledge  LEK – Local Ecological Knowledge – xv –  Glossary of Technical Terms Number of Identified Specimens – Used to describe the total number of skeletal specimens which have been confidently identified to at least family, genus or species level (i.e., not including ‘unidentified’ specimens). This is not the same as miscellaneous unidentified bone specimens which are sometimes referred to as simply Number of Specimens (NSP).  Specimen – skeletal fragment. The term ‘specimen’ is used rather than ‘bone’ because skeletal elements are often broken into multiple pieces and can therefore be counted more than once in a zooarchaeological assemblage.  Taxa - a term used to describe different animals (or plants) that are not necessarily all of the same taxonomic category. For instance ‘Northern anchovy’ is a species while ‘rockfish’ (Sebastes spp.) is a genus designation (following Linnaean taxonomy).  Relative abundance (relative frequency, %NISP or % weight) – simply the percentage of a particular item (e.g., herring, clam weight) relative to all other specimens within the same category (e.g., identified fish remains, identified shell fragments).  Rank order abundance (%NISP) – the relative rank order of a particular item (e.g., herring) in comparison to all other specimens within the same category (e.g., identified fish remains, identified shell fragments).  Ubiquity – Frequency of occurrence or the percentage of contexts in which a certain item is present versus absent. For example, herring would be ‘ubiquitous’ if it was present in every level of every excavation unit at the site.  Column sample – Small (typically 10x10 cm) vertical ‘column’ of archaeological sediment excavated from the sidewalls of an excavation area.   Archaeological ‘site’ – An archaeological term describing a place with material and human altered sediments indicative of past human use and settlement. In this dissertation, I use the terms site, shell midden, and settlement interchangeably.  Shell midden – A widely used archaeological term to describe places in coastal, near-coastal, and aquatic areas containing anthropogenic (human created) sediments. On the Northwest Coast, shell middens are typically composed of varying quantities of shell, bone, fire altered rocks (e.g., boiling stones), artifacts, charcoal, greasy silt, decomposed plant and other organic remains, architectural features, and a host of other cultural materials, including isolated and multiple human interments. Shell midden sediments can range widely in volume from small instantaneously deposited lenses to massive deposits enveloping entire landforms and spanning broad periods of time (ca. 100-10,000 years).   Midden – a term that “has its roots in the Scandinavian languages, meaning material that accumulates about a dwelling place” (Stein 1996:638). – xvi –  Acknowledgements This dissertation was completed only because a number of supportive, inquisitive, positive, patient, and hard working people entrusted me with their time, equipment, and funding. I am *very* grateful to have had this opportunity to think about food with good people and at several amazing places.   I cannot thank my supervisors Andrew Martindale and Michael Blake enough for their support and mentoring over the years, for enabling and expanding my research horizons, and making time at massively important moments. Thanks to Andrew for finding ways to support me in the field, in the lab, and in energizing discussions with scrap paper and coffee. Thanks to Mike for his sagely wisdom, timely advice, and for sharing his expertise in archaeological cartography and visualization which has been useful in each chapter. Thanks also to committee members Bruce Miller and Aubrey Cannon for inspiration, insight and support in numerous timely ways over many years, and to other examiners for thoughtful critique and constructive feedback (Robert Losey, David Pokotylo, and Greg Crutsinger).   I have been especially fortunate to have had fantastic feedback from a number of co-authors on Chapters 2, 3, and 4. I owe Dana Lepofsky and Madonna Moss an inordinate amount of gratitude for their support and efforts in navigating the process of writing Chapter 2. I thank Dana in particular for her boundless energy, for bringing me into the Herring School at SFU, and timely encouragement during a re-think of the structure of this dissertation. I am grateful to Ken Lertzman, Virginia Butler, and Trevor Orchard for productive correspondence, and to Gary Coupland, Megan Caldwell and Fred Foster for contributing and organizing data. Rebecca Wigen at UVic and Morley Eldridge at Millennia were amazing collaborators during numerous revisions in Chapters 3 and 4 respectively. Thank to editors Andrew Martindale, Susan Marsden, Duncan McLaren, Torben Rick & Todd Braje.   Alan McMillan and Denis St. Claire are owed large debts of my gratitude for their support of this dissertation throughout its several stages and for being great mentors (and editors). Denis St. Claire’s boundless energy and ability to coordinate boats and equipment has been particularly vital to the Hiikwis Archaeological Project, to which this dissertation contributes. Alan and Denis were also the directors of the Huu7ii Archaeological Project – generously supported by Huu-ay-aht First Nations.  I am especially grateful to Tseshaht and Huu-ay-aht First Nations who provided me with the privilege of conducting fieldwork in their territories. Particular thanks to Les Sam, Boyd Gallic, Darrell Ross, and Chris Robinson at Tseshaht for their support of the Hiikwis Archaeological Project and to Barry Watts, Jordan Dick, Hank Gus, Jeff Gallic, and Cody Robinson for their help in the field. I especially thank Wanda Robinson for her amazing organizational and culinary ability at the Nettle Island field cabin. I thank Robert Dennis at Huu-ay-aht for enabling fieldwork at Huu7ii and Judy Johnson, Henry Williams, Arthur Peters, Gabe (Hip) Williams, and Alec Frank and numerous others for help in the field, and many volunteer UVic ‘Rockwashers’ for help in processing column samples in the lab.   Ian Sellers, Ted Knowles, Pete Dady, and Ian Sumpter are most thoroughly thanked for their patient persistence for volunteering long days of coring, augering, and mapping in the Broken Group in 2008, 2009, & 2011. Thanks also to Stephanie Sketchley for her organizational perseverance, to Bryn Letham, Alex Clark, Kelsey McLean, Butch Neilson, Mike Blake, Andrew Martindale, Ben Nelson, Audrey Dallimore, Quentin Mackie, Daryl Fedje, Adam Love, Rodney Steadman, Jacob Earnshaw, Byron Malloy, Byron Fry, Tim Sumpter, Russ Markel, and Nicole Smith for help in the field. At Parks, I thank Karen Haugen, Jen Yakimishin, Arlene Armstrong, Darren Salisbury, Mike Collyer, and Kim St. Claire (nee Seward-Hannam) and to Gwyn Langemann in Calgary. Thanks to Karey and Henk at Sechart Lodge for use of their showers and grocery delivery. Thanks to John Southon for radiocarbon dates and to Paula Reimer and Maarten Blaauw for answering 14C queries. I am further thankful to Dongya Yang, Camilla Speller, and Antonia Rodrigues for their Ancient DNA analyses. – xvii –  Patricia Ormerod, Wayne Point, Mike Blake and Andrew Martindale are thanked for facilitating the use of key UBC Lab of Archaeology mapping/coring/augering equipment and enabling boat and van transportation which was absolutely critical to this project. I owe massive thanks to Steve Weisman for hundreds of hours of sorting and conversation in the LOA lab at MOA as well as Mary Vickers, Alexandra Tucker, Kelsey Timler, Kav Golka, Sara Wilson, Kathryn Youngs, Abby Manuel, Brett Patychuk, Cora denHartigh, Chenise McClarty, and other LOA Lab night volunteers for all their help.  Tremendous thanks to Susan Crockford and Becky Wigen for vital assistance with fish/bird/ mammal bone identification at the University of Victoria Zooarchaeology Lab and for their many years of mentorship. Thanks to Ann Stahl, Quentin Mackie, and Becky for enabling department access and to Yin Lam, Dorothy Wong, Daryl Fedje, and Quentin for hosting me (often w Nicole/Flora) in Victoria.   The opportunity to conduct research in the Broken Group Islands was made possible by the support of the Tseshaht First Nation and Pacific Rim National Park Reserve of Canada as well as Parks Canada’s Western and Northern Service Centre. Further support was provided by the Nuu-chah-nulth Tribal Council, BC Hydro, the Laboratory of Archaeology at UBC, Denis St. Claire, and Audrey Dallimore at Royal Roads University. The Broken Group research was conducted under Parks Canada Agency Research and Collection Permits #PRN-2008-1579 and 2009-2737. Research at Huu7ii was supported by the Huu-ay-aht First Nation with in-kind support from SFU and the University of Victoria and was conducted under BC Archaeology Branch Permit 2004-162. I have been extremely fortunate to have the support of a SSHRC CGS scholarship, a Charles and Alice Borden Fellowship, and awards from the UBC Faculty of Arts to support this dissertation research.  I thank Audrey Dallimore for geological enthusiasm and capability, for enabling fieldwork in the Broken Group in 2010 and 2011, and for coordinating Malcolm Nicol and Randy Enkin’s scanning of percussion cores at the Geological Survey. Thanks to Brad Anholt and Beth Rogers at the Bamfield Marine Sciences Centre and Larry Johnson, Johnson Ginger, and Rita Johnson at Huu-ay-aht First Nations and as well as numerous BMSC staff for the opportunity to teach Coastal Field Archaeology. Phenomenal thanks to Jane and Ewen McKechnie and to Steve and Kay Weisman who enabled this.  I am thankful to the many other UBC anthropology people such as Sue Rowley, Charles Menzies, Pat Moore, Brian Chisholm, Leslie Robertson, Julie Cruikshank, RG Matson, Mike Richards, Eleanore Asuncion, Kyla Hicks, Joyce Ma, Vinay Kamat, Peter Johansen, and William McKellin for various important conversations, opportunities, and acts of kindness. I particularly thank Irena at the MOA Café and Rima Wilkes, Jennifer Kramer and Pat for the many lunches as well as Charles and the Anth Running Club. Further thanks to Hisham Zerriffi, Cornelia Scheffler, Chris Arnett, Nigel Haggan, Jasmine Johnstone, Molly Malone, Duncan McLaren, Anne Salomon, Bill Angelbeck, Gwaliga Hart, Lee Brown, Russ Markel, Jessie Morin, Marina La Salle, Rich Hutchings, Gerald Singh, Margot-Hessing Lewis, Ale Diaz, Sue Formosa, Steve Daniel, Ed Gregr, Jerry Maedel, Tony Pitcher, and Daniel Pauly. Thanks to the department for hosting inspiring visiting speakers and for sushi & coffee.   Blake Edgar and Greta Lindquist at UC Press and Roy Carlson at SFU Archaeology Press kindly provided permission to include Chapters 2 and 5 in this dissertation. Grant Keddie at the Royal BC Museum, Tanya Anderson at the Canadian Museum of Civilization, and Jessica Desany Ganong at the Peabody Museum of Archaeology and Ethnology kindly provided image permissions.  Finally, I thank my parents Jane and Ewen McKechnie for priceless grandparenting, Nicole’s parents, Ron and Pat Smith for sharing many meals, editorial advice, and help with English as a first language, and to Linda and Bill Murdoch for hosting me from the island my first term in Vancouver. Nicole is an award-deserving contributor to this dissertation. her positive editing and countless moments of support have made this dissertation entirely possible. I thank you and Flora immensely for enduring far too much archaeology talk and for completing my contemporary world in our mutuality of being. – xviii –  Dedication  This dissertation is dedicated to Flora Frances McKechnie and Nicole Fenwick Smith.   – 1 –  Chapter 1. Archaeological Scales of Analysis on the Northwest Coast  “Archaeological synthesizing on the Northwest Coast and elsewhere is always a process in which we move back and forth between scales of analysis. The ecological foundation for patterns found through analysis of a single site or sites occupied by a local group is often lost when ‘scaling up.”  Madonna Moss (2012:16).  Introduction This dissertation examines multiple scales of Indigenous Northwest Coast history from the disciplinary perspective of archaeology. I investigate social life materially represented in subsistence and settlement data to examine temporal and geographic scales of historical variability over the mid-to-late-Holocene (ca. 5,000-200 BP).  Scholars have long argued that the Northwest Coast is not a homogenous place but exhibits cultural and historical variation that is, in part, influenced by environmental and ecological differences (Boas 1887; Suttles 1987). These factors have long been central to archaeological investigations of human history on the coast but the temporal and spatial scale of this variation remains unknown, yet must have major consequences for archaeological interpretation (Ames 1991; Cannon 2001; Moss 2012). Confronting this interpretive challenge requires moving beyond the archaeological tendency to generalize from a single case or a small set of cases to examine many examples at a variety of scales and evaluate how robustly aspects of human and environmental history are archaeologically supported and conversely, which aspects of archaeological interpretation are in need of revision. As stated by Madonna Moss in this chapter’s epigram, a central interpretive challenge is the issue of scale. As a conceptual heuristic, scale is both a virtue and a constraint in the archaeological – 2 –  interpretation of Indigenous histories on the Northwest Coast. Scale is virtuous because archaeological data are effective at illuminating the enduring presence of Indigenous People occupying vast landscapes over hundreds of generations throughout the coast. However, a spatially and temporally coarse scale also severely constrains the analytical and interpretive resolution of the conventional archaeological record (Bailey 2007) and correspondingly, can contribute to a lack of richness and dynamism in Indigenous history when this is often a result of inadequate archaeological methods and a lack of preservation. Applying a large-scale archaeological perspective also has the particularly negative consequence of homogenizing a host of potentially significant aspects of cultural, historical, and environmental variation. This analytically erases the possibility of identifying archaeological patterning and social practices as expressed at smaller scales.  This dissertation confronts the issue of scale across six chapters within the unifying theme of – the everyday over millennia. This theme focuses on two fundamental domains of human existence that have a pervasive archaeological expression: food and settlement. My methodology focuses on the everyday over millennia through six case studies that variously draw on zooarchaeological, settlement, and radiocarbon data from multiple localities and time periods on the Northwest Coast. These linked case studies query the strength of archaeological patterning at multiple spatial and temporal scales and demonstrate the interpretive utility of contextualizing archaeological history alongside the intergenerational history represented in Indigenous oral historical narratives and place names.  Theoretical Framework The theoretical framework I employ to examine the everyday over millennia is inspired by the historical approach of Braudel (1970; 1980, 1981), particularly the Annaliste – 3 –  concept of the longue durée, which is the broad centennial and millennial scale of history beyond ‘the history of events.’ This broad scale and inclusive conception of historical analysis is applicable to the narration of archaeological history in part because:  1) Archaeological data are exceptional for their encompassing representation of time as well as a range of temporal scales. This ‘depth’ ensures that a broad assemblage of items and human circumstances are encountered in archaeological sequences. 2) Archaeological data often represent the material legacy of a multitude of people and this collective representation (at a community scale over long periods of time) constitutes a rich representation of human history. 3) Archaeologists intuitively recognize that many aspects of the archaeological record produce strong patterning at millennial scales but often lack a nontrivial explanation to account for such patterning. 4) Adopting a broad spatial and temporal scale facilitates the exploration and recognition of smaller-scale variability and encourages the comparison of phenomena across multiple scales to account for conjunctures and/or disjunctures in historical phenomena. Archaeologists in western North America periodically invoke Braudel and the Annaliste concept of the longue durée (Ames 1991; Cannon 1996; Hull 2005) but also recognize that emphasizing patterning at a millennial-scale can be simultaneously amorphous, homogenizing, and under-specific (Silliman 2012). Archaeologists have also commonly evoked a similarly amorphous term of ‘long-term history’ (Lightfoot 1993; Mitchell and Scheiber 2010), which can also underspecify why or how certain aspects of the archaeological record persist over millennial-scale time frames. Such concepts remain useful descriptive heuristics under which archaeological data exhibit patterning but often lack detailed exploration and explanation. On the Northwest Coast, those who focus on millennial-scale history do not feel compelled to adopt Braudel’s historical scales, but rather emphasize the Indigenous robusticity and stability of cultural institutions (Cannon 2002, – 4 –  2011a; Martindale and Letham 2011). This approach considers the agency of Indigenous peoples and their ritual cosmology as an explanatory factor in the persistence of certain archaeological patterns rather than the more common characterization of Northwest Coast societies as shaped and “subject to biological and environmental constraint[s]” (Cannon 2011a:57).  This dissertation engages with millennial-scale and centennial-scale patterning and variability in Northwest Coast archaeology. I draw on Braudel’s (1970; 1981) concepts of the longue durée and conjunctures (medium-term change over human lifetimes) but do so in combination with Bourdieu’s concept of practice (1977). I combine these concepts to investigate how the habitus (i.e. culturally situated orientations or dispositions) of the ‘everyday’ is expressed as a persistent practice unfolding over human generations and over millennia of human history. As stated by Ortner (2005:33), habitus emerges from practice and represents “a system of dispositions that incline actors to act, think, and feel in ways consistent with the limits of the structure.” This structure, rooted in history, is inherently cultural and: “Culture is a product of acting social beings trying to make sense of the world in which they find themselves, and if we are to make sense of a culture, we must situate ourselves in the position from which it was constructed”  (Ortner 1984:130, emphasis in the original).  Similarly, in this dissertation I attempt to identify the practice-based outcome of everyday life as represented in the fine-scale archaeological record of food and settlement. Like Ortner’s subjectivity-oriented view of culture, I seek to adopt the perspective “from which it was constructed” by applying it to interpreting the structured depositional practices of the – 5 –  archaeological record of food and settlement (cf. Garrow 2012; Thomas 2012). I argue that regular and individually practiced elements of everyday life over millennia can be recognized in archaeological deposits when examined at a fine depositional scale and collectively across millennia and broad geographic regions. Aspects of the longue durée are archaeologically and anthropologically significant as they are reflective of actions expressed on the highly regular timescale of the ‘everyday’ which nevertheless durably persist over the ‘long-term.’ Everyday life is an important but understudied domain of history which “consists of the little things one hardly notices in time and space” such as “the ways people eat, dress, or lodge” (Braudel 1981:29). Such durable practices are often under-recognized or remain un-questioned (Giddens 1984:41) but importantly transcend the “event-centered” focus of conventional historical discourse (Fogelson 1989). This under-recognition extends to archaeological interpretation wherein everyday quotidian practices are often considered historically inconsequential relative to ritual and symbolic practices. In the words of Garrow (2012:85, emphasis added), such “material signatures of ‘everyday’ practice have been under-theorized and all too often ignored” and yet it is “vital to consider the ‘everyday’ as well as the ‘ritual’ processes which lie behind the patterns we uncover in the ground.” Thus, certain pervasive forms of archaeological data are under-considered given their interpretive relevance to understanding the ‘everyday’ and such a lack of focus inhibits exploration of patterning in such data at a finer scale as well as the integration of such data across larger scales of history.  Bourdieu’s (1977) concept of practice, as an iterative unfolding of culturally constructed subjectivities and dispositions (Ortner 2005), provides a direct causal linkage between the everyday and the longue durée. That is, regularized everyday practices exhibit – 6 –  remarkable persistence across the vast timescales of the longue durée, even amidst change in their form and expression, because macro-scale patterns are the collective product of everyday actions and not vice versa. This is particularly applicable to those intergenerational secular rituals of food and settlement that curate and recreate ‘traditional’ practices, the material signatures of which are richly represented in archaeology. Thus, an archaeological analysis of the longue durée has the potential to highlight the power of individuals as causal agents in the construction of history. At the same time, given the volatility of individual agency, evidence of millennia-long stability in everyday practices speaks to the power of cultural traditions and the conventions that foster such continuity.  Food and Place as Multi-scalar  This dissertation examines two forms of archaeological data that are fundamentally regular aspects of everyday life, that stretch across human generations and are pervasive aspects of the archaeological record on the Northwest Coast: 1) food and 2) the places people live. The animals people eat and the places their remains are deposited offer particular insight into the quotidian daily practices of ancient communities as we can be confident that animal and/or animal products are eaten on a daily basis and that consumption most often occurs in settlements where people spend a great deal of time (homes). Evidence of food consumption and deposition at specific places has enduring interpretive value in archaeology precisely because such data are the result of everyday practice (Atalay and Hastorf 2006; Gifford-Gonzalez 2008). Moreover, because consumption likely occurs on such a fine temporal scale, archaeological evidence of food can be examined at a large variety of scales: from a handful of sediment deposited in a day or a week to an entire midden deposited over centuries or millennia. While there is an unknown degree of temporal averaging and taphonomic bias – 7 –  built into such a multi-scalar comparison (Bailey 2007; Kowalewski, et al. 1998; Lyman 2003a), observations of the repetitive occurrence and stable proportionality for certain foods in archaeological deposits (aka. structured deposition) represents a strong archaeological pattern and creates a basis for inferring the consistent importance of certain foods and thus a continuity in practice over generations.  Food Deposition as Social Practice This dissertation focuses on the identification of such collective depositional patterns and argues that a focus on food at particular places provides insight into how “one is disposed to do and to be in certain ways because of one’s experiences in social settings” (Pauketat 2001:4). I argue that such a simple insight – the repetitive continuity of a cultural practice – is often lost amid the more common tendency to emphasize dynamic historical change, the symbolic dimensions of practices, and especially to avoid the equally hazardous narrative trope of an unchanging Indigenous history. A contrary perspective on food is that it can be an abstraction that may have little or no consequence or relationship to major historical and social transformations. In other words, a tsunami of social change may have just occurred but breakfast may remain the same for the next hundred years. However, it is also the case that breakfast wouldn’t be possible without that skilled but often mundane activity of collecting and preparing food on which those ‘higher-level’ historical dynamics unfold. While this is not to suggest, for instance, that a history of medieval Europe can be deduced from an analysis of farming practices alone, but rather, that such a history cannot be holistically achieved without attentiveness to everyday activities on a broad spatial and temporal scale (Braudel 1981). It is additionally noteworthy – 8 –  that large-scale social transformations often begin from everyday circumstances that become magnified through “the amplification of minor differences” consequentially turning “small issues into big events,” and thus “how microhistories become macrohistories and vice versa” (Sahlins 2013:161). Such disruptions can alter the ways people and communities relate to one another and this correspondingly has the potential to be reflected in the proportion and intensity of food being collected, processed, consumed and deposited into the archaeological record.  Food pervades most social relations due to its regularity of consumption and its often fundamental centrality in social and ceremonial gatherings (Dietler and Hayden 2001). Food has human and historical agency as it both represents a culturally unique transformation of various animals and plants into a category of ‘food’ and in so doing, both reflects and facilitates the perpetuation of a “specifically cultural and historical consciousness” (Ortner 2005:34). As in Bourdieu’s habitus, such a consciousness does not structurally determine a given future, but it likely informs, influences, and guides practices into the future. Put more simply, such a regular aspect of everyday life also defines what is considered ‘acceptable’ in a given setting, or conversely, what is not considered acceptable1. Such a formulation is equally applicable to how a person ‘goes about living’ in a place and what and how one prepares and consumes food and disposes of waste.  Animals consumed as human food are often themselves considered to have levels of agency in that they require skill and attentiveness to catch or harvest, prepare and/or care for                                                 1 For instance, in Nuu-chah-nulth territory on western Vancouver Island, an ethnographic account states that a “common punishment for eating alone was to have one’s mouth ‘stretched’ by two husky individuals, one of whom hooked a finger in either corner of the mouth and gave a yank. Food boxes and dishes of the offender would be smashed up, contents and all” (Drucker 1951:388) – 9 –  (e.g., Ingold 1980; Russell 2012). On the Northwest Coast, animals are often associated with a human-like animism which emerges from specific social and cosmological histories with people (e.g., Drucker 1951:151; Hill 2012; Losey 2010). Animal products and products made from animals also obtain a spiritual significance and serve a host of other practical and secondary uses (e.g., Hodgetts and Rahemtulla 2001). The disproportionate abundance of certain foods combined with exceptional preservation conditions further structures archaeological perspectives on food and food waste. The presence of these foods represents actions of processing and provisioning and involves connections between people and landscapes and of people to each other. It is through such quotidian, under-recognized routines, that communities are commonly defined and persist through time. Place and the Practice of Settlement An even more elemental aspect to the human history of everyday life concerns place. As stated by Casey (1993:21) “there is no (grasping of) time without place” and thus, place is a primary category and a fundamentally ‘prior’ condition. Places emerge over the long-term through routinized practices in a “landscape of habit” that is both consciously and non-discursively built upon the “well-trodden (or well-paddled) paths” laid out by one’s predecessors (Mackie 2001:25). Places are central to the expansive history of the longue durée as articulated by Braudel and the Annaliste school of historians. Places are not simply a backdrop to the arc of human history but act as “a major continuous constraint on human activity” (Thornton 2008:20).  Archaeological sites, such as shell midden settlements on the Northwest Coast, are particularly significant as human places, as they are constructed localities where groups of – 10 –  people have “invest[ed] themselves into their landscape” (Momaday 1974:80) and literally have mixed “their labour with the earth” (Williams 2005:76). Such physically altered places acquire a broader social and historical significance that can come to signify “points in the geography of a community where time and space intersect and fuse” (Bakhtin 1981:84).  Similarly, the surrounding landscape in which archaeological sites develop and in which human histories unfold, both shape people’s history and are correspondingly shaped by people over time (Ingold 1993). In this fashion, human places and landscapes can be said to be ongoing co-productions, reflecting the regular actions of people and natural processes alongside more periodic but larger-scale transformations. Under these conditions, an attempt “to separate natural history from social history becomes extremely problematic” as “the idea of nature contains, though often unnoticed, an extraordinary amount of human history" (Williams 2005:67-76). For the purposes of this dissertation, shell midden deposits are considered human places or settlements. I consider shell middens to be physically created places where people spent a great deal of time, and whose everyday practices have cumulatively produced a physical assemblage of material spanning generations of persistent human activity (Blukis Onat 1985; Cannon 2003). I recognize that shell middens are only one of a large variety of other ‘types’ of sites on the Northwest Coast and comprise a relatively narrow aspect in what was undoubtedly a much more widely used coastal and inland landscape (e.g., Oliver 2007). Moreover, materials preserved in these deposits do not encompass the full spectrum of human activities that took place there (e.g., Croes 2003) and neither do the materials represented reflect purely incidental utilitarian deposition or an exclusively ritualized deposition (cf. Klokler 2008; Luby and Gruber 1999; Luby, et al. 2006; McNiven 2013; – 11 –  Villagran, et al. 2011). However, shell middens deposits do contain an immense amount of material evidence of past human activity (Moss 2011a) and persist as settled places because these built environments act to structure future deposition (Mackie 2001:62). Moreover, because shell middens are observed to be localities where people spend a great deal of time, it can be reliably assumed that many of the materials deposited in shell midden sites are related to the everyday human activity that occurred there. As physical deposits, shell middens are places where “time takes on flesh and becomes visible for human contemplation” and thus may serve as “monuments to the community itself” (Bakhtin 1981:84).  Dissertation Structure This dissertation is divided into two parts consisting of six individual chapter contributions as well as an introduction and conclusion. The first half of the dissertation examines Indigenous subsistence and settlement histories as archaeologically expressed on large, medium, and small geographic scales on the Northwest Coast (Chapters 2, 3, and 4). The second half of the dissertation consists of three chapters that focus on archaeological data specific to Barkley Sound on the southwest coast of Vancouver Island (Chapters 5, 6, and 7). These multiple temporal and spatial scales and mixture of regional and local case studies (Figure 1.1 and Figure 1.2) focus on foodways and settlement practices and indicate that Indigenous history on the Northwest Coast exhibits remarkable continuity at the scale of the everyday.  – 12 –   Figure 1.1 Map of the Northwest Coast showing geographic scales of analysis that are the focus for chapters in this dissertation. – 13 –   Figure 1.2 Temporal scales of analysis that are the focus for chapters in this dissertation. Vertical scale approximates north-south gradient.  Regional Scales of Human and Ecological Histories Chapter 2 adopts a historical ecological approach (e.g., Jackson, et al. 2001) to investigate the archaeology of Indigenous herring fisheries and ecological baselines within much of the Northwest Coast as a singular region over the Holocene. I, along with several co-authors, demonstrate that Pacific herring (Clupea pallasii), have an under-recognized importance in the history of Indigenous fisheries on the Northwest Coast. We compile and integrate a wealth of formerly scattered black and grey-literature data on fine screen fish remains (171 sites) and identify a pervasive coast-wide pattern with a high archaeological importance of herring relative to all other fish species. We observe herring to have a more significant role in Indigenous subsistence economies than has been previously appreciated and further demonstrate that herring exhibit strong spatial variability in abundance across this large region. We explore variability at the finest archaeologically reported scale within 50 individual sites across the coast and argue this indicates herring use and herring habitat was – 14 –  broadly consistent over time and within regions and micro-regions. We further hypothesize that pre-industrial herring population variability differs from modern variability that characterizes industrial fisheries and advocate that research at an archaeologically finer temporal scale is needed to qualify the strength of these observations.  Chapter 3 narrows in geographic scope and shifts from Indigenous fisheries to Indigenous hunting traditions on the southern British Columbia coast, from the Strait of Georgia in the Salish Sea around to western Vancouver Island (Figure 1.1). Along with co-author Rebecca Wigen, I examine the spatial and temporal coherence of mammalian hunting practices as represented in a compilation of existing zooarchaeological data within the linguistically distinct Nuu-chah-nulth and Coast Salish culture areas. These regions have long been recognized as distinct by Indigenous Peoples and anthropological researchers (e.g., Boas 1887), but this study compiles a wealth of existing zooarchaeological data to further demonstrate how this distinctiveness is archaeologically represented in hunting practice. Thus, within each of these two regions, we observe an unexpected degree of geographical and temporal coherence in the proportionality of zooarchaeological data. These results yield further insight into the ecological contribution that human hunting had on the marine and terrestrial environments over the Holocene and offer an improved basis for interpreting how these ‘traditions’ and/or ‘cultural adaptations’ contributed to and/or complemented other domains of human subsistence and ritual practice (e.g., burning, plant harvesting, spiritual training and preparation). Although these hunting practices have an obvious environmental distribution (outer coast versus an inland sea), this by no means explains the cultural persistence of these practices which are direct products of repetitive, culturally embedded, – 15 –  intentional social action (e.g., Ingold 2000) operating over an intergenerational time scales and at the scale of the everyday or at least every year.  Chapter 4 refines the temporal and spatial scale of analysis to examine settlement history in the heavily investigated area of Prince Rupert Harbour. I explore the issue of radiocarbon calibration and the chronological resolution of archaeological phenomena in marine settings with co-author Morley Eldridge. We employ available marine reservoir data from the Dundas Islands and Prince Rupert Harbour to re-evaluate the archaeological settlement dynamics of two long-standing radiocarbon chronologies that have played a significant interpretive role in Northwest Coast archaeology, particularly concerning the antiquity of Indigenous oral history and regional conflict (Cybulski 1999; Martindale and Marsden 2003). This re-analysis narrows and affirms the previous identification of this medium-scale temporal patterning in Coast Tsimshian settlement dynamics and burial practices. Our re-analysis advocates for continued improvements in the use of calibration techniques for marine and marine influenced dates in order to clarify and strengthen previously published chronologies by re-engaging with the cumulative legacy of existing data. We conclude by emphasizing the value of acknowledging ambiguities and contingencies in archaeological claims to knowledge (Gero 2007; Martindale n.d.), which can limit archaeologists’ ability to resolve the temporal scale in key periods in human history. Case Studies of Indigenous History in Barkley Sound The second half of the dissertation examines the archaeological history of subsistence and settlement in two different First Nation territories in Barkley Sound on Southwest – 16 –  Vancouver Island and demonstrate how these reflect both an enduring cultural continuity at a broad scale and provide vital insight into intermediate timescales including periods of dramatic historical change in settlement and more subtle change in foodways. Chapter 5 examines fine-scale coherence and variability in archaeological fisheries data from a single Nuu-chah-nulth ‘big’ house. I explore the depositional patterning in these everyday foods at multiple depositional and spatial scales within this very large house (35 x 17 m, or 115 x 50 ft) in the centre of a village in the named origin place of the Huu-ay-aht First Nation (McMillan and St. Claire 2012). Two additional column sample time series dating to the mid-Holocene (ca. 5000 to 3000 years BP) and associated with a period of higher sea levels, further contextualize these later household associated data. This chapter also evaluates the interpretive utility of converting the number of fish bones (NISP) into numbers of fish (MNI) and extrapolates these in terms of fish per cubic metre. I also examine temporal trends in the use of key species (hake, salmon, rockfish) that indicate coherent shifts in the marine climate during a period of when there is a regional increase in dated defensive sites on western Vancouver Island (ca. 800-500 years BP). This study demonstrates the relevance of exploring subsistence and climatic variability in a single large house over a 1000 year period. Zooming out to a multi-site scale but remaining in Barkley Sound, Chapter 6 examines the settlement history in an archipelago identified in numerous Indigenous oral historical accounts as the territory of an autonomous Nuu-chah-nulth polity (the Maktl7ii7ath local group) in a portion of the Broken Group Islands (Sapir and Swadesh 1955; St. Claire 1991). This group of islands has a series of archaeologically anchored Indigenous place names and an oral historical record describing it as the defended year-round territory of the – 17 –  Maktl7ii7ath prior to the contact era (St. Claire 1991). I examine the archaeological evidence of settlement in this portion of the archipelago and evaluate how this corresponds to oral historically documented social history. I employ multiple measures (e.g., midden surface areas, radiocarbon dates) to archaeologically characterize this human settlement history. Collectively, these observations demonstrate these islands were enduringly occupied over multiple generations, consistent with the Nuu-chah-nulth concept of ‘home’ or ‘chiefly territory’ (hahuulhi). This dynamic settlement chronology, reveals considerable shifts exemplified in the archaeologically ‘sudden’ occupation of the elevated fortress of Huts’atswilh. This micro-regional settlement history is an archaeological example of the intermediate scale of intergenerational history akin to Braudel’s conjunctures and that transcends ‘event-centred’ history (cf. Braudel 1970; Fogelson 1989). Chapter 7 builds from the previous chapter but focuses on the fundamentally regular activity of collecting and consuming food as a frame for contextualizing the everyday over millennia. Centering on zooarchaeological data from numerous auger samples collected from two settlements on Dicebox Island as well as associated villages on adjacent islands, this chapter details how archaeological evidence of ubiquitous foods are represented at multiple temporal and spatial scales over the last 2,500 years. This includes data from the territory as a whole between sites, and within individual levels at a single site. I demonstrate that certain foods constitute a consistent majority of the repetitively deposited items. I then evaluate if the relative proportion and/or density (bones per litre) of these foods persist over time, and explore the ecological factors by which they may differ spatially and/or temporally. I relate these to community processes identified in the settlement organization explored in the previous chapter.  – 18 –  Conclusion  These linked multi-scalar studies examine the most regularly practiced aspects of daily life at a variety of scales on the Northwest Coast. I identify the outcome of traditional foodways as represented in the archaeological record both over long temporal scales but also at small depositional scales (e.g., handfuls of dirt). The results provide evidence of both historical continuity and considerable stability of practice. I interpret this as evidence of both a thoughtful cultural adaptation to the circumstance of people’s lives in these environments and as evidence of an extraordinary effort at maintaining historical continuity via a construct of tradition. Thus, the stability of the longue durée is achieved not as a fixed cultural entity but is iteratively reproduced within a cultured continuity of everyday practice. This approach seeks to confront the analytical tendency to consider certain aspects of the archaeological record to be static, essentialized and normative. Rather, I seek to demonstrate how archaeological attentiveness to identifying regular and structured depositional practices in a community setting and its variation holds potential for enriching archaeological understandings of the everyday practices of Indigenous peoples. Finally, I examine how both the Indigenous oral historical record and archaeological chronology of settlement enrich archaeological interpretations of continuity and change in settlement practices on the Northwest Coast.    – 19 –  Chapter 2. Archaeological Data Provide Alternative Hypotheses on Pacific Herring (Clupea pallasii) Distribution, Abundance, and Variability on the Northwest Coast2 Introduction Low trophic-level fish (“forage fish”) are experiencing global declines with increasing recognition of widespread and cumulative ecological, cultural and economic impacts (Pikitch, et al. 2012; Pinsky, et al. 2011; Smith, ADM, et al. 2011). Both Indigenous and non-Indigenous peoples on the Northwest Coast of North America recognize Pacific herring (Clupea pallasii) as a foundation species in coastal food webs (Menge, et al. 2013) that serve an essential role in maintaining social–ecological systems (e.g., Garibaldi and Turner 2004; Platten and Henfrey 2009). Herring and its roe are critical prey for a host of fish (e.g., hake, Pacific cod, dogfish, salmon), bird and marine mammal predators (Robinson and Ware 1999; Therriault, et al. 2009; Ware and Schweigert 2002). Herring is also central to the social, cultural, and economic relations of coastal indigenous communities, many of which seek to continue their traditional fisheries for herring and herring roe on kelp or other substrates (Harris 2000; Thornton, et al. 2010a; Thornton, et al. 2010b). Since 1882 and continuing into recent decades, industrial fishing of herring has helped support many communities across the Northwest Coast (Funk 2010; Tester 1935).  Populations of this once highly abundant forage fish have been dramatically reduced across much of its North Pacific range relative to levels seen in the mid 20th century (Funk                                                 2 This multi-authored chapter was submitted for consideration for publication on September 16, 2013 and accepted for publication on December 12, 2013 (see preface for more information). Due to length, the tables associated this chapter are presented in the appendix to this dissertation.  – 20 –  2010; Schweigert and Linekin 1990; Schweigert, et al. 2010). Fisheries scientists have proposed various factors accounting for these declines and sustained low abundances even after reductions of fishing pressure. These include climate-induced ecological changes in distribution of predators and prey (Ware and Schweigert 2002), disease (Marty, et al. 2010), over-fishing (Pauly, et al. 2001), and the rebound of marine mammal populations that prey on herring (Schweigert, et al. 2010). Assessing these potential drivers and moving forward with conservation requires baseline information on herring abundance and distribution prior to its depletion. Yet current knowledge of Pacific herring distribution and abundance is based on biomass estimates that date back only to the mid 20th century – well after the onset of industrial fishing.  Most modern ecological data lack sufficient time depth for establishing baselines for marine ecosystem management and/or recovery (McClenachan, et al. 2012); this has the potential to dramatically underestimate the degree of population loss (the ‘shifting baseline’ problem [Pauly 1995]) and inhibit recovery efforts. As a result, traditional and local ecological knowledge (TEK/LEK) as well as paleoecological and archaeological data are increasingly important for informing these baselines (Rick and Lockwood 2013). Such data provide both the long-term perspective needed to assess pre-industrial ecological states, and ecologically and culturally salient baselines for conservation (Jackson, et al. 2011; Turner, et al. 2008; Wolverton and Lyman 2012).  Here, we compile the archaeological record of fisheries in the Northeast Pacific from Alaska, British Columbia, and Washington (Figure 2.2) to assess the spatial and temporal distribution of pre-industrial herring abundance. The abundance and spatial distribution of archaeological fish bones reveals the widespread importance of herring to indigenous peoples – 21 –  throughout the region, and indicates the abundance of herring in coastal ecosystems over the past several thousand years. Herring were highly abundant along southwest Vancouver Island and in the Salish Sea. The archaeological record indicates that places with abundant herring were consistently harvested over time, and suggests that the areas where herring massed or spawned were more extensive and less variable in the past than today. Although archaeological data are expressed over different spatial and temporal scales than modern fisheries population estimates, archaeological data highlight both a disjunction between pre-industrial and contemporary herring abundances and distribution, and the need to revise and expand the data on which current fisheries management and conservation are based. History of Herring and Herring Roe Fishing For many First Nations and Native American groups from Alaska to Washington, the nutritionally valuable and readily harvested herring and its roe were integral to daily lives and worldviews (e.g., Boas 1932; Bouchard and Kennedy 1989; Brown and Brown 2009). This is reflected in photographs, interviews, oral histories, and indigenous place names (e.g., Teeshoshum, “milky waters from herring spawn;” K’i:na?a, “herring guts on rocks” (Sapir and Swadesh 1955:35); Yaaw Teiyi “Herring Rock,” the sacred place where the first herring arrived; Shaan Daa “White Island,” also known as “Fish Egg Island”, named for the whiteness created by the spawning activity each spring; and “Silver Bay”, because in the winter, there were so many herring “if you looked at it in the moonlight you’d see the backs of the herring... and it would look silver” (Thornton, et al. 2010a:60, 334, 489). Many coastal groups maintained family-owned locations for harvesting herring and herring roe from anchored kelp fronds, eelgrass, or boughs of hemlock or cedar trees (Drucker 1951; Powell – 22 –  2012). Herring was harvested at other times of the year than the spawning period when massing in local waters (Grant 1857:300; McKechnie 2005b:103; Moss, et al. 2011; Sapir and Swadesh 1939:223) but most ethnohistorical observations identify late winter and springtime spawning as a key period of harvest for both roe and fish. Processed herring and roe were consumed in large quantities and traded widely among coastal First Nations (Bouchard and Kennedy 1989:8; Jewitt 1807:6). Sustainable harvests were encouraged by building kelp gardens (Brown and Brown 2009:xviii), wherein some roe-covered fronds were not collected, by minimizing noise and movement during spawning events, and by elaborate systems of kin-based rights and responsibilities that regulated herring use and distribution (Bouchard and Kennedy 1989; Powell 2012; Sproat 1868:224; Thornton, et al. 2010a; Thornton, et al. 2010b). Industrial commercial fishing of herring began in earnest on the Northwest Coast in the late 19th century primarily using beach and drag seines to catch fish and then render (“reduce”) oil and meal (Stick and Lindquist 2009; Taylor 1955; Tester 1935; Thornton, et al. 2010a). In British Columbia, officials prohibited the reduction of herring into oil and fertilizer in 1910 (Carrothers 1941:111) and noted that the larger bays were “being gradually deserted by the larger schools where they were formerly easily obtained” (Tester 1935:7). In 1927 (23 years prior to the current management baseline for estimating historical biomass), the fishery on eastern Vancouver Island, British Columbia, processed 31,103 metric tons of herring (Munro and Clemens 1931:5), which is roughly two times the annual harvest rate in 2012 and approximately 38% of the current biomass estimate for the entire Strait of Georgia (Department of Fisheries and Oceans Canada 2012b; 2013:43). In Alaska, the herring reduction industry began in 1882, with state-wide harvests reaching a peak of 75,000 metric – 23 –  tons in 1929 (Funk 2010:257). By the 1930s, the effects on population numbers and/or behaviour caused fisheries scientists to express deep concern about the effects of over-fishing on herring (Carrothers 1941:111; Taylor 1955:110; Tester 1933:289), as the fishery shifted to deeper water harvest technology (Tester 1935:7) and underwent a considerable spatial expansion in search effort (Taylor 1955:110).  By the late 1960s, as a result of increased fleet efficiency, reduction fisheries, and poor recruitment, the herring populations of British Columbia and Washington collapsed (Outram and Humphreys 1974). This led Canada to permanently close its reduction fishery in 1968 (Pearse 1982:15), followed by the closure in Washington state in the early 1980s (Stick and Lindquist 2009). In the early 1970s, the herring industry in Japan collapsed (Morita 1985), which increased East Asian demand for herring eggs taken from North American waters. As a result, a Sac Roe Fishery (SRF) began throughout the Northeast Pacific, targeting female herring just prior to spawning. In the last decade, coast-wide declines in herring numbers (Cleary, et al. 2010; Stick and Lindquist 2009) have led to a greatly reduced SRF, now limited to a few regions such as the Strait of Georgia (Salish Sea) and Sitka and Togiak, Alaska (Funk 2010). In addition to these larger ventures, a relatively small food and bait fishery has persisted since the early 20th century. Since the 1970s licenses and legal judgments have been issued to First Nations in Canada (Newell 1999; Powell 2012), Native Americans in Washington (Boxberger 1999), and Alaskan Natives (Thornton, et al. 2010a) that support food, social, and ceremonial fisheries and in some cases commercial fishing. Government fishery managers, scientists, and local and indigenous peoples lack consensus on the cumulative consequences of ongoing commercial fisheries on herring populations. Many First Nations, Native Americans, Alaska Natives and other local fishers, – 24 –  based on personal observations and traditional knowledge, hypothesize that local herring stocks, on which they consistently relied for generations, have been dramatically reduced and made more difficult to access following 20th century industrial fishing (Haida Marine Traditional Knowledge Study Participants, et al. 2011; Hebert and Thornton 2010; Thornton, et al. 2010a; Thornton, et al. 2010b). In contrast, fisheries managers (Hay, et al. 2008; Schweigert, et al. 2010) identify commercial fishing as only one of several potential causes for the coast-wide decline in herring and/or persistent lack of recovery since implementing conservation measures (Hay, et al. 2009). In the Strait of Georgia (BC), managers hypothesize that populations have in fact not been depleted, but rather have shifted spatially due to climatic factors and predator abundance (Hay, et al. 2009; Ware and Schweigert 2002). The unresolved nature of these alternative hypotheses regarding the primary factors responsible for temporal and spatial shifts in herring populations represent a barrier to achieving consensus on the need and strategies for improving herring conservation and management. Zooarchaeological data offer a record of the pre-industrial abundances and distribution of herring, providing a longer-term perspective that can illuminate and contextualize these debates.   Herring and the Archaeological Record  On the Northwest Coast, stratified, shell-bearing archaeological sites (shell middens) provide long-term records of human-animal interactions. Although animal bones can enter archaeological sites through a range of cultural and non-cultural processes (Erlandson and Moss 2001), the majority of fish bones in coastal middens are associated with human processing or consumption of fish. Even though animal products were widely traded in the – 25 –  past, the bulk of zooarchaeological remains in shell midden deposits tend to be composed primarily of resources harvested nearby (Cannon 2000a; Lepofsky, et al. 2007; Szpak, et al. 2013). Thus, in most cases, zooarchaeological remains, including marine fish bones, can be used as proxies for local distribution and abundance. This synthesis of zooarchaeological fisheries records builds on the increasing number of recent analyses employing rigorous methods of fish bone recovery and quantification, particularly the use of column sampling and fine-screen mesh (≤ 3.2 mm) that is critical to ensuring adequate proportional representation of small-bodied fish such as herring (Casteel 1972). Our estimates of herring relative abundance and rank order are based on a standard zooarchaeological measure of “number of identified specimens” (NISP), which is not equivalent to biomass or meat weight. Rather, NISP is correlated with numbers of individual animals (Grayson 1984) and thus can be used as a culturally and taphonomically filtered proxy of past, local fish populations (Lyman 2008). Most archaeological deposits, including the zooarchaeological data presented here, have relatively low chronological resolution as calibrated radiocarbon age-range estimates often span more than a century; thus, specific deposits incorporate a degree of time averaging (Kowalewski, et al. 1998; Lyman 2003a).  Despite interpretive challenges, ancient fish bone records provide long-term and spatially explicit data on past use and abundance of herring and other fish. This is particularly so in the case of our dataset, which compiles rigorously screened data from numerous sites across the Pacific Northwest Coast that represent multiple temporal and regional scales. This is the largest available dataset of fine-screened archaeological fish assemblages from the Northwest Coast and offers new insight into the taxonomic composition of indigenous fisheries. – 26 –   Figure 2.1. Percent abundance of herring bones in archaeological sites with >50 identified fish bones (N = 171 sites). Only two sites lack herring bones. Herring is abundant (>60% of total fish NISP) in sites throughout the Strait of Georgia in southern British Columbia. In 71% of sites, herring makes up at least 20% of the site’s total assemblage of fish bones.  Material and Methods3  Data Sources To examine the distribution and abundance of herring represented in archaeological sites of the Northwest Coast (SE AK, BC, WA), we compiled a database of all well-sampled sites with adequately recovered and identified fish bone assemblages located within 1 km of                                                 3 Due to length, the tables for this chapter (Chapter 2) are presented in an appendix to this dissertation. – 27 –  the current marine shoreline (Table 2.1). This entailed an extensive literature review of published and grey literature on zooarchaeological analyses completed over the past 40 years. Given the small size of herring bones (e.g., vertebral centra are 4 mm or less in diameter), we only included sites where the zooarchaeological remains were systematically recovered using a fine-screen mesh (equal to or less than 3.2 mm). All zooarchaeological remains were identified by established analysts or students working under the analysts’ direct supervision using comparative collections (Table 2.1).  Only sites containing a minimum of 50 fish bone specimens identified to at least family level were included. Over 91% of the 171 sites have more than 100 fish bones. This is a reasonable threshold for assessing the relative abundance of the three-most ubiquitous and abundant taxa at a site (Butler and Campbell 2004:340). For the site-based analyses presented in Figure 2.1 and the regional analyses presented in Figures 2.3-2.6, we combined zooarchaeological data from sites with multiple excavated sub-assemblages to ensure adequate sample sizes (Tables 2.1 and 2.2).  We used the relative abundance of herring bones among all identified fish (number of identified specimens, NISP) as a proxy for the relative abundance of herring in archaeological sites. We did not convert herring bone NISPs to estimates of ‘meat-weight’ or biomass as suitable taphonomic, allometric, and stratigraphic data are lacking for the majority of sites. We judge our assessment of the abundance of identified herring bones as a proportion of all fish bones to be conservative given that herring have fewer vertebrae than some larger fish taxa that can fracture into numerous identifiable fragments (e.g., Oncorhynchus spp., Squalus acanthias). While quantitative data on herring bone density measurements have not been conducted, their bones are notably smaller and more delicate – 28 –  relative to most other measured fish species (Smith, RE, et al. 2011a). Moreover, the cellular structure of herring bone likely makes them more susceptible to microbial degeneration relative to other fish (Szpak 2011:3367) indicating this species would likely be less well-preserved in the archaeological record. Chronological Assessments To assign age-ranges to archaeological assemblages, we used stratigraphic and chronological information provided in the original research reports and in subsequent published archaeological research (Table 2.1). The majority of sites contain at least one radiocarbon date (88%) while the remainder contain temporally diagnostic artifacts and are on landforms consistent with late Holocene sea levels (N = 21 sites). Radiocarbon age estimates were assigned age categories based on dates re-calibrated using the Intcal09 curve (Reimer, et al. 2009) and represent the 2-sigma calibrated range. Most radiocarbon samples were taken from the basal layers of individual site deposits.  Time Series Analysis To examine within-site temporal patterns at the finest possible depositional scale, we identified a sub-set of 50 sites that report chronologically distinct vertical levels or stratigraphic layers (Figures 2.6 and 2.7, Table 2.2). We included only those sites that have been dated in their basal and terminal levels either by radiocarbon dating or the presence of historic artifacts of European manufacture. Assemblages had to have 50 or more identified fish bones per vertical level or stratigraphic layer and three or more levels in sequence. To ensure each datapoint had 50 or more specimens, some assemblages from contiguous strata or levels with less than 50 identifiable specimens were combined (Table 2.2).  – 29 –  Due to differences in how assemblages were collected in situ, and the level of detail in which they are reported, we present these data in four analytically separate categories in Figure 2.5. These are; 1) assemblages collected from multiple areas within a site according to temporally distinct strata (ca. 500-2000 years per stratigraphic layer, n=14 sites); 2) sites with a single column sample of contiguous levels (N = 19 sites); 3) sites with multiple column samples with multiple contiguous levels spanning broad temporal intervals (N = 8 sites), and 4) Sites with multiple column samples with multiple contiguous levels spanning broad temporal intervals in Barkley Sound (N = 9 sites).  Comparison to Modern Spawning Records To evaluate archaeological data in relation to historic and contemporary herring spawning records from British Columbia (Figure 2.7.), we calculated the distance between our archaeological study sites and historically documented herring spawning localities (Table 2.3). We restricted this analysis to sites within areas with modern herring spawning observations as these represent a long-term ecological record of cumulative spawning activity and are derived from annual surveys of spawning length, thickness, timing and other variables collected by the Department of Fisheries and Oceans Canada (DFO) (Department of Fisheries and Oceans Canada 2012a) since 1937 (McCarter, et al. 2005). Where available, we also documented the proximity to spawning areas identified in Traditional and Local Ecological Knowledge studies in BC (Haida Marine Traditional Knowledge Study Participants, et al. 2011; Hessing-Lewis, et al. 2011). These are based on interviews with knowledgeable indigenous and non-indigenous local residents with lifetimes of perspective on the fisheries in their traditional territories. Distance to closest documented spawn was – 30 –  determined by placing geo-rectified maps of cumulative spawning localities into Google Earth and measuring the shortest route by water (± 500 m) to an archaeological site.  To assess the potential disjuncture between archaeological data and the cumulative record of historical spawning activity, we compared the six quantitative categories used by DFO (ranging from ‘vital’ to ‘minor’(McCarter, et al. 2005)) to correspondingly ranked archaeological abundance categories (e.g., 100-80% NISP = vital, 1-20% = minor). If there was more than one record of cumulative spawning within a distance of 2 km, we chose the most abundant ranking for the comparison. If the ordinal grouping in the two records differed by more than one category, we considered the two records to represent a disjuncture in abundance (Table 2.3).   Figure 2.2. Rank order of herring bones in 171 adequately sampled archaeological sites. Herring is among the top three most abundant fish taxa in 88% of the sites and is absent in only two sites. – 31 –  Results In the study area, 171 coastal archaeological sites have been sufficiently sampled to evaluate the past distribution and archaeological abundance of herring (Figure 2.1, Table 2.1). These sites span the early Holocene (10,700 cal BP) to the contact-era (AD 1740-1860) with the bulk of sites dating to within the past 2,500 years (76%). The dataset contains 435,777 fish bones confidently identified to family, genus, and/or species. These specimens represent a wide range of taxa, with each site containing a minimum of 50 identified bone specimens. Coast-wide similarities in the taxonomic richness of assemblages and relative abundance of numerically dominant fish taxa indicate that the archaeofaunal sample sizes are adequate to assess the distribution and relative abundance of herring (Table 2.1).  Herring Ubiquity, Abundance, and Rank Order Within this dataset, herring is the single-most ubiquitous fish taxon in archaeological sites throughout the Northwest Coast, occurring in 169 of the 171 assemblages (99%) representing various physiographic settings. It is also the single-most numerically abundant taxon in the dataset representing 49% of all identified fish bones, with a site average of 47% (±33% NISP). Herring is similarly the 1st ranked taxon by NISP in over half (56%) of the 171 assemblages and is among the two most numerous taxa in 80% of assemblages (Figure 2.2). Herring comprises more than 20% of fish bones in 71% of the assemblages (NISP). The high ubiquity, rank order, and relative abundance of herring in most sites is remarkable given the many other taxa present in each site (mean N of fish taxa = 10.2, SD = 3.2) and that some ubiquitous genera, such as Oncorhynchus spp. (salmon and trout) and Sebastes spp. (rockfish) comprise many species (7 and 36 respectively). The occurrence and relative abundance of herring in the majority of sites surpasses any other fish taxa, and demonstrates – 32 –  a pervasive and previously under-recognized role for this species in indigenous economies spanning the Holocene.   Figure 2.3. Boxplots showing percent abundance of herring bones in adequately sampled archaeological sites (N = 171) by major region (arranged north to south). Boxes encompass the middle 50% of cases, with the median represented as a vertical line. Whiskers encompass the 1.5 of the interquartile range, excluding outliers (represented as circles) and extreme outliers (represented as stars).   Figure 2.4. Boxplots showing percent abundance of herring bones in archaeological sites by sub-regions (arranged north to south) within the Salish Sea region (N = 78). Note the lower abundance of herring in the Fraser Delta and Puget Sound. Note: Gulf Islands includes the Canadian Gulf Islands and the US San Juan Islands. – 33 –  Spatial Distribution of Herring Geographically, nearly half the sites in the database (46%) are concentrated around the Salish Sea, reflecting more intensive archaeological investigation in this region. Only two assemblages, both in the northern portion of the study area, do not contain herring bones (Table 2.1). The occurrence of sites with significant proportions of herring throughout the study area (Figure 2.3) indicate that the spatial gaps in the distribution of analyzed sites reflect gaps in archaeological sampling, rather than gaps in the ancient distribution of herring. Thus, as a proxy for the number of fish harvested in coastal waters, these data reflect a widespread abundance of herring in marine ecosystems in the deep past.  Archaeological sites numerically dominated by herring occur throughout the study area. Sites in the Central Coast, West Coast Vancouver Island, and Salish Sea areas are consistently dominated by herring bones with median abundance values representing over 40% of the assemblage (Figure 2.3). Most sites where herring represents less than 20% of NISP (n=50) are in the northernmost portion of the study area (Haida Gwaii, northern BC Coast, and SE Alaska) or in the southernmost portion of the study area (Puget Sound and the lower mainland of British Columbia) (Figures 2.1, 2.3, and 2.4). Even in the northern portion of the study area, herring still ranks among the three most ubiquitous and abundant fish in 75% of sites (Table 2.1). The proportion of herring in Haida Gwaii exhibits high variability amongst nearby sites with salmon, rockfish, and dogfish (Squalus acanthias) representing the other dominant fish taxa. On the northern BC Coast, sites around Prince Rupert and on the Dundas Islands tend to be dominated by salmon and oil-rich smelts (Osmeridae). In southeast Alaska, salmon tends to dominate followed by herring, Pacific cod (Gadus macrocephalus), and sculpins (Cottidae). – 34 –  The Salish Sea region exhibits exceptionally high archaeological abundance of herring, but with substantial variation among sub-regions (Figures 2.3 and 2.4). Sites along the northwestern and northeastern Strait of Georgia (eastern Vancouver Island, E. Strait of Georgia, respectively), the southwestern Strait of Georgia (southern Vancouver Island), and the Gulf Islands, have the highest mean relative abundances for herring (Figure 2.4). These high numbers contrast with sites in Puget Sound (N = 10, x̅ = 20.6%) and along the Fraser Delta (N = 10, x̅ = 33.7%), which tend to have abundance values less than 40%. This spatially variable distribution of herring across broad regions of the Northwest Coast and around the Salish Sea suggests that although herring was a ubiquitous component of coastal ecosystems, herring biomass (vis-à-vis other fish taxa) was relatively greater in some regions than others. – 35 –   Figure 2.5. Percent abundance of herring bones over time in sites with dated samples (N = 50, Table 2.2). Sites are arranged into three groups a) Multiple samples within a site aggregated into broad temporal groupings (ca. 500-2000 years per stratigraphic layer, N = 14 sites), b) Sites with a single vertical ‘column’ of contiguously sampled levels (N = 19), and c) Sites with multiple column samples from multiple site areas including Barkley Sound (N = 17). Each site is associated with one or more radiocarbon dates and/or contain historic-era artifacts in the upper levels indicating they date ca. AD 1778-1850 (Table 2.1). – 36 –  Temporal Patterns  The sites in the dataset span the past 10,700 calendar years. The majority of the sites (78%) have temporal components that fall within the past 2,500 years whereas much smaller numbers of sites have components that date prior to this period and only a handful that date before 5,000 years ago (Table 2.1). This temporal distribution broadly corresponds to the proportion of dated archaeological sites in the region (Ames and Maschner 1999:54), and is a product of site taphonomy, relative search effort, and geographically variable Holocene sea levels and shoreline locations (Mackie, et al. 2011). Of the 171 sites, 50 (29%) meet our criteria for examining detailed intra-site temporal patterning (Figures 2.5 and 2.6, see Materials and Methods). Similar to the overall dataset, over half of these sites are from the Salish Sea region (N = 27, 54%) and the majority date to within the past 2,500 years. Collectively, the 50 sites are comprised of 571 vertically distinct levels, each with sample sizes of >50 NISP (Table 2.2).  Herring occurs in the overwhelming majority of the 571 individual levels (99.3%) and the mean relative abundance of herring exceeds 20% in the majority of the 50 sites (88%). Variance in the relative abundance of herring is less than ±10% within 96% of individual sites (Figure 2.6, Table 2.2), further supporting an interpretation of broad consistency over time. Variance is correlated with sample size, however (Spearman’s rho=0.303, p=0.032, n=50), which reflects the influence of several thoroughly sampled sites from a single region with moderate variability (Barkley Sound, see Figure 2.5D). Herring ranks as the first or second-most numerous fish taxon in the majority (92.1%) of the 571 stratigraphic levels. Seven sites exhibit mean abundances lower than 20% and low rank orders (less than 3) for herring over time (Figure 2.6). Herring is only absent from four – 37 –  individual levels in four sites which also have consistently low mean abundance values (<4% NISP; Figure 2.5). The broad consistency in abundance through time is particularly striking considering these archaeological deposits likely represent a variety of cultural and ecological contexts, seasons of occupation, and depositional processes over multiple temporal scales (Cannon 2000a; Lyman 2003a). Of the two sites in our sample that exhibit highest intra-site variability (more than ±10% variance), one (Burnaby Narrows, Figure 2.5A5) spans a key period of subsistence change (Orchard 2011b) and sea-level change (Mackie, et al. 2011). In the other (Spring Cove, Figure 2.5B10), dramatic increases in anchovy in two of five occupational levels reduce the relative abundance of herring, which correspondingly increases variance (cf., Lyman 2003a). As with modern fisheries landing data, social, economic, and environmental circumstances as well as analytical resolution and ecological drivers, needs to be considered when interpreting fine-scale variability within a specific site.  Sites in close proximity and which are broadly contemporaneous display similar patterns of herring abundance over time (e.g., Tsawwassen, Beach Grove, and Crescent Beach around Boundary Bay [Figure 2.5A8-10]; Cape Lazo, Q’umu?xs, and DkSf-4 in Comox Harbour [Figure 2.5B6-7, 2.5C2a and 2.5C2b ]; and the Nanaimo Harbour / Departure Bay area [Figure 2.5C4a-b]). This similarity suggests sub-regional-scale coherence in herring abundance over time and indicates that the formation processes associated with each site assemblage (either human harvest of herring or natural taphonomic processes within a site) do not obscure the overall ability of the zooarchaeological data to accurately estimate abundance of herring in these settings. Furthermore, the observations of similar herring abundance values from multiple columns from spatially separate but temporally overlapping – 38 –  deposits within a single site (Figure 2.5C and Table 2.2) strengthens the inference that single column samples can document site-wide patterning for such abundant and ubiquitous taxa (McKechnie 2005b). These overall patterns thus support the value of zooarchaeological data as a proxy for past herring abundance.  Figure 2.6. Mean abundance and rank order of archaeological herring bones from levels in adequately sampled and dated sites (N = 50; arranged north to south). Error bars represent ±1 standard deviation (see Table 2.2 for supporting information). Collectively, these multiple analytical scales (Figures 2.5 and 2.6) and abundance measures indicate a pattern of relatively low temporal variability and broad consistency in – 39 –  herring use as represented among and within archaeological sites distributed across a large coastal region. In areas where herring was archaeologically abundant, it appears to be consistently available and abundant throughout the mid-to-late Holocene. Given the limits of radiocarbon dating and calibration, we cannot determine the extent to which the temporal resolution of our data may inhibit our ability to detect finer-scale (e.g. annual, decadal, multi-decadal) variation in herring abundance through time. If high variation in the local availability of herring over time was characteristic for this region, however, we hypothesize this would be apparent in more of our site sequences. Overall, this large-scale archaeological dataset indicates that herring was a desired and reliable source of food in coastal waters in most locations across the Northwest Coast.  Figure 2.7. Boxplot showing variability in the archaeological abundance of herring bones in relation to proximity by water to historically surveyed spawning locations (±500 m). Distance data calculated only for sites in BC within monitored areas (N = 79) and with components dating to within the last 2,500 years (Table 2.3). Archaeological data are grouped into relative abundance categories as per Figure 2.1. Comparison to Modern Spawn Location Data  To evaluate the ecological longevity of herring spawning habitat and the extent that herring spawning areas have changed historically, we determined the proximity of – 40 –  archaeological sites to herring spawning areas documented as a result of federal fisheries monitoring efforts that support stock assessment in British Columbia (Department of Fisheries and Oceans Canada 2012a). We limited our analysis to archaeological deposits dating within the last 2,500 years (‘late period sites’) and located within areas annually monitored for spawning activity since the mid 20th century (N = 79 sites). We observe that archaeological sites with greater than 80% herring are significantly closer (Mann-Whitney U = 525, Z=-2.936, p=0.003) to documented spawning locations than archaeological sites with less than 80% herring (Figure 2.7, Table 2.3). The significance of this difference is strengthened when we include data from nine sites with traditional and local ecological knowledge [TEK/LEK] (Haida Marine Traditional Knowledge Study Participants, et al. 2011; Hessing-Lewis, et al. 2011) on the BC Coast (Mann-Whitney U = 478, Z = -3.490, p=0.000). The strong correspondence between high archaeological abundance of herring bones and the multi-decade monitoring record of modern spawning sites is a compelling argument for the long-term site fidelity of herring spawning.  In British Columbia, over 30% (N = 35) of the 114 late period archaeological study sites are in areas that have not been monitored for herring spawn by Department of Fisheries and Oceans (DFO) (mean distance from closest monitored shoreline is 7.5 km, SD = 7.9 km). Of these, 14% (N = 5 sites) have herring abundance values over 80%. Based on the correspondence between sites with >80% herring bone abundance and documented spawning sites, we hypothesize that these five sites were also close (i.e., within ± 2 km) to former herring spawn locations. Thus, archaeological sites can be used to expand the ecological baseline for spawning populations and identify localities that may no longer support spawning or are missed by modern monitoring efforts. – 41 –  We also note a general correspondence between six abundance categories used by federal fisheries monitoring in British Columbia (see Materials and Methods and Table 2.3) to characterize the cumulative location and intensity of herring spawning, and six ordinal categories of archaeological abundance of herring. That is, the majority of late period archaeological assemblages (48 of 79 sites; 60%) have similar categorical abundance values to the fisheries monitoring dataset (± 2 km). However, 25% of these archaeological sites (N = 20 sites) contain a categorically greater abundance of herring than fisheries monitoring records whereas, only 10% have substantially lower abundance values than cumulative spawning records (N = 8 sites).  Collectively, these comparisons both provide information on the longevity of spawning habitats as well as the potential depletion of spawning habitats in historic times. The predominant correspondence between ordinal categories in the ancient and modern data indicates correspondence between ancient productivity of herring and the productivity of contemporary spawning habitats in a given location. This in turn indicates consistency in the geography of spawning habitats from the distant to the more recent past. Furthermore, the quarter of archaeological sites that have higher abundances of herring than is predicted by the federal cumulative spawn records provides a basis for expanding our estimate of potentially productive spawning habitat beyond what is indicated by historic census records.  Discussion This broad compilation of zooarchaeological data provides a previously unconsidered record of the past abundance and distribution of fisheries along the Northwest coast of North America. It establishes herring as among the most important marine fish on which coastal indigenous people relied in the past. These data expand greatly upon the extant – 42 –  paleoecological fisheries record (O’Connell and Tunnicliffe 2001; Tunnicliffe, et al. 2001; Wright, et al. 2005) and indicate that herring was both widespread across the coast, and a mainstay of ecological and socio-economic systems over the Holocene. Herring occurred in abundance (>40% NISP) in numerous sites across all areas of the coast but exhibited super-abundance (i.e., > 60% NISP) in a few regions such as the central and northern Strait of Georgia and west coast of Vancouver Island. Spatially clustered sites reflect similarly consistent abundance values and a few sites exhibit modest trends over time that may relate to physically driven habitat change, such as the geomorphological evolution of the Fraser Delta (e.g., Beach Grove and Tsawwassen). Only a small number of sites exhibit high variability whereas the more common pattern is of broadly sustained abundance over space and time (Figures 2.5 and 2.6).  Accounting for Differences Between Archaeological and Modern Herring Dynamics Could the contrast between the consistency seen in the archaeological record and patterns in recent fisheries showing overall decline and increasing fluctuation in spawning distribution arise from insufficient temporal resolution in the archaeological data? Even though modern population fluctuations are observed at interannual to decadal timescales and the likely resolution of the archaeological record is at the centennial scale, it is unlikely that the temporal consistency in the archaeological record is an artifact of insufficient chronological resolution for several reasons. First, if extreme fluctuations in abundance and periodic absences were common in the past (for instance, such that sites with great abundance in some periods commonly declined to low levels) we would expect to see this reflected in our data, even if we couldn’t resolve the actual temporal trajectory in an – 43 –  individual population. Alternatively, our data may also represent subannual temporal variability, which is masked by the coarseness of existing calibrated radiocarbon records. Both scenarios indicate that in our sample of 571 levels, across 50 sites and representing ca. 7,000 years of harvesting data, we should observe some indication of absences and wide fluctuations in herring abundance if they were present. Yet these data show a remarkably consistent presence of herring with 99.3% of the 571 levels containing herring and a mean relative abundance exceeding 20% in 86% of the sites, with within site variance of less than ±10% in 48 of 50 sites (Figure 2.7, Table 2.3). Second, because of the regularity with which ancient peoples ‘sampled’ their marine environments, the archeological record should be a sensitive indicator of marked declines in local abundance of herring as harvesters would conceivably switch to alternative prey when a target species declines to low levels. Such resource switching is observed in the regional zooarchaeological record, but is expressed over centennial scales and among a limited range of animal species other than herring (e.g., Butler and Campbell 2004; Orchard 2011b). In contrast, even substantial variability around a high mean abundance might not be reflected in our data if herring populations never declined to a threshold where people could meet their needs for consumption, storage, and trade. Likewise, another factor potentially contributing to consistency in the archaeological abundance of herring may be multi-season herring harvests, where fish, fish oil, and roe were processed for longer-term storage and extended consumption within a year (e.g., McKechnie 2005b:103; Moss, et al. 2011:283). This scenario may further dampen fluctuations in abundance over decadal or centennial time scales but also depends on consistent interannual availability of herring.  – 44 –  We envision three alternative hypotheses to account for the difference we observe between archaeological and modern data in the apparent variability and abundance of herring populations (Figure 2.8). The first hypothesis asserts that the modern pattern of interannual dynamics is an accurate reflection of the past and that the archaeological record overestimates abundance and underestimates variability in ancient herring populations (Figure 2.8, Hypothesis 1). For the reasons advanced above, we argue that this hypothesis can be rejected. Our data seem most consistent with hypotheses that ancient herring populations had a higher mean abundance than in the last century (Hypotheses 2 and 3, Figure 2.8). Variability could have been damped relative to the industrially exploited populations of the 20th century (Hypothesis 2, Figure 2.8), or it could have been similar to those harvested populations (Hypothesis 3, Figure 2.8). In either case, abundance in the more distant past was sufficiently high to meet the needs of harvesters, leading to consistency in the archaeological record. – 45 –   Figure 2.8. Schematic representing three alternative hypotheses regarding the relationship between modern and archaeological abundance and variability of herring populations. The left side of the figure combines the smooth archaeological signal (blue lines) with three hypothesized trajectories for actual populations (black lines), each shown with its range of variability (grey stippled rectangles). In each case, the end of the timeline following the onset of industrial fishing represents a caricature of the recent pattern of herring population dynamics, characterized by great variability, an overall decline in abundance, and some periods of very low abundance.  While the archaeological record shows consistency through time, the paleooceanographic record for the Northwest Coast illustrates considerable variability (e.g., Ivanochko, et al. 2008). Both ocean temperature and productivity vary throughout the late Holocene at different spatial and temporal scales, with some periods of interregional convergence (Anderson, et al. 2005a; Ersek, et al. 2012; Wright, et al. 2005). If ancient herring populations fluctuated in response to high frequency oceanographic variability (e.g., – 46 –  Tanasichuk 1997), the fluctuations were not of sufficient amplitude to influence the overall catch of ancient fishers (cf. Reitz, et al. 2008); the substantial variability in the paleooceanographic record is not matched by the archaeological record of herring. Moreover, it seems unlikely that climatic change alone is sufficient to account for substantial declines in Pacific herring populations and spawning distribution over the last 100 years (cf. Hsieh, et al. 2006; Lindegren, et al. 2013). The most parsimonious explanation for the difference between the modern pattern of variability in herring abundance and the long-term archaeological record is the onset of industrial-scale commercial fishing. Dramatic reductions and spatial shifts in herring populations were observed by fisheries managers in British Columbia and Alaska in the 1930s (Carrothers 1941:111; Funk 2010; Munro and Clemens 1931; Taylor 1955:110; Tester 1933:289; 1935:7). A period of decline between AD 1910 and 1970 is evident in the well-dated sediment cores from southeast Vancouver Island (O’Connell and Tunnicliffe 2001:187). Since the 1970s, further population constrictions have been observed by non-Native bait fisherman in the Georgia and Johnstone Straits (BC), including the widespread absence of herring in over 170 locations that previously supported spawning (Schweigert and Linekin 1990). Modern declines and contractions of the spawning range of Pacific herring are supported by recent syntheses of the traditional ecological knowledge of indigenous peoples in Alaska and British Columbia (Haida Marine Traditional Knowledge Study Participants, et al. 2011; Hebert and Thornton 2010; Hessing-Lewis, et al. 2011). These syntheses document quantifiable reductions in the magnitude of spawning events and loss of spawning locations in living memory. Furthermore, this shift in abundance is reflected in indigenous place – 47 –  names, which highlight locations of formerly abundant herring, but where few herring are found today (e.g., Ch'axa'y or “Sizzling [with herring] Water”]) (Reimer/Yumks 2011:237). Archaeological abundance of herring is additionally mirrored in indigenous place names and origin narratives. For instance, the 2,400 year-old site of Nulu on the Central BC Coast, where herring made up 85±11% of the fish assemblage (Cannon, et al. 2011), is the place where, according to Heiltsuk oral tradition, Raven first obtained herring (Boas 1932). Conversely, a culturally affiliated site at the Koeye River 25 km away from Nulu (Table 2.1), is not associated with herring or herring spawn either in ethnographic or modern ecological data and exhibits low archaeological herring abundance (10% NISP). In the northeastern Salish Sea in southern British Columbia, the place name of Teeshoshum (“waters white with herring spawn”) is associated with a ca. 800 year old assemblage comprised of 93% herring bones (Table 2.1). However, extensive herring spawning in this ecologically suitable location have not been documented since 1998 (Department of Fisheries and Oceans Canada 2012a).  Collectively, these sources of historical and ethnobiological data reinforce the contrast between the archaeological record and the dynamics of the modern fishery. In response to locally reduced or absent spawning populations, the current herring fishery in BC has contracted from its mid-20th century coast-wide focus and is now concentrated along the east coast of Vancouver Island, overlapping an area in which herring were super-abundant archaeologically (Department of Fisheries and Oceans Canada 2013; Therriault, et al. 2009). While other factors are also responsible for range contraction of harvested fish populations, range contractions are recognized as one of the effects of over-fishing (MacCall 1990; Worm and Tittensor 2011). – 48 –  Historical Baselines As Context for Modern Dynamics Historical baselines in natural resource management serve as reference conditions for understanding the context for modern population or ecosystem dynamics, providing both assessments of abundance and ranges of natural variability (Landres, et al. 1999; Rosenberg, et al. 2005). Our archaeological data, in combination with traditional ecological knowledge and early historic observations (Carrothers 1941; Funk 2010; Munro and Clemens 1931), suggest that late 20th century census data alone do not provide a sufficient baseline for assessing the abundance, distribution, and dynamics of Pacific herring in relation to industrial fishing since the 1880s. This discrepancy between recent and past dynamics is expressed in both temporal and spatial domains. In the temporal domain, the archaeological data challenge the notion that large fluctuations in abundance, including extremely low levels of abundance, are a regularly occurring component of population variability. In the spatial domain, our data argue against the idea that spawning was spatially erratic in the past, with little site fidelity (Hay, et al. 2009:1662). The archaeological data, in combination with the oral historical knowledge and early historic observations (Carrothers 1941; Funk 2010; Munro and Clemens 1931), suggest that industrial fishing already had a significant ecological impact on herring abundance and spawning location well before the initiation of coast-wide spawning censuses in British Columbia in 1938. Moreover, the historical baseline currently used for annual stock assessments that underpin current harvest allocations in BC begin only in 1951 (Department of Fisheries and Oceans Canada 2012b:11). The recent history of erratic population fluctuations, declines, and shifting spawning distributions exhibited by Pacific herring populations are not unusual among industrially harvested populations of forage fish worldwide (Anderson, et al. 2008; Hsieh, et al. 2006; – 49 –  Pikitch, et al. 2012; Pinsky, et al. 2011; Smith, ADM, et al. 2011). What is unique is our ability to provide a long-term temporal context for these recent dynamics. This is particularly relevant because these data reflect sustained continuous harvesting of herring populations for millennia prior to more than a century of modern industrial exploitation. Similar to historical research on the impacts of early industrial-era fishing on Atlantic cod (Rosenberg, et al. 2005) and herring in northern Europe (Barrett, et al. 2004; Poulsen 2008), and the long-term effects of human use of coral reefs (Kittinger, et al. 2011), the archaeological analyses synthesized here critically extend the temporal depth of ecological baselines for contemporary fisheries management.  Conclusion Spatially and temporally extensive archaeological data on the relative abundance of herring bones in coastal archaeological sites along the Northwest Coast provide insight into the past distribution and abundance of Pacific herring and long-term human-herring interactions. Herring bones exhibit a remarkable degree of dominance within the archaeofauna across space and time in the majority of these records. Over the period represented well by the archaeological record (ca. 2,500-200 years BP), Pacific herring populations also appear to have exhibited higher abundance and greater consistency in their distribution than is indicated by the dynamics of industrially harvested populations over the past 50-100 years. The archaeological data indicate that in most parts of the study area, and particularly in the Strait of Georgia, herring remained consistently available to harvesters over thousands of years. Of the hypotheses posed in Figure 2.8, we reject hypothesis 1, that the archaeological data misrepresent the actual abundance and variability of herring. At present, we cannot distinguish between hypotheses 2 and 3: herring abundance appears to – 50 –  have been consistently high, but we cannot resolve the magnitude of variability in abundance. The archaeological record, in combination with local and traditional knowledge, early historical records, and paleoecological records, suggest that spawning locations were formerly more abundant and geographically extensive than is recorded by modern surveys. These data, particularly in the context of high harvest levels during the early industrial fishery and the subsequent contraction of effective spawning range, indicate that the currently utilized ecological baseline of the mid 20th century is inadequate for modern management.  Our data support the idea (Hypotheses 2 and 3, Figure 2.8) that, if past populations of Pacific herring exhibited substantial variability, this variability was expressed around a high enough mean abundance such that there was adequate herring available for indigenous fishers to sustain their harvests while avoiding the extirpation of local populations. These records thus demonstrate a fishery that was sustainable at local and regional scales over millennia, and a resilient relationship between harvesters, herring, and environmental change that has been absent in the modern era. Archaeological data have the potential to provide a deep-time perspective on the interaction between humans and the resources on which they depend. Furthermore, it can contribute significantly towards developing temporally meaningful ecological baselines that avoid the biases of shorter term records.     – 51 –  Chapter 3. Towards a Historical Ecology of Pinniped and Sea Otter Hunting Traditions on the Coast of Southern British Columbia4 Introduction Marine mammals (pinnipeds, cetaceans, and sea otters, Enhydra lutris) have been important to First Nations people in coastal British Columbia for millennia but their archaeological distribution is poorly known. While archaeological evidence of marine mammal hunting is known for numerous locations over the past 10,000 years of human occupation on the British Columbia Coast (e.g., Cannon 1991; Carlson 2003; Fedje, et al. 2005a; Matson 1976), it is remarkable that few studies have examined archaeological evidence of mammalian hunting traditions on broad regional and/or temporal scales. Considering the importance of these animals to the modern marine ecosystem, understanding the long-term human use and past distribution of marine mammals can add considerable perspective to contemporary knowledge of these ecologically important species and this highly valued marine ecosystem (cf. Jackson, et al. 2001; Lotze and Worm 2009; Pitcher 2005). The occurrence of marine mammal bones in archaeological contexts reflects the direct use and long-term occupation of this region by coastal First Nations people. Such occurrences indicate that humans have been participants in this ecosystem for at least the past 10,000 years and as such, likely directly or indirectly affected the distribution, growth, behavior, and relative densities of marine mammals and their prey (cf. Crockford, et al. 2002;                                                 4 This chapter has been published previously as: McKechnie, Iain and Rebecca J. Wigen (2011) In Human Impacts on Seals, Sea Lions, and Sea Otters: Integrating Archaeology and Ecology in the Northeast Pacific, edited by T. J. Braje and T. C. Rick, pp. 129–166. University of California Press, Berkeley. (c) 2011 by the Regents of the University of California. See preface for more information. – 52 –  Etnier 2002a; Gifford-Gonzalez, et al. 2005; Lyman 2003b). In this chapter, we compile archaeological data on the distribution of pinnipeds and sea otters from archaeological assemblages along the coast of southern British Columbia. We evaluate the spatial and temporal extent of human hunting and explore the possible influence humans may have had on this aspect of the marine environment and conversely, discuss the potential significance that hunting these animals had to First Nations cultures in the region. Drawing on archaeological and ethnographic information, we ask three questions regarding the use of marine mammals by First Nations People over the past 8,000 years:  • What marine mammals did aboriginal people in Southern British Columbia most commonly utilize? • How similar or how different are species occurrences and proportions relative to today? • Is there evidence of specialized or regional hunting traditions and if so, what might have been the potential impacts of these activities on the ancient marine ecosystem?  Context Recent assessments of Northwest Coast archaeology have emphasized the need to adopt a regional approach to characterizing patterns of aboriginal resource use (Butler and Campbell 2004; Cannon 2001). In discussing the role of salmon on the central Northwest Coast, Cannon (2001:185) has noted that: “More important than improved recovery and analysis of fish remains from individual site locations is the pressing need to expand research strategies beyond the individual sites to encompass a variety of site locations within specific regions. This is the only way to gain a better appreciation of the extent of regional, seasonal, and longer-term temporal variability in the focus and intensity of local fisheries.”  – 53 –  Although Cannon’s suggestion is directed towards the archaeological assessment of fisheries, it is broadly applicable to archaeological research on the Northwest Coast. To date, only a few regional studies of indigenous resource use have been conducted (Butler and Campbell 2004; Crockford, et al. 2002; Hanson 1991; Hanson and Kusmer 2001; Hebda and Frederick 1990; Hobson and Driver 1989; Moss 2008; Orchard and Clark 2005). These efforts have provided a host of insights but have been hindered by the small number of published zooarchaeological studies, a methodological concern on sampling adequacy and taxonomic identification, and a focus on characterizing the ‘full range’ of animal consumption including birds, fish, mammals, and shellfish (e.g., Croes and Hackenberger 1988; Driver 1993; Hanson 1991; Mitchell 1990a). Additionally, many of these studies have focused on demonstrating quantitative differences in tabular form making it difficult to assess the potential continuity and spatial associations between adjacent sites and regions. In this chapter, we adopt an explicitly spatial approach to examining the archaeological expression of aboriginal hunting practices in southern British Columbia, focusing specifically pinnipeds and sea otters. We explore compositional patterns using zooarchaeological assemblage data from 75 temporally distinct assemblages from 58 sites spanning the past 8,000 years. We group assemblages into four broad temporal categories variously corresponding to broadly defined archaeological cultural historical ‘periods’ (300–1,200; 1,200–2,400; 2,400–5,000; 5,000–8,000 calibrated years BP) following Lepofsky et al. (2005; 2007) and Ames and Maschner (1999). These data provide context for considering the regional extent and potential influence that First Nations hunting practices had on the marine and terrestrial environments of southern British Columbia. – 54 –  The Modern Marine Ecosystem Our study area includes the western and eastern coasts of southern Vancouver Island, the Gulf Islands, and the British Columbia mainland and adjacent areas on the Olympic Peninsula in Washington State (approximately 40,000 km2, Figure 3.1). This region is temperate with occasional winter snowfall at sea level, high annual rainfall on western Vancouver Island (260-340 cm), and much lower rainfall on southeastern Vancouver Island and the Gulf Islands (70-110 cm) (Environment Canada 2009). The terrestrial environment is heavily forested with the densest forest cover on western Vancouver Island. The least densely forested areas are on southeastern Vancouver Island and the Gulf Islands, which contains patchy oak savanna woodlands and prairies with evidence of anthropogenic fire regimes extending back at least 2,000 years (Brown and Hebda 2002). The waters of the Strait of Georgia are heavily influenced by the seasonal input of freshwater from the Fraser River and other major rivers (Thomson 1981). Marine primary productivity is further influenced by nutrient-rich deep-water drawn into the Strait of Juan de Fuca from the north flowing California counter-current. This hyper-saline and oxygenated water is vertically mixed by the strong tidal currents in the Gulf Islands at which point this mixed surface water then flows back out the strait and northwards along the outer coast of Vancouver Island (Thomson, et al. 2007). The greatest annual primary productivity in the study area occurs off the coast of southwestern Vancouver Island (Ware and Thomson 2005), but strongly increases during Summer when wind-driven upwelling and prolonged sunlight along the continental shelf break facilitate massive blooms of plankton attracting a host of marine predators (Allen, et al. 2001).  – 55 –  The contemporary distribution of pinnipeds on the southern British Columbia coast is represented by six species occupying a range of habitats and exhibiting seasonal fluctuations in abundance (Table 3.1). Harbor seals (Phoca vitulina) are the most numerous and ubiquitous species which are present year-round (Department of Fisheries and Oceans Canada 2010). Northern fur seals (Callorhinus ursinus) are slightly more abundant but do not come ashore and are found almost exclusively on the continental shelf off western Vancouver Island, primarily between December and May (Olesiuk 2009b:57). Steller Sea lions (Eumetopias jubatus) while resident year-round, do not have rookeries in the study area and are only seasonally present in the Strait of Georgia during Winter and Spring (Bigg 1988; Olesiuk 2009a). California sea lions (Zalophus californianus) are present in lesser numbers, and elephant seals (Mirounga angustirostris) migrate to the southern BC coast seasonally (Table 3.1). Elephant seals have recently established a year-round haul-out on Race Rocks on the southern tip of Vancouver Island where at least two elephant seals pups have been born in 2009 and 2010 ( 2010). The contemporary British Columbia sea otter population consists of 4,182 animals (Nichol, et al. 2009), which have increased from a re-introduced population of 89 animals transported from the Aleutian Islands to northern Vancouver Island in between 1969-1972 (Bigg and MacAskie 1978). This population has been increasing annually at a rate of 15% but has yet to entirely re-colonize its former habitat (Nichol, et al. 2009). A recent population reconstruction (Gregr, et al. 2008), incorporating habitat characteristics and contact-era (ca. AD 1774-1825) fur trade records, suggests that the British Columbia coast may have once supported a population of between 34,000 and 74,000 otters. Interestingly, this estimate – 56 –  excludes the Strait of Georgia due to the absence of sea otters in published archaeological assemblages (i.e., Hanson and Kusmer 2001).  Until 1970, the federal Department of Fisheries and Oceans supported ‘population control’ programs, which offered incentives to reduce pinniped populations (particularly Harbor seals and Steller sea lions) (Jeffries, et al. 2003; Olesiuk 2009a). These controls ended when the marine mammal protections were implemented in 1970 (Olesiuk 2009b). Since this time, Harbour seal populations have increased dramatically from 10,000 animals to 105,000 animals (Department of Fisheries and Oceans Canada 2010). Harbour seals in particular were viewed as competitors with human fisheries, especially salmon, despite the fact that the bulk of their prey is composed of non-salmonids (Olesiuk, et al. 1990).  Historic Changes to the Marine Ecosystem Marine mammals, particularly the trade of their highly valued pelts, were central throughout the process of early historic contact between First Nations and Europeans on the Northwest Coast of North America (Fisher 1977; Lutz 2008). Sea otters, fur seals, and then whales sequentially became the focus of international commercial industries, which rapidly diminished these culturally and ecologically significant animals over several decades. The corresponding impact of the sudden removal of these ecologically important animals is poorly known (due in part to the paucity of archaeological data and lack of long-term ecological studies) and this issue is a contemporary research concern for ecologists and managers throughout northern Pacific (e.g., Misarti, et al. 2009; Springer, et al. 2003). The early ‘contact period’ on the Northwest Coast (ca. AD 1774-1812) focused on the trade of sea otter pelts obtained from the western coast of Vancouver Island as well as the – 57 –  northern and central British Columbia Coast (Busch and Gough 1997; Dick 2006). Sea otter pelts were recognized as wealth items and chiefly regalia within aboriginal communities prior to contact with European trading vessels. Aboriginal exchange of pelts for highly valued European trade goods was therefore immediately incorporated into the indigenous economy and caused a large expansion of this particular hunting activity (Dick 2006; Lutz 2008). Over a few decades, the burgeoning trade led to an increasing scarcity of sea otters and increasing conflict between European and American traders and First Nation communities. European trading ships ceased to seek sea otter pelts on the southern British Columbia coast after 1812 following a series of dramatic incidents involving the destruction of European ships and crew as well as reprisal attacks on aboriginal villages (e.g., McMillan 1999:188). After a period of several decades of infrequent interaction, the European colonial presence on the coast increased substantially during the mid 19th century. In the 1880’s through the early 1900’s, commercial demand for seal pelts in the European market fueled a commercial sealing industry, which targeted northern fur seals off British Columbia Coast and Washington State (Crockford 1996). Dozens of schooners operating out of Victoria and Seattle employed hunters from First Nations communities on western Vancouver Island and the Olympic peninsula to travel offshore where hunters would launch traditional dugout canoes and use harpoons to approach and capture fur seals ‘sleeping’ in the surface waters (Swan 1883). In 1880 alone, over 12,000 fur seals were harvested off the coast of Vancouver Island (Swan 1887:397). Growing international pressure concerning the declining stocks of fur seals at their breeding grounds in the Bearing Sea in the late 19th century (Elliott 1886) – 58 –  eventually resulted in one of the first international treaties protecting marine wildlife in 1911, and ended the commercial harvest of fur seals in Canadian waters (Gay 1987). Also during this time, a nascent whaling industry developed on the British Columbia coast during the mid 19th century but expanded significantly after 1905, regionally diminishing whale populations, particularly sperm whales (Physeter macrocephalus), fin whales (Balaenoptera physalus), and humpbacks (Megaptera novaeangliae) (Gregr, et al. 2000). Non-migratory populations of humpback whales residing in coastal inlets were among the first whales targeted (Merilees 1985; Webb 1988). Whaling continued on the British Columbia coast until it was halted in 1967. Historic Changes Among Coastal First Nations The dramatic changes described above occurred in the context of tumultuous and tragic social, economic, and demographic changes within aboriginal communities over the last two centuries. Particularly devastating were introduced diseases (particularly smallpox and measles), which swept through the coastal population in the 1790’s and later in the 1850’s and 1860’s (Boyd 1999). These epidemics severely reduced the aboriginal population and correspondingly destabilized the highly structured social and political networks and dramatically reduced the scale and extent of the aboriginal subsistence economy. A large influx of European and Asian colonists in the mid to late 19th century introduced a host of new opportunities and challenges for sustaining and generating wealth in aboriginal communities. Many aboriginal people began to participate in the developing wage economy in forestry, fisheries, and agriculture on a seasonal basis (Lutz 2008). Among the highest wages were those offered to aboriginal hunters employed in the fur seal industry who earned – 59 –  seasonal incomes well above European counterparts (Lutz 2008). This industry persisted until the 1911 international fur seal treaty between America, Japan, Russia and Canada mandated a halt to this once central economic activity. Also during the late 19th century, anthropological fieldworkers began to document aboriginal peoples of the coast (e.g., Boas 1887). Numerous researchers compiled painstakingly detailed description and oral historical accounts of aboriginal lifeways including marine mammal hunting practices and rituals associated with hunting (e.g., Sapir and Swadesh 1939; Waterman 1920). Whaling was a particularly honored tradition among Nuu-chah-nulth peoples on the west coast of Vancouver Island and Makah peoples on the Olympic Peninsula (McMillan 1999) but sealing was also a respected skill in coastal aboriginal communities on both the east and west coasts of Vancouver Island (Drucker 1951; Elmendorf 1960; Suttles 1952). Hunters were typically chiefs or men of high status who underwent elaborate ritual and physical preparation (Arima 1988; Drucker 1951; Elmendorf 1960:86; Suttles 1952; Waterman 1920). Hunter’s wives were obligated to observe ritual practices prior to and during the hunt and contributed to labor to butchering and cooking as well as preparing materials which comprised all aspects of hunting technology (e.g., preparing sinew, seal skin floats, cedar bark rope, etc.). Hunting took place from dugout canoes using harpoon technology with seal-skin floats or from land using clubs. A host of historic and ethnographic accounts are available for further investigating hunting practices and techniques, some of which are described further in the sections below. Methods To investigate the regional expression of marine mammal hunting practices, we compiled available zooarchaeological data from academic papers, graduate student theses, – 60 –  and provincially reviewed archaeological permit reports on file at the British Columbia Archaeology Branch in Victoria or federally reviewed reports on file with Cultural Resource Management Services Division of Parks Canada (Calgary, AB) (Table 3.2). All sites are within 500 m of the coastline or along the banks of lower Fraser River (a major tidally influenced river draining into the Strait of Georgia). Sites that are more than 500 m from the ocean (‘inland’ sites) are not included in the present analysis because they have not been investigated to the same degree as shoreline shell midden sites and generally contain poorly preserved faunal assemblages. All assemblages included in the analysis are from pre-contact archaeological deposits (ca. > AD 1774) and date to with the past 8,000 years. However, the majority of the assemblages date to the last 2,400 years. Most sites in the study sample (>90%) are from deposits which have been radiocarbon dated. Assemblages without radiocarbon ages are either from the upper layers of ethnographically identified village locations (assumed to date within the past 1,200 years) or are associated with temporally diagnostic artifact assemblages. To enable comparability between regions, we collapse assemblages into one of four broad temporal categories (300–1,200; 1,200–2,400; 2,400–5,000; 5,000–8,000 calibrated years BP, Table 3.2).        – 61 –  Table 3.1 Ecological characteristics for pinnipeds and sea otters in British Columbia. Common Name Northern Fur Seal Harbour Seal Steller Sea Lion California Sea Lion Elephant Seal Sea Otter Species  Callorhinus ursinus Phoca vitulina richardsi Eumetopias jubatus Zalophus californianus Mirounga angustirostris Enhydra lutris Mean Female Wt. (kg) 35-45 58 200-300 94 650 44 Mean Male Wt. (kg) 100-200 72.5 400-800 375 1800 33 Modern Period of Pup Dependency 7-8 months One month 12-24 months 12-24 months One month 5-8 months Modern Birthing Season, Location July – Nov., Bearing Sea - California May - June, Throughout BC May - June, N BC May – July, California Oregon Dec -March, California & Mexico Year-round with peak in Spring Modern Seasonal Occurrence in  March - June, with peak in May Resident in BC year round Resident in BC year round Year round with peak Oct – May Recently present year-round Resident year-round Prevalence in Archaeological Sites in Southern BC Ubiquitous on western Vancouver Island Ubiquitous Common Rare Rare but increasing Present on Western VI, rare/absent on Eastern VI Estimated Contemporary Population in BC (2000-2010) 123,000 105,000 18,400-19,700 2,000-3,000 Fewer than 100 4,712 Management Status in Canada  Threatened Not at Risk Special Concern Not at Risk Not at Risk Special Concern IUCN Red List Status (2009) Vulnerable Least Concern Endangered Least Concern Least Concern Endangered References (Banfield 1974; Department of Fisheries and Oceans Canada 2007:5) (Baird 2001: 668; Olesiuk 1999: 28) (COSEWIC 2003) (Bigg 1985:17; Hancock 1970; NOAA Fisheries Service 2009)  (Stewart 1999: 217-218) (Estes 1999:182; Nichol, et al. 2009)  Zooarchaeological assemblages included in the analysis were identified using comparative osteological collections based at large universities in the region (e.g., Simon Fraser University, University of British Columbia, University of Victoria, and Washington State University) and/or at the Royal BC Museum in Victoria (RBCM). Individual bone specimens were identified to the lowest taxonomic level possible on the basis of their morphological similarly to comparative specimens. A majority of analyses have been – 62 –  conducted by or with the consultation of a few individuals who have a long history of work in the region (e.g., Susan Crockford; Jon Driver; Gay Frederick [formerly Calvert & Boehm]; Gregory Monks; Rebecca Wigen). Researchers at the University of Victoria and the RBCM have assembled a comprehensive comparative collection of vertebrates in the North Pacific including marine and terrestrial mammals from multiple individuals of different age and sex classes. Other universities have less comprehensive collections, but include common mammals like deer and harbor seal. Identification and quantification criteria have been reviewed and standardized for consistency. For identifications that appear questionable, as in the case of northern fur seals identified from Burrard Inlet in North Vancouver (i.e., Galdikas-Brindamour 1972), revised assessments of earlier identifications are used (i.e., Trost 2005). Faunal remains from the majority of sites included in the analysis were recovered from controlled excavations using standard 1/4” (6.35 mm) mesh. However, in one instance of salvage data recovery, mammal bones were collected by hand from excavated sediments (Site 46, Table 3.2). Mammal remains recovered from fine mesh bulk samples were not included in the analysis due to the small sample sizes and their lack of comparability with fauna from larger volumes of excavated sediment. In most cases, mammal bones were not recovered or were in very low numbers in these fine mesh bulk samples. Faunal identification data are quantified by number of identified specimens (NISP), which includes individual skeletal elements or fragments that can be confidently attributed to taxonomic categories of family, genus, or species level designation (e.g., cervidae, deer, or mule deer). Highly fragmented remains that cannot be confidently assigned to family, genus, or species level are not included in the analysis (e.g., unidentified sea mammal). – 63 –  Quantification of relative abundance is evaluated by comparing the percentage of identified specimens (NISP) from individual taxonomic categories to all other positively identified specimens (e.g., %NISP of pinnipeds). Species or genus-specific taxonomic identifications are shown in Table 3.2 but for the purpose of this analysis, specimens have been lumped into larger categories of ‘marine’ and ‘terrestrial’ mammals. The marine category includes all pinnipeds (seals and sea lions), delphinids (porpoises, dolphins), and sea otters. River otters (Lontra canadensis) are also included in the marine mammal category, as they inhabit coastal waters throughout the study area (Banfield 1974). Whales have been excluded due to the highly fragmentary nature of these large elements, which make identification and quantification problematic (Huelsbeck 1988; Monks 2001). Canids and domestic dogs (the only domesticated animal in the region) have been excluded from this analysis since it is not clear they are used for dietary consumption (Crockford 1997). Age and sex classification (based on epiphyseal fusion, eruption and wear of teeth) is not consistently recorded for most of the assemblages in this study. However, sex information is available for a small selection of the examined sites for adult-sized sea lions and fur seals (summarized in Table 3.3). Ideally, future analyses will examine incremental skeletal structures, such as growth lines in non-deciduous teeth as well as measuring the dimensions of diagnostic elements, which has been published for a handful of sites (e.g., Crockford, et al. 2002; Etnier 2002b).  The compiled assemblage data have been entered into to a geographic information system (GIS) and plotted over a series of geospatial layers depicting the major river drainages, the terrestrial topography and the 200 and 500 m bathymetric contour lines (Figure – 64 –  3.2). Mammalian fauna from individual sites are represented as pie charts showing the relative abundance of mammals for different taxonomic categories. Our analysis examines the differences in the relative composition between marine and terrestrial mammals for all sites and time periods and then examines species-specific patterns for marine mammals.   Figure 3.1. Map of the study area showing the distribution of archaeological assemblages used in the analysis and their chronological age-range.  Results Seventy-five temporally distinct assemblages from 58 archaeological sites are included in this analysis, representing a combined total of 67,500 mammalian skeletal specimens (Table 3.2). Fifty-one assemblages are from the Strait of Georgia region, 21 are from the west coast of Vancouver Island, and three are from the Olympic Peninsula in Washington State. Site distribution is relatively more concentrated in areas impacted by modern development and sparse in other areas, particularly north of the city of Vancouver and along on southwestern Vancouver Island (Figure 3.1). Slightly more than half of the sites – 65 –  (53%, N=31) were excavated as part of academically driven research projects while the remainder were cultural resource management projects mandated under heritage legislation in the province of British Columbia. The most recent temporal period (ca. 300-1200 cal BP) contains the greatest number of assemblages (N=36) and this progressively decreases with assemblage age (1200-2400 cal BP, N=26; 2400-5000, N=11; 5000-8000, N=2). This trend reflects a sampling bias towards younger sites associated with modern sea levels and a lack of research devoted to locating and sampling earlier Holocene archaeological assemblages (cf., Fedje, et al. 2004; 2011b). Nine sites span multiple temporal periods whereas the remaining sites fall into one of the four broadly defined temporal periods (Table 3.2). Assemblage size (NISP) ranges widely from a low of five mammalian elements (Site 38) to a high of 51,937 (Site 50) but over 70% of the assemblages have more than 50 identified specimens. The largest assemblage from Ozette (Site 50), was the product of a decades-long collaborative project between the Makah tribe and a team of archaeologists from Washington State University (Huelsbeck 1994; Samuels and Daugherty 1991). Sites with small assemblages (< 50 NISP) are included in the analysis as they represent additional locations with confidently identified taxonomic information. – 66 –   Figure 3.2. Relative composition of marine and terrestrial mammals (%NISP) for all sites in the study area (n=58). Chronologically distinct periods have been combined in this figure. Spatial and Temporal Patterning Figure 3.2 illustrates the relative abundance of marine and terrestrial mammals among all examined assemblages spanning the past 8,000 years and shows a sharply defined regional difference in the use of marine mammals and terrestrial mammals between the west and east coasts of Vancouver Island. Sites in the Strait of Georgia show remarkable consistency, with terrestrial mammals comprising the majority of all mammalian assemblages. Sites on western Vancouver Island in contrast, are composed, primarily, of assemblages with more than 75% marine mammals (Figure 3.2). Notable exceptions to this pattern occur at two sites on western Vancouver Island, which contain a higher percentage of terrestrial than marine mammals. One of these sites is a rockshelter in a sheltered area of Hesquiat Harbour (Site 35) and the other is located on a highly exposed linear coastline near a river that is navigable for 10 km upstream but where the surf makes it hazardous to launch canoes in most conditions (Site 46). In the Strait of Georgia, sites that show the greatest – 67 –  percentage of marine mammals are in the Gulf Islands and the Saanich and Tsawwassen peninsulas (Sites 15-18, 21-24). Sites situated on the east coast of Vancouver Island show a consistently low relative percentage of marine mammals relative to elsewhere in the Gulf of Georgia. Shoemaker Bay (Site 45), a site at the head of an inlet, which deeply incises the western coast of Vancouver Island, has an assemblage dominated by terrestrial mammals that appears more similar to eastern Vancouver Island assemblages. Saltery Bay (Site 51), a multi-component site in the north of the study area, shows a notably high percentage of marine mammals relative to other sites in the Strait of Georgia.  Figure 3.3 depicts the chronological sequence of the relative composition of marine and terrestrial mammals over the four distinct time periods. Terrestrial mammals represent a majority of the two earliest assemblages in the Strait of Georgia and this pattern follows in subsequent time periods with the notable exception of Saltery Bay (Site 51). Similarly, sites on western Vancouver Island have a majority of marine mammals and this is consistently represented over the three temporal periods with available data (Figure 3.3). These persistent compositional patterns in regionally distributed mammal assemblages indicate remarkable continuity over long periods of time. However, temporal trends within particular sites are also apparent, although more subtle. Some individual sites have suggestive chronological changes in the relative composition of marine versus terrestrial mammals. For instance, Helen Point (Site 21) shows a relative decrease in marine mammal composition over three successive time periods which speculatively correlates with greater terrestrial mammal composition on southeastern Vancouver Island (Sites 8-18). In contrast, sites on southwestern Vancouver Island and Washington State appear to show a progressive increase in marine mammal percentages over the three time periods (sites 34-50). While these trends – 68 –  are suggestive of regionalized subsistence economies, it is not within the scope of this paper to evaluate each of these possible local patterns in detail.  – 69 –   Figure 3.3. Relative composition of marine and terrestrial mammals (%NISP) for all sites in the study area according to temporal period. Figure 3.4 shows the same temporal sequence of analyses shown in Figure 3.3, but depicts only the relative composition of three most common pinniped species (harbour seal, Steller sea lion, northern fur seal) and sea otter. This analysis purposefully excludes terrestrial and cetacean (delphinid) specimens, which correspondingly reduces sample size – 70 –  and alters the representation of relative percentage (i.e., %NISP refers to just these species). Nevertheless, this analysis indicates that assemblages on western Vancouver Island are predominantly composed of northern fur seal, followed by Steller sea lion and harbor seal. Northern fur seals are strikingly absent from all sites in the Strait of Georgia and harbor seals predominate the marine mammal assemblages in the Strait of Georgia followed distantly by Steller sea lions. Sea otters are present at eight sites on western Vancouver Island and Washington and comprise substantial proportions of four of these assemblages (Sites 34, 36, 44, & 45). Sea otters are present at five sites in the Strait of Georgia dating to between 2400 and 300 yr BP (Sites 13, 23, 52, and 53), which contrasts with the findings of Hanson and Kusmer (2001) who observed a near absence of sea otters in this region. Where sea otters are present, they represent small proportion of the assemblages except for one small marine mammal assemblage (NISP=4), which is composed entirely of sea otter (Site 13). Steller sea lions also are present at sites on southern and western Vancouver Island but are found in greater densities on western Vancouver Island and in lower percentages along the Strait of Georgia (Figure 3.3).  California sea lions are identified in only six sites in the study area, all on the exposed Pacific coast (Table 3.2). Elephant seals are similarly present in small numbers at four sites in the western portion of study area (Sites 41, 42, 48, and 49). Guadalupe fur seals, whose skeletal morphology closely resembles the northern fur seal (Etnier 2002b), are only identified at one site, which also has the largest identified assemblage (Ozette, Site 50).  – 71 –   – 72 –   Figure 3.4. Relative abundance of marine versus terrestrial mammals for sites dating to within the four examined periods (5000-8000, 2500-5000, 1200-2500 and 300-1200 BP). – 73 –  Table 3.2 Site Names, Locations, References, and Number of Identified Specimens (NISP) used in the analysis. All Mammal totals exclude whales and domestic dogs and canids. No. Region* Site # Period Site Name Reference Harbor S. Steller SL N Fur Seal C. Sea Lion Guadelupe Fur Seal Elephant seal Sea Otter Terrestrial Mammals All Mammals 1 ecVI DjSf-13 1200-2400 Buckley Bay Wigen 1980 4 10 0 0 0 0 0 221 235 2 ecVI DjSf-14 1200-2400 Tsable River Wigen 1980 2 4 0 0 0 0 0 194 203 3 ecVI DkSg-2 300-1200 Sandwick midden Wigen 1980 citing Capes,1964 3 0 0 0 0 0 0 64 75 4 ecVI DiSc-1 300-1200 Little Qualicum River Bernick and Wigen 1990 18 1 0 0 0 0 0 243 263 5 ecVI DiSc-26 1200-2400 Qualicum Beach Golf Course Willows et al. 2008 0 0 0 0 0 0 0 10 11 6 ecVI DhSb-3 1200-2400 Dogwood St., Parksville Wilson et al. 2006 0 0 0 0 0 0 0 12 12 7 ecVI DhRx-16 1200-2400 Departure Bay Wilson and Crockford 1994:118 6 0 0 0 0 0 0 733 743 8 scVI DcRu-74 1200-2400 Esquimalt lagoon Wet site McKechnie, 2004:51, Table 19 3 0 0 0 0 0 0 63 67 9 scVI DcRu-2 300-1200 Esquimalt Lagoon Hanson 1991:Table 6 citing Stevenson 1978 9 3 0 0 0 0 0 254 266 10 scVI DcRu-2 300-1200 Esquimalt/ Ft Rodd Hill Crockford, 1997b 17 0 0 0 0 0 0 337 358 11 scVI DcRu-78 300-1200 Ft Rodd Hill Mitchell 1981 0 5 0 0 0 0 0 88 96 12 scVI DcRu-136 300-1200 Esquimalt Harbour Wigen 2002 5 0 0 0 0 0 0 53 58 13 scVI DcRu-4 300-1200 Kosapsom Stewart and Wigen 2003 0 0 0 0 0 0 4 19 23 14 scVI DcRu-75 300-1200 Dallas Road Wilson et al. 2003:Table 28 14 0 0 0 0 0 0 126 143 15 scVI DcRt-16 300-1200 King George Terrace Wilson et al. 2004:107, Table 38 3 4 0 0 0 0 0 111 119 16 scVI DcRu-71 1200-2400 Eagles Nest Eldridge 2000:40 11 2 0 0 0 0 0 54 77 17 scVI DdRu-18 300-1200 Eventide Rd. Weathers et al. 2007:29, Table 6 2 5 0 0 0 0 0 37 44 18 scVI DdRu-5 1200-2400 Patricia Bay Kanipe et al. 2006:264, Table 46 9 10 0 0 0 0 0 145 165 – 74 –  No. Region* Site # Period Site Name Reference Harbor S. Steller SL N Fur Seal C. Sea Lion Guadelupe Fur Seal Elephant seal Sea Otter Terrestrial Mammals All Mammals 19 Gulf Is DgRu-3 1200-2400 Dionisio Point Ewonus, 2006:39, Table 2 3 0 0 0 0 0 0 88 96 20 Gulf Is DfRv-106 300-1200 Galiano Island Mason et al. 1995:21, Table 4 0 1 0 0 0 0 0 20 21 21.1 Gulf Is DfRu-8 300-1200 Helen Point  Boucher, 1976: Assemblages 12 ,13, & 14 12 1 0 0 0 0 0 180 194 21.2 Gulf Is DfRu-8 1200-2400 Helen Point  Boucher, 1976: Assemblages 5 & 6 18 5 0 0 0 0 0 93 119 21.3 Gulf Is DfRu-8 2400-5000 Helen Point  Boucher, 1976: Assemblages 1, 2, 3, & 4 187 54 0 0 0 0 1 360 664 22 Gulf Is DhRt-1 300-1200 Pender Canal Hanson, 1995:34 62 0 0 0 0 0 0 406 525 23 Gulf Is 45SJ169 1200-2400 Decatur Island-169 Lyman, 2003 24 0 0 0 0 0 1 154 179 24 Gulf Is 45SJ165 1200-2400 Decatur Island-165 Lyman, 2003 10 0 0 0 0 0 0 77 87 25.1 Mnlnd DgRr-2 300-1200 Tsawwassen Butler & Campbell 2004:353, citing Kusmer, 1994 4 0 0 0 0 0 0 16 20 25.2 Mnlnd DgRr-2 1200-2400 Tsawwassen Butler & Campbell 2004:353, citing Kusmer, 1994 1 0 0 0 0 0 0 5 6 26 Mnlnd DgRr-1 1200-2400 Crescent Beach Hanson 1991:Table 6, citing Ham 1982 2 1 0 0 0 0 0 805 820 27 Mnlnd DhRr 18 1200-2400 Point Grey Coupland 1991:89 0 0 0 0 0 0 0 16 16 28 Mnlnd DhRr-6 300-1200 Belcarra (Tum-tu-may-whueton) Trost 2005:116 citing Charlton 1977:226 &Galdikas-Brindamour 1972 10 0 0 0 0 0 0 227 239 29 Mnlnd DhRr-8 300-1200 Cates Park (Whey-Ah-Wichen) Trost 2005:116 citing Williams 1974 14 0 0 0 0 0 0 227 251 30 Mnlnd DhRr-18 300-1200 Cove Cliff (Say-Umiton) Trost 2005:112 7 0 0 0 0 0 0 123 133 31.1 Mnlnd DgRr-2 300-1200 St. Mungo - late Boehm 1973: Table VIII 4 0 0 0 0 0 0 15 19 31.2 Mnlnd DgRr-2 1200-2400 St. Mungo - mid Boehm 1973: Table VIII 3 0 0 0 0 0 0 6 9 31.3 Mnlnd DgRr-2 2400-5000 St. Mungo - early Boehm 1973: Table VIII 23 0 0 0 0 0 0 212 235 32.1 Mnlnd DgRr-6 1200-2400 Glenrose Cannery-late Imamoto 1976: Table 2.1 6 0 0 0 0 0 0 70 76 32.2 Mnlnd DgRr-6 2400- Glenrose Cannery- Imamoto 1976: Table 2.1 23 0 0 0 0 0 0 193 216 – 75 –  No. Region* Site # Period Site Name Reference Harbor S. Steller SL N Fur Seal C. Sea Lion Guadelupe Fur Seal Elephant seal Sea Otter Terrestrial Mammals All Mammals 5000 mid 32.3 Mnlnd DgRr-6 5000-8000 Glenrose Cannery-early Imamoto 1976: Table 2.1 3 0 0 0 0 0 0 32 35 33 Mnld DhRp-17 1200-2400 Port Hammond Rousseau et al. 2003:101 3 0 0 0 0 0 0 264 267 34 wcVI DiSo-9 1200-2400 Hesquiat DiSo9Combined Calvert 1980: Tables 43 & 47 38 1 100 2 0 0 68 88 328 35 wcVI DiSo-16 300-1200 Hesquiat rockshelter Calvert, 1980: Table 39 0 0 0 0 0 0 0 66 126 36 wcVI DiSo-1 300-1200 Hesquiat Village combined Calvert, 1980: Tables 51, 55, 59, 63, & 67 86 41 228 29 0 0 90 155 844 37 wcVI DgSl-67 300-1200 Chesterman Beach Wilson 1994: Table 5 0 0 25 12 0 0 0 6 52 38 wcVI DgSk-1 300-1200 Long Beach Wigen, 2003: Table 2 0 1 0 0 0 0 0 0 5 39 wcVI DfSj-100 2400-5000 Little Beach Weathers et al. 2008; Wigen, 2008: Table 5 0 1 19 0 0 0 0 3 25 40 wcVI DfSj-57 1200-2400 Spring Cove, Ucluelet Spady& Wigen 2008: Table 15 1 0 49 0 0 0 0 5 57 41 wcVI DfSi–5 300-1200 Ma'acoah Monks 2006:224 32 2 84 14 0 1 6 71 224 42.1 wcVI DfSi-16+17 300-1200 Tsishaa Late McKechnie 2007a: Tables 17 & 18, including Frederick & Crockford 2005 28 13 214 1 0 0 3 29 344 42.2 wcVI DfSi-16+17 1200-2400 Tsishaa Mid McKechnie 2007a: Tables 17 & 18, including Frederick & Crockford 2005 14 0 49 0 0 1 1 25 109 42.3 wcVI DfSi 16+17 2400-5000 Tsishaa Early McKechnie 2007a: Tables 17 & 18, including Frederick & Crockford 2005 2 7 35 0 0 0 2 131 408 43 wcVI DfSi-26 300-1200 Clarke Island McKechnie 2007: Tables 17 & 18 6 7 56 0 0 0 0 9 80 44.1 wcVI DfSh-7 300-1200 Huu7ii late Frederick et al. 2006: Tables 6, 8 & 10 21 24 46 0 0 0 12 49 298 44.2 wcVI DfSh-7 1200-2400 Huu7ii mid Frederick et al. 2006: Tables 6, 8 & 10 5 1 1 0 0 0 0 3 17 44.3 wcVI DfSh 7 2400-5000 Huu7ii Early Frederick et al. 2006: Tables 6, 8 & 10 9 16 39 0 0 0 3 65 169 45.1 wcVI DhSe-2 300-1200 Shoemaker bay Late Calvert and Crockford 1982:186 75 0 11 0 0 0 0 781 889 45.2 wcVI DhSe-2 1200-2400 Shoemaker bay Mid Calvert and Crockford 1982:186 13 2 0 0 0 0 0 182 201 45.3 wcVI DhSe 2 2400- Shoemaker bay Calvert and Crockford 1982:186 7 0 1 0 0 0 0 91 101 – 76 –  No. Region* Site # Period Site Name Reference Harbor S. Steller SL N Fur Seal C. Sea Lion Guadelupe Fur Seal Elephant seal Sea Otter Terrestrial Mammals All Mammals 5000 Early 46 wcVI DeSf-6 300-1200 Klanawa McKechnie, 2007b:7 7 0 8 0 0 0 0 59 75 47 wcVI DeSf-2 300-1200 Tsuxwkwaada McKechnie 2005: Table 1 3 0 185 0 0 0 0 3 191 48 WA  45CA213 1200-2400 Hoko R. Wet/Dry site Croes 1995: 71 7 1 3 0 0 0 0 10 22 49 WA  45CA21 300-1200 Hoko R. Rockshelter Wigen 2005: Table 4.4 114 40 2425 0 0 1 40 423 3318 50 WA  45CA24 300-1200 Ozette Huelsbeck 1994:28 w additions from Etnier 2007:200 377 10 47296 1 34 2 501 708 51007 51.3 Strait of Georgia DkSb-30 5000-8000 Saltery Bay Pegg et al. 2007: Table 20 17 0 0 0 0 0 0 108 129 51.2 Strait of Georgia DkSb-30 2400-5000 Saltery Bay Pegg et al. 2007: Table 20 6 2 0 0 0 0 0 76 163 51.1 Strait of Georgia DkSb-30 300-1200 Saltery Bay Pegg et al. 2007: Table 20 3 0 0 0 0 0 0 14 27 52 Gulf Is 45-SJ-24 300-1200 English Camp Pegg 1999:69 20 0 0 0 0 0 1 310 372 53 Strait of Georgia DKSf-4 300-1200 Comox Harbour Site Simonsen 1990:35, Table 11 2 7 0 0 0 0 2 262 292 54 wcVI DfSg-2 300-1200 Aguilar House Engish 2006:28 2 4 19 0 0 0 2 6 36 55 Gulf Is DgRv-2 1200-2400 Shingle Point Matson et al. 1999:61-71 4 0 0 0 0 0 0 107 116 56 Mnlnd DhRt-6 2400-5000 Locarno Beach Brolly and Muir 1993:52, Table 12, excluding 2mm mesh 1 0 0 0 0 0 0 67 68 57 Gulf Is DfRu-3 1200-2400 Harbour House, Saltspring Brolly et al. 1993:72, Table 8 0 1 0 0 0 0 0 127 129 58.1 ecVI DiSe-10 300-1200 Denman - Southern Rckshltr Eldridge 1987:5-57, Table 5-9 0 0 0 0 0 0 0 7 7 58.2 ecVI DiSe-10 2400-5000 Denman - Northern Rckshltr Eldridge 1987:5-59, Table 5-10 2 0 0 0 0 0 0 57 59 58.3 ecVI DiSe-10 2400-5000 Denman - Blufftop Hunting Ritual Eldridge 1987:5-67, Table 5-16 1 0 0 0 0 0 0 26 27 * ecVI-East Coast Vancouver Island, scVI-South Coast Vancouver Island, Gulf Is-Gulf Islands, wcVI-West Coast Vancouver Island, MnInd-Mainland of British Columbia, WA-Washington State. – 77 –  Age and Sex Data Due to the striking disparity between the archaeological and modern distributions of northern fur seal, age and sex data for this once commercially significant species have been discussed elsewhere in great detail (Burton, et al. 2001; Calvert 1980; Crockford, et al. 2002; Etnier 2002a, 2007; Gifford-Gonzalez, et al. 2005; 2011; Gustafson 1968; Moss, et al. 2006; Newsome, et al. 2007). These studies have identified adult males and females as well as immature pups less than 6 months of age in a variety of sites in the study area. Combining archaeological data from Ts’ishaa (Site 42) with historic accounts from schooner captains, aboriginal hunters, and merchants involved in the sealing industry of the 1880s (Swan 1883, 1887), Crockford et al. (2002:170) argue for the presence of a non-migratory breeding population of fur seals capable of giving birth in kelp-beds or kelp rafts at sea. Subsequent studies have used stable isotopes to posit the existence a mid-latitude breeding distribution in the Pacific Northwest (Newsome, et al. 2007). However, definitive evidence of a terrestrially-based northern fur seal rookery remains to be identified.  Age and sex data from few species other than fur seal have been compiled. For sites in the study sample, a small portion of Steller sea lion skeletal remains from a limited number of sites has been confidently sexed. Table 3.3 shows the breakdown of sexed elements for sites where the information was available from controlled excavations. Although the sample is small (NISP=19), 95% of the sexed elements are from adult males with only two elements out of 40 either unknown or female. This pattern is similar to data from the Oregon Coast, which indicates that Steller sea lions from archaeological contexts are primarily adult males (Lyman 2003b). The incidence of females and/or juveniles of either – 78 –  sex are very low with only one definitively female element found on southeast Vancouver Island (Table 3.3).  Additional Steller sea lion sex and age information comes from fauna collected from archaeological contexts during monitoring of construction projects (Table 3.3). Monitoring samples are collected without screening and have only general provenience. However, males dominate four large assemblages from the southern Strait of Georgia region. For example, at Patricia Bay (Site 18), at least 74 male Steller sea lion elements have been identified in contrast to only one female and three juveniles of indeterminate sex (Kanipe, et al. 2007; Wigen 2007b:24). Explanation for this pattern of male dominance could relate to seasonal availability of Steller sea lions and the differential behavior of the females and males at rookeries and haulout sites and/or preference on the part of the human hunters.  Steller sea lions breed at communal rookeries in northern British Columbia during the summer but spend the winter at haulout sites throughout British Columbia (Olesiuk 2009a). Since the first survey in British Columbia in 1913 (Newcombe and Newcombe 1914), the most southerly sea lion breeding rookery is located on the north end of Vancouver Island. Non-breeding animals of both sexes use a limited number of established year-around haulouts found mainly on the west coast of Vancouver Island. In the winter, individuals of both sexes use a wider range of haulouts including the Strait of Georgia and southern Vancouver Island. This pattern suggests both males and females should be available to hunters in the winter, so it appears seasonal distribution does not explain the hunting pattern. Recent studies of the behavioral responses to human disturbance of hauled out Steller sea lions note that females and pups are the first to leave followed by the sub-adult males and, finally, the bulls (COSEWIC 2003; Szaniszlo 2005). The dominance of males in the – 79 –  archaeological sites may be a result of this behavioral pattern, with the males being easier for the hunters to target. Additionally, haulouts of ‘bachelor males’ are commonly present near rookeries and may have also been targeted by hunters. The extremely large size of males and these behavioral characteristics suggest that males were hunted while onshore rather than offshore as documented for fur seals. Table 3.3 Sex Distribution of Steller Sea Lion Remains from Selected Sites. Region* Site Total NISP Male Female Unknown Reference ecVI DiSe-7, Deep Bay 1 1     Wilson et al. 2004b:44 scVI DcRt-16, King George Terrace 4 4     Wilson et al. 2004a:94 scVI DdRu-5, Pat Bay 10 10     Kanipe et al. 2006:269 scVI DcRu-136, Esquimalt Lagoon 2 2     Wigen 2002 Gulf Is DfRv-106, Galiano Island 1 1     Mason et al. 1995:22  wcVI DiSo9-II, Hesquiat 1   1   Calvert 1980:155 wcVI DiSo1-I, Hesquiat 13 2   11 Calvert 1980:156 wcVI DiSo1-II, Hesquiat 1 1     Calvert 1980:157 wcVI DiSo1-III, Hesquiat 12 3   9 Calvert 1980:158 wcVI DiSo1-IV, Hesquiat 14 2 1 11 Calvert 1980:159 wcVI DjSf100, Little Beach  1   1   Wigen 2008:11 wcVI DfSi-26, Clarke Island 7 4   3 McKechnie 2007a:30 wcVI DfSi-16, Ts’ishaa 20 7   13 Unpublished data in Frederick and Crockford 2005 and McKechnie 2007 wcVI DfSh-7, Huu7ii 9 7 2   Unpublished data in Frederick et al. 2006   Total  96 44 5 47     % Sexable 90% 10%   * ecVI = East Coast Vancouver Island, scVI = South Coast Vancouver Island, Gulf Is = Canadian Gulf Islands, wcVI = West Coast Vancouver Island,   – 80 –  Harbor seals and sea otters are more difficult to sex, as sexual dimorphism is less extreme. In most cases, no information has been recorded about the sex of individuals. Age for harbor seals has been recorded in a few site reports (e.g., Frederick and Crockford 2005; Frederick, et al. 2006a; Wigen 2007b), indicating that harbor seals of all ages were regularly hunted in all areas. No age or sex information is available for the few identified California sea lion or elephant seal elements. Steller sea lions are present in sites along the southwestern portion of the study area, suggesting they may have been moderately abundant in the ancient environment. Elephant seals, California sea lions, and Guadalupe fur seals rarely occur in the assemblages, suggesting they were not commonly present at this latitude. However, taxonomic identifications of male California sea lions may be confused with female Steller sea lions (personal observations) and similarly, Guadalupe fur seals appear similar to northern fur seals (Etnier 2002b).  Discussion This analysis has identified distinct regional patterns in marine and terrestrial mammal assemblages from archaeological sites on western Vancouver Island and the Strait of Georgia. These differences provide a basis for inferring that aboriginal hunting practices dramatically differed in focus between these two regions. In particular, aboriginal peoples in the Strait of Georgia region did not hunt marine mammals to the same degree as was practiced by aboriginal peoples along the Pacific coast of Vancouver Island and Washington state. Rather, terrestrial mammals appear to be relatively more important in the Strait of Georgia region and relatively less so on the exposed Pacific coast.  – 81 –  These geographic patterns are robust at least over the past 5,000 years and have implications for a number of perspectives in anthropology and ecology. Of particular anthropological relevance is the apparent coherence and continuity of aboriginal marine and terrestrial hunting patterns within each of the two areas. This difference is most strongly apparent in the relative composition of marine versus terrestrial mammals (Figure 3.2 and Figure 3.3) and among the individual pinniped species (Figure 3.4). For specific localities within these two regions, variation in marine mammal use also appears to be spatially patterned, such as the slightly higher abundance of marine mammals in assemblages along southeastern Vancouver Island and in the Gulf Islands (Figure 3.2). These archaeological patterns are an outcome of the direct participation of generations of aboriginal people in this marine ecosystem. Such patterns reflect a continuity of skillful cultural knowledge and practical engagement with these marine environments. These regional differences are consistent with a host of ethnographic accounts and observations (e.g., Boas 1887; McMillan 1999; Suttles 1952), which identify the aboriginal peoples who occupy these two regions as belonging to two culturally and linguistically distinct groups (Wakashan speaking peoples [Nuu-chah-nulth, Ditidaht, Makah] and Salish speaking peoples). During the post-contact era (ca. AD 1774-1860) and continuing among First Nations today, these cultural differences include distinct patterns of residence, conventions of artistic representation, ritual practice, and a host of other regionally unique forms of cultural expression (ibid.). Our analysis provides insight into an aspect of these rich cultural histories, suggesting that cultural distinctions between regions and cultural similarities within regions have great antiquity in regards to hunting practice. Indeed, since our data relate to an important subsistence activity (hunting), such regional expressions can – 82 –  be hypothesized to have contributed structure to cultural history and community identity. This is illustrated in a 1922 quote discussing the biography of Tom Sayach’apis (ca. 1835–1927), a respected cultural historian of the Tseshaht, a Nuu-chah-nulth Nation on southwestern Vancouver Island: Tom ate very little meat of land mammals in his early days. Indeed, like most of the [west] Coast people, he had a prejudice against deer meat and it was not until as a middle-aged man, he had come into contact with some of the deer hunting tribes of the interior of the island, that he learned to prize it, though to this day, venison has not for him the toothsome appeal of a chunk of whale meat (Sapir 1922:304). From an ecological perspective, such strong geographic patterning as seen through a lens of human hunting activity is relevant for considering the long-term ecological role of marine mammal species in the study area. Our data represent a unique and otherwise unattainable record of marine mammal abundance and distribution over the Holocene. This information helps contextualize contemporary marine mammal ecology by greatly extending temporal and geographic knowledge of these important species. In this regard, it is notable that the three most abundant pinnipeds in the archaeological assemblages (northern fur seals, harbor seals, and Steller sea lions) are also the three most numerically abundant in the contemporary environment (Table 3.1). The similarity in rank-ordered abundance between our archaeological data and contemporary pinniped population estimates indicates these three species have exhibited remarkable demographic resilience to human harvesting.  The stable long-term trends observed in the archaeological data stand in contrast to the dynamic historical events documented over the past 200 years which resulted in the rapid extirpation of the sea otter (ca. AD 1774-1811) and culling of pinniped populations (ca. AD 1868-1970). In this context, it seems important to note that the sudden cessation of marine – 83 –  mammal hunting over the past 40 years appears to be unprecedented in a record spanning at least the past 5,000 years. The closest possible parallels to this sudden reduction in hunting intensity and corresponding population increase likely occurred after European diseases dramatically reduced aboriginal populations in the early contact era (ca. AD 1774-1811) and again in the mid-19th century, just prior to the beginning of the commercial fur seal industry (Boyd 1999).  While it is striking to note the ongoing demographic consequences of historic-era commercial hunting, equally important are possible behavioral and ecological changes that occurred among targeted taxa over the Holocene. For instance, our compiled archaeological results indicate that aboriginal peoples regularly hunted fur seals for a minimum of 5,000 years on the exposed Pacific coasts of Vancouver Island and northwestern Washington. These observations correspond with the locations of fur seal hunting mentioned in historic-era harvest log-books (Olesiuk 2009b:42) indicating that the continental shelf off southwestern Vancouver Island has long been a major focus of fur seal foraging activity. However, many of these same historic-era harvest records (ca. AD 1891-1911) also indicate a strongly seasonal presence of fur seals at this latitude between January and May (Olesiuk 2009b:54), which contrasts with zooarchaeological data including osteometric evidence for localized breeding (Crockford, et al. 2002; Etnier 2002a; Gustafson 1968), and geochemical evidence for non-migratory mid-latitude foraging and prolonged pup weaning dating from the mid-Holocene (Newsome, et al. 2007). Tantalizingly suggestive historic accounts compiled from interviews with commercial fur sealers conducted a decade prior to the historic log-book-based records mentioned above, describe non-migratory behavior involving giving birth at sea in kelp beds and observing fur seals in the Strait of Georgia during late-– 84 –  summer when the migratory populations are in the Bering Sea (Crockford et al. 2002 citing Swan 1883). These data suggest historic-era commercial harvests may have eliminated or altered a behaviorally unique sub-population of fur seals and archaeological data yield important insight into the previously more diverse ecology of this currently strictly pelagic pinniped. Although the absolute number of animals harvested cannot be quantified with our zooarchaeological data alone, our analysis has several implications concerning potential human impacts on pinniped and sea otter populations. Firstly, our data indicate a regional focus on specific pinniped species, which implies that aboriginal peoples in these regions targeted specific taxa with regularity (e.g., annually or seasonally) as suggested by ethnographic accounts (e.g., Arima 1988; Suttles 1952). Secondly, our data represent only a small percentage of the approximately 4,000 currently documented coastal shell midden sites present in the study area (British Columbia Archaeology Branch n.d.). Thus, the collective impact of aboriginal hunting may be considerably greater than previously recognized if these regional patterns apply to several thousand unexamined sites. Finally, the temporal and spatial consistency of hunted marine mammals in these two regions indicates a degree of stability and continuity to aboriginal resource harvesting practices, which demonstrates the capacity for sustained harvesting in antiquity (Etnier 2007).  Repetitive and consistent human participation in an ecosystem provides structure to ecological interactions (Balée 2006; Hobbs and Fowler 2008; Liu, et al. 2007). However, determining and/or identifying ‘human impacts’ in such an archaeological context is challenging as these ecosystems may represent the end-point of centuries or millennia of human hunting. Future analyses that can incorporate more fine-grained osteometric – 85 –  information on age and sex distribution will yield much greater insight into these ecological impacts. Considering the ecological dimensions of human subsistence activity in coastal environments, Fedje et al. (2004) offer a distinction that seems to encompass the regional differences we observe in our analysis of mammalian hunting practice; the concept of “maritime” and “coastal adaptations.” These authors define a “maritime adaptation” as on which is:  “heavily reliant upon marine and coastal resources for the majority of subsistence needs. In contrast, a ‘coastal adaptation’ involves the fluent use of coastal and marine resources for at least some, and probably a significant portion, of the subsistence needs of a group of people. Coastal adaptations should therefore include the possibility of considerable use of terrestrial resources and inland areas” (Fedje, et al. 2004:112). This synthetic statement helps characterize the regional differences observed in our analysis suggesting that aboriginal peoples in the Strait of Georgia ‘adapted’ themselves to terrestrial mammal hunting more extensively than aboriginal peoples on western Vancouver Island. This observation is consistent with ethnographic and archaeological information from Coast Salish peoples in the Strait of Georgia region where widespread aboriginal burning practices aimed to enhance habitat for economically important plants such as camas and berries as well as entice foraging ungulates into recently burned areas with young re-growth (Brown and Hebda 2002; Suttles 1987; Weiser and Lepofsky 2009). Conversely, on western Vancouver Island, the logistics of hunting pelagic pinnipeds likely facilitated other forms of offshore subsistence activities (e.g., deep water fishing or birding) as well as voyaging and trading or vice versa (Arima 1988; McMillan 1999).  – 86 –  However, while the concept of “adaptation” (defined as ‘the suiting of one thing to another’) may help characterize human hunting activity expressed over thousands of years, a term that is more appropriately scaled to human lifetimes is ‘tradition’ (defined as ‘the act of handing down’). Considering long-term hunting practices as ‘traditions’ accommodates the intergenerational transference of complex hunting knowledge in a way that anthropological uses of ‘adaptation’ often under-specify or assume outright (cf., Moss 2008). The regionally consistent patterns of terrestrial and marine mammal use were additionally constrained or enabled by environmental differences between the two regions. For instance, the open prairie oak savanna woodlands of the Strait of Georgia versus the very moist dense coastal rainforest on western Vancouver Island (Suttles 1987) or the higher primary productivity on the exposed Pacific coast versus the Strait of Georgia (Ware and Thomson 2005). However, while it is essential to consider the environmental factors that may contribute how aboriginal peoples may have focused on marine or terrestrial mammals, these factors cannot adequately explain how or why generations of people choose to participate in an ecosystem in a particular manner. Rather, explanations that posit an environmental imperative for human action risk imposing a deterministic assumption about human actions when that is not the case. Thus, although the relative contribution of hunting was undoubtedly constrained or enabled to some degree by the presence or absence of animals in different regions within the study area, human agency cannot be ignored. Rather than a simple consequence of available resources, social factors critically underpin the hunting practice in ways that may be much more meaningful than the simple availability of resources. Hunting was a conscious decision first and then a reflection the environmental availability. – 87 –  In contrast to an environmentally determined model of human subsistence, aboriginal peoples the Northwest Coast are increasingly recognized as highly active in “managing” subsistence resources through explicit cultural definitions of harvesting practice and etiquette including selective harvesting, seasonal restrictions on use and/or consumption, and proprietorship over resources which was contingent on sustained productivity (Berkes and Turner 2006; Hunn, et al. 2003; Moss 2011a; Trosper 2009; Weiser and Lepofsky 2009). The data presented here suggest that generations of people actively chose to pursue an elaborate, highly structured, and specialized activity – primarily pelagic marine mammal hunting or primarily terrestrial mammal hunting in different regions. These hunting traditions embody the skill, training, and cooperation, which develop over years, and decades among groups of individual hunters and which are passed on to subsequent generations in a community of continuous social practice (cf., Bourdieu 1977). The consistency and continuity with which certain species were utilized in the archaeological assemblages over 5,000 years suggests that similar cultural practices informed aboriginal peoples’ hunting behavior in the archaeologically documented past. Conclusions We posed three questions at the beginning of this chapter: • What marine mammals did aboriginal people in Southern British Columbia most commonly utilize? • How similar or how different are species occurrences and proportions relative to today? • Is there evidence of specialized or regional hunting traditions and if so, what might have been the potential impacts of these activities on the ancient marine ecosystem?  – 88 –  In answer to the first question, fur seals, harbor seals, and Steller sea lions respectively appear to be the most heavily utilized pinnipeds on the southern BC coast over the last 8,000 years. Sea otters appear to be utilized by peoples on western Vancouver Island and the Olympic Peninsula, and are present only in low numbers at several sites in the Strait of Georgia. The least utilized pinnipeds appear to be California sea lions, elephant seals, and Guadalupe fur seals, which are rare but are found in sites along the southwestern margin of the study area. These are relative differences however, and thus even small relative percentages may represent large numbers of harvested animals. In regards to the second question, our data expand knowledge of the range and relative composition of marine mammal species that are currently threatened or endangered (Table 3.1). Examining the archaeological distribution of these species provides context for understanding their recovery since historical over-exploitation and the extent to which aboriginal peoples may have influenced the marine ecosystem before the historic-era fur trade. Considering the fact that the 58 examined sites represent less than 1% of the approximately 4,000 currently documented coastal shell midden sites present in the study area (British Columbia Archaeology Branch n.d.), the perspective gained here hints at the potential cumulative influence of large indigenous human populations in the region. The extent of this influence is unknown at this point, but did not appear to have reached the stage where people dramatically change their hunting patterns. Such basic insights help refine our understanding of the influence of human activity on the environment and, conversely, the influence of animals on human activity. And finally, this study indicates that aboriginal people on the outer coast of Vancouver Island and Washington state developed a specialized hunting tradition targeting – 89 –  northern fur seals. This conclusion is further supported by several ethnographic and historic sources (e.g., Arima 1988; Sapir and Swadesh 1955; Swan 1887; Waterman 1920) and by artifactual data from pre-contact archaeological sites in the study area (cf., McMillan 1999; Samuels and Daugherty 1991). Ethnographically, such hunting activity involved paddling considerable distances offshore, sometimes beyond sight of land, to stealthily approach ‘sleeping’ fur seals to within range for launching a harpoon. Such specialized journeys entailed considerable physical risks and logistical preparation. Ethnographic accounts emphasize how participants ritually engaged in elaborate physical and spiritual training and how hunters utilized specifically designed canoes, sharpened paddles, unique harpoons, yew-wood throwing sticks, seal skin floats to tire harpooned animals, as well as long lengths of cedar-bark and elk-hide rope (Arima 1988; Sapir and Swadesh 1955; Waterman 1920).  In contrast to western Vancouver Island and northwestern Washington, aboriginal peoples in the Strait of Georgia appear to have developed marine mammal hunting traditions that targeted harbor seals and to a lesser extent Steller sea lions. As discussed previously, harbor seals are non-migratory, nearshore foragers who regularly ‘haul-out’ on land and are the most common seal present on the coast today (Department of Fisheries and Oceans Canada 2010; Jeffries, et al. 2003). Ethnographic accounts describe hunting ‘hauled-out’ harbor seals from boats, by hand with clubs, or by stealthily placing nets in the water below haul-out areas, particularly outside sea caves (Suttles 1952; Waterman 1920). Steller sea lions, on the other hand, occur seasonally in the Strait of Georgia during winter and spring when herring aggregate prior to spawning (COSEWIC 2003). Ethnographically, sea lions were hunted by numerous Coast Salish peoples although a few groups were recognized as particularly skilled specialists (Suttles 1952). Ethnographic descriptions of hunting both – 90 –  Steller sea lions and harbor seals were associated with elaborate forms of physical and spiritual preparation, which occurred prior to and during the hunt and were observed by both hunters and their wives (Elmendorf 1960; Suttles 1952). While hunting practices documented in archaeological contexts represent only a small portion of the rich ancestral pasts of coastal First Nation communities, zooarchaeological data represent a readily available source of information that broadens our contemporary understanding of ancient coastal lifeways and cultural environments. Such ecologically significant marine data are essentially non-existent outside of archaeological contexts and future research will benefit from additional analyses of new sites, regions, and time periods as well as more detailed osteometric observations. The compilation of zooarchaeological information explored in this paper suggest coherent patterns indicating the persistence of hunting traditions practiced continuously over 5,000 years. These patterns help unravel the complex interwoven cultural and ecological histories and environments on the Northwest Coast.    – 91 –  Chapter 4. Re-Calibrating Archaeological Chronologies on the Northern Northwest Coast: Radiocarbon Data from Prince Rupert Harbour5  This chapter investigates the methods by which radiocarbon dating and radiocarbon calibration has been practiced and interpreted by archaeological researchers on the northern Northwest Coast. We note that past efforts to report and interpret archaeological chronologies have used a variety of approaches and we suggest this affects the current understanding of the region’s human past. This paper attempts to add clarity to the process of radiocarbon dating and radiocarbon calibration with a focus on interpreting radiocarbon dates obtained on marine or marine-influenced organisms (particularly marine shell and bone collagen). We first discuss the process of radiocarbon dating and radiocarbon calibration, highlighting some of the complexities and nuances of the method. We then discuss three important radiocarbon datasets that underpin the archaeological chronology of social history and culture change in the region; 1) terminal village occupation dates on shellfish compiled by Archer (2001), and 2) human burial dates compiled by Cybulski (1975), and 3) this same dataset reanalyzed by Ames (2005). We then compile a database of currently existing paired marine and terrestrial dates to derive a regionally specific marine reservoir estimate (Delta-R or ∆R) that allows us to integrate these different datasets into the same chronological timescale.  Our resulting re-calibrations refine previously identified chronological and settlement patterns and in some cases shift the understood age of past events by up to 500 years. These chronological shifts                                                 5 This chapter is co-authored with Morley Eldridge. See preface for more information. – 92 –  and patterns have implications for archaeological understandings of the antiquity of social complexity, warfare, and settlement patterns within the region. The Importance of Calibration in Radiocarbon Dating  Radiocarbon dating is the most widespread method for evaluating the chronology of archaeological phenomena within the past 40,000 years. However, many levels of uncertainty underlie this chronological method and these are not always well understood by archaeologists (Bronk Ramsey 2008). Most archaeologists are familiar to a degree with how radiocarbon dating works – it is a method that measures the rate of decay of radioactive carbon isotopes (denoted as 14C) relative to stable carbon isotopes (13/12C), a measurable difference that begins when organic tissue stops growing – hence the cessation of carbon uptake.  The half-life of radiocarbon is approximately 5,730±40 years, meaning that half the radioactive carbon (14C) present in a living organism will have decayed in this time (Godwin 1962). Radiocarbon dates provide a result in ‘years before present’ and an instrumental measurement error (e.g., 4,500±50 14C yr bp)6. However, a crucial aspect of radiocarbon dating is that radiocarbon time (i.e., radiocarbon years) is not equivalent to conventional time (years measured by the astronomical 365 day-long year). Thus, calibration is necessary to convert radiocarbon years into historically equivalent and comparable measures of time. This is because radiocarbon ‘years’ (14C yr bp or RYBP) are variably affected by changes in the                                                 6 We encourage researchers to follow international convention when reporting normalized radiocarbon dates (“radiocarbon years bp” ± error), and calibrated dates (“cal yr BP” or “cal yr BC/AD”) along with a one or two sigma age range. We also encourage researchers to submit their dates to both the Canadian Radiocarbon Database (CARD) and any shell wood pair data to the CHRONO Marine Database. The greater the available information, the greater the chance our collective archaeological knowledge will appropriately incorporate both the ambiguity and the qualified certitude in our regional archaeological chronologies (cf. Gero 2007).   – 93 –  global carbon cycle, which is a product of global climate change and solar radiation that influences the concentration of 14C in the atmosphere and ocean at different points in the earth’s history (Guilderson, et al. 2005). This fluctuation is particularly influenced by changes in the global carbon cycle such as forests expanding following deglaciation, major changes in ocean circulation, increased carbon emissions during the industrial revolution, atomic bomb testing, and ongoing land clearance and burning of fossil fuels. Radiocarbon calibration (often known simply as calibration) is a probabilistic method for reconciling the variation in 14C with conventional time, thereby allowing for a measurement of radiocarbon years (14C yr bp or RYBP) to be converted into an age estimate in chronological years before present or calendrical (BC/AD) years (Bowman 1990). Due to the uncertainties inherent in the dating method (measurement error) and the uncertainties of past global carbon (calibration error) the resulting value of a given calibrated radiocarbon age is more appropriately considered an age-range within which a dated sample has a probability of occurring (rather than a single number with a standard error). Since radiocarbon varies depending on global climatic change, the same period in time can be represented by different amounts of 14C variation and vice versa. Calibration of radiocarbon years is further complicated for samples taken from marine or marine influenced contexts; these require a second form of calibration for the marine reservoir effect discussed at length in a following section. Calibrating Radiocarbon Age Estimates  While the radiocarbon method was invented in the mid-20th century (Libby, et al. 1949), it was not until the 1980s that radiocarbon calibration ‘curves’ became commonly – 94 –  available for use in archaeology (Bowman 1990). The most widely used calibration curve developed from an increasingly refined series of dendrochronologically dated tree-ring sequences constructed from the Pacific Northwest (Stuiver 1982) and preserved in peat bogs in Germany and Ireland (Pearson and Stuiver 1986; Stuiver and Pearson 1986; Stuiver, et al. 1998). Researchers radiocarbon dated individual tree rings, each of which had a known-age and compared the difference between the measured radiocarbon age and the actual age to produce a graphical representation of this relationship (Figure 4.1).  This ‘curve’ essentially documents the concentration of atmospheric 14C as preserved in the individual tree growth rings. The most crucial feature of the curve is the overall rate of 14C decline over time but also the undulations that occur over shorter-time periods (due to variations in atmospheric 14C which is influenced by variations in solar activity and other processes of global carbon circulation) and the varying thickness of the curve itself that reflects the relative confidence that a date of a certain age falls within a calendar age-range (due to measurement limitations). These features are discussed in two short examples. – 95 –   Figure 4.1 A 2,400 year-long section of the atmospheric radiocarbon curve covering the period between 1,400 and 3,000 years before present (Reimer et al. 2009). Figure modified from a graphical output of CALIB 6.1 (Stuiver and Reimer 1993). Variability in Radiocarbon Years Consider a radiocarbon date obtained from the leaf of a terrestrial plant (one that derived its carbon from the atmosphere), yielding a measure of 4000 ± 25 radiocarbon years before present (14C yr bp). When calibrated using the 2009 radiocarbon curve7 at a 95.5%                                                 7 The internationally recognized radiocarbon curve is refined approximately every four years at an international meeting. The latest curve (INTCAL13) was published September 22, 2013 (Reimer, et al. 2013) and has not yet been incorporated into the dates reported in this dissertation.  – 96 –  confidence interval (i.e., the 2-sigma calibrated range8), the resulting age estimate falls between 4,520 and 4,080 years ago. Thus, the process of calibration adds between 80-520 calendar years to this particular radiocarbon age and also significantly increases the apparent uncertainty (from ±25 to up to ±220 years). In addition, differences between radiocarbon and calibrated radiocarbon ages vary over time as a result of changes in the shape of the curve.  Variable Uncertainty in the Atmospheric Calibration Curve Consider another date from a terrestrial plant leaf, 9,500 ± 25 14C yr bp. When calibrated at a 95.5% confidence interval, this date has a calibrated age-range estimate of between 10,670 and 11,070 years ago. Thus, in contrast to the example above, this calibration increases the numeric difference between the radiocarbon date and calibrated age range by 1,170 to 1,570 years. The statistical uncertainty of this calibrated sample at the a 95.5% confidence interval is slightly less, approximately ± 200 years. It could be reduced further by examining the fluctuations in the curve for the period of time in question as shown in Figure 4.2. These transformations and increased uncertainties are integral to translating radiocarbon time into conventional (calendrical/astronomical) time and illustrate how some periods in time have increased uncertainty regardless of the precision of individual radiocarbon measurements (Guilderson, et al. 2005).                                                 8 Probability is simply an expression of the likeliness of something occurring. The convention in analytical statistics is to calculate uncertainty in terms of standard deviations (which can be approximated as the average difference of a set of values from its average) often called ‘sigmas’ or ‘standard deviations’. In a normal distribution (where there is a central tendency to the values) 1-sigma encompasses about 67% of the probability and 2-sigmas encompasses about 95%. Sigmas are thus notations for quantifying confidence and are sometimes displayed using the symbol σ. In describing the 2-sigma calibrated range, this is specific to the calibration probability and does not refer to the uncertainty for the individual date. – 97 –   Figure 4.2. Example of calibration effects on a radiocarbon date with a large error range. The vertical height of the ‘calibrated age-range’ shown on the x-axis represents the relative probability of the date occurring during a particular period.  ‘Correcting’ for Isotopic Fractionation In addition to the fluctuations in the global carbon cycle, radiocarbon measurements are further influenced by molecular transformations as carbon is differentially incorporated into organic tissues and ‘hardparts’ such as bones and shells. These alterations have a smaller magnitude than the differences between marine and terrestrial carbon, but can still result in an age estimate being shifted by hundreds of years. Since the carbon-14 isotopes are slightly heavier (because of their additional neutrons), they have a greater inertia and react slower – 98 –  than carbon 12 (12C) and carbon 13 (13C) isotopes (Stuiver and Polach 1977).  In other words, a plant that absorbs atmospheric carbon will have a particular carbon 14 value; but that same carbon 14 will be slightly depleted when absorbed by a herbivore consuming plant tissues. Likewise this difference would be slightly larger in a carnivore that regularly consumes herbivores that in turn regularly consumes plant tissues and so on. Thus, even though these different organisms may have lived at exactly the same time, the 14C levels in their tissues will exhibit slight differences, thus requiring further adjustment based on their tropic level. Fortunately, these differences in normalization have been determined by measuring the ratio of the stable carbon isotopes (13C to 12C) in a variety of organisms (Stuiver and Polach 1977). Since 13C and 12C stable isotopes do not change over time, the measured difference is exactly half that of the difference between the 12C and the 14C atoms, allowing for an accurate estimate of 14C in the original organism from the measured levels of 13C to 12C in the sample. Although direct measurements are often not available for samples analysed prior to the 1970s values can be retroactively estimated if the type of dated organism is known (e.g., 0.0 for shellfish, -25.0% for terrestrial charcoal)9 (Stuiver and Polach 1977) but see Southon ( 2011) for further discussion.                                                 9 Most modern radiocarbon laboratories tend to provide a direct measurement of the C12/13 ratio prior to processing a sample so that the value can be factored into the normalization formula but exceptions still exist. In those cases, without directly measuring the isotopic fractionation, radiocarbon samples are ‘normalized’ according to assumed values (Stuiver and Polach 1977) which for C3 plants is typically -25.0 and for C4 plants ranges between -9 and -12 . Assumed values are commonly indicated by a lack of a decimal place (e.g., -25 rather than -24.5).   – 99 –  Calibrating Radiocarbon Age Estimates for Marine Samples The global carbon cycle has two primary domains— the atmosphere and the ocean —within which carbon circulates at dramatically different rates. For instance, atmospheric carbon circulates virtually instantaneously throughout the globe but marine carbon circulates at much slower rates as much of the carbon is ‘trapped’ in deep-ocean currents for hundreds of years before it is released once again into the biosphere through upwelling.  This temporal lag in marine radiocarbon is known as the marine ‘reservoir’ (Stuiver and Braziunas 1993). On a global scale, marine carbon circulates an average of 405 years behind the rate of terrestrial carbon (Hughen, et al. 2004) but there is considerable variation regionally and over time. The difference between the global average and the local average of marine reservoir time is known as the ∆R (hereafter Delta-R). Thus, marine or marine influenced samples require additional calibration compared with terrestrial/atmospheric samples, as shown by the following example.  Example 3 - Calibrating a Marine Date Consider a date on a marine clamshell from the same archaeological deposit as the 4000±25 14C yr bp plant leaf discussed in Example 1. Because this clam absorbed its carbon from the marine environment (via filter feeding on marine plankton), the radiocarbon age will likely appear significantly ‘older’ in radiocarbon time than a contemporaneously deposited terrestrial sample (e.g., wood charcoal). Thus, estimating the age of this ‘marine date’ requires additional calibration for the local marine reservoir, incorporating both the global reservoir value (R) and the Delta-R value, the values for which are derived from two methods discussed below. – 100 –  Method 1: Dating Modern Known-Age Shells Collected in the Historic Era  The first of the two principal methods for estimating the age of the marine reservoir is to radiocarbon date living shells that were collected from a known place on a known date prior to the commencement of atomic bomb testing in the 1940s (which significantly disrupted the global carbon cycle). Because oceanic currents are relatively stable, this method provides a historical estimate of the local marine reservoir (Delta-R) that can be applied to radiocarbon calibration of marine samples. However, marine reservoir can be particularly variable in large estuary environments where there is considerable influx of freshwater from rivers, which dilutes (and thus lowers) the marine carbon absorbed by organisms living in those environments (Rick, et al. 2012).  Conversely, deepwater fjords in proximity to glacial-fed rivers and tidewater glaciers may contain significant quantities of ‘old’ meltwater ‘trapped’ in glacial ice for thousands of years (hence older carbon) (Hutchinson, et al. 2004). On the Northern Northwest Coast a recent study (McNeely, et al. 2006) has dated a series of historic marine shells collected from Haida Gwaii and Southeast Alaska prior to atomic testing in the mid-20th century (Figure 4.3). For example, they obtained a radiocarbon date on a giant razor clam (Ensis siliqua) collected in AD 1937 from Masset on northern Haida Gwaii that returned a radiocarbon age of 850±40 14C yr bp. Thus, even though this clam lived during the early 20th century (and died in 1937), it absorbed ‘older’ marine carbon that had been circulating in the ocean for approximately 690 years. This 690-year difference includes the global average of marine carbon circulation of 405 years plus 285 years for the local marine reservoir effect. This local difference is due to ‘upwelling’ of ‘older’ carbon as deep oceanic currents collide with the continental shelf immediately west and south of Haida Gwaii and then circulate throughout Hecate Strait (Thomson 1981).  – 101 –  McNeely et al.’s (2006) data are relevant to the interpretation of archaeological chronologies based on marine shell dates but unfortunately lack observations from Prince Rupert Harbour. Nevertheless, results from this study reveal regional differences and sub-regional variability in the marine reservoir age for both Southeast Alaska and Haida Gwaii (Figure 4.3). Of particular significance are the comparatively low Delta-R values from Haida Gwaii relative to deep-water fjords and channels of Southeast Alaska, which have Delta-R values consistently above 400 (Figure 4.3). This is likely a reflection of several large active glaciers and ice fields in coastal Southeast Alaska releasing particularly old freshwater carbon into the deep and potentially slowly circulating coastal fjords. While the sample remains small, the Southeast Alaska data also indicate a west-to-east trend of increasing Delta-R values closer to the mainland coast (Figure 4.3).  Currently, there are no available data for Prince Rupert Harbour, thereby limiting the direct utility for interpreting marine influenced radiocarbon ages for sites in this area. A further complicating factor in the use of this method is that oceanic circulation patterns during the early 20th century or historic era may not be very representative of marine circulation during previous millennia. On the Northwest Coast, the early 20th century is the ‘end’ of the ‘Little Ice Age’ and is associated with significant receding of glaciers and an increase in associated meltwater (Menounos, et al. 2009). – 102 –   Figure 4.3. Map of the northern Northwest Coast showing the location and local marine reservoir age (Delta-R value) for radiocarbon dated shellfish of known age collected from the northern Northwest Coast prior to atomic testing in the mid 20th century (data from McNeely et al. 2006). Also shown are the locations of late 20th century glaciers (data from  Method 2: Dating ‘Shell-Wood’ Pairs from Archaeological and Paleoecological Contexts The second method for obtaining a local estimate of marine reservoir age is to date contemporaneously deposited marine and terrestrial carbon samples to identify the marine reservoir age from a pre-modern context by the difference in age between the two samples. Thus, in contrast to dating early 20th century marine shell discussed above, this method can assess the magnitude of the marine reservoir prior to the modern era and therefore, is more appropriate for assessing the marine reservoir during the archaeological period of interest. – 103 –  This method consists of estimating the total marine reservoir age at particular points in the past (expressed as R[t]) by dating depositionally ‘paired’ terrestrial and marine samples from coastal archaeological or paleoecological contexts and subtracting the global reservoir estimate from the combined age difference. This method is appropriate only if marine and terrestrial organic remains are short-lived and were contemporaneously deposited in archaeological contexts such as rapidly accumulating shell midden or a paleoecological context such as wave cut terrace or delta (Southon, et al. 1990; Southon and Fedje 2003).  Because archaeological and paleoecological deposits contain long temporal records, this method allows for the assessment of changes in reservoir age at a variety of temporal and spatial scales (Ingram and Southon 1996). For instance, Kennett et al. (1997) dated multiple paired archaeological marine and terrestrial samples from southern California and estimated temporal fluctuations in the reservoir age over the past 10,000 years that could be linked to changes in paleoclimate and upwelling intensity. This in turn allowed for a more refined chronology of the palaeoceanography and the pre-contact cultural dynamics of pre-colonial Southern California (Kennett and Kennett 2000).  The drawback to this method is that geological and archaeological researchers must be very attentive to the strength of the stratigraphic association for each sample and recognize the potential interpretive hazards of each data point.  In addition, the age difference between samples is less precise than Method 1 because both must rely on age estimates rather than known dates. To recap, the total marine reservoir is a combination of the global reservoir and the local correction (Delta-R) and can be estimated over time and regionally. The local component of reservoir correction is expressed as Delta-R. Considering our 4600±25 14C yr – 104 –  bp shell date example once more, if we know that the Delta-R for this period is 350±50 radiocarbon years for this point in time, it is possible to more accurately estimate the calibrated age of this shell using online calibration software (e.g., CALIB). This sample produces an age range estimate between 4180-4510 calibrated years BP (at 95.5% probability) and therefore appears more recent than the original measured age and is much more consistent with the original 4,000 ± 25 14C yr bp estimate on the (presumably) contemporaneous leaf in the example that started this discussion. Animals who consume both marine and terrestrial carbon Humans, bears, and dogs are among many organisms that ingest substantial amounts of foods (and hence carbon) ‘mixed’ from both the marine and terrestrial environment. Radiocarbon ages obtained on their bones require a modified form of marine calibration involving estimating the proportion of marine carbon in their diets. Fortunately, the proportion of marine carbon in the diet can be estimated using the commonly applied 13/12C stable isotope ratio measurement that documents the differential uptake of marine and atmospheric carbon by comparison to a regionally specific food web model (e.g., Chisholm, et al. 1982, 1983; Chisholm 1986; Szpak et al. 2009). Calibration programs such as CALIB and OXCAL offer a way to incorporate the percentage of marine carbon in the diet (if known) and calibrate the result using on marine curve. Figure 4.4 provides a summary of the interpretive steps needed in calibration a ‘mixed’ marine and terrestrial sample as well as for strictly terrestrial and strictly marine samples. – 105 –   Figure 4.4. Schematic depicting the steps required for radiocarbon calibration. Note the high number of steps required for calibrating mixed marine and terrestrial samples. Assuming an isotopic value of -25 for charcoal is really only appropriate for northerly latitudes that lack C4 plants (Teeri and Stowe 1976). Estimating Marine Reservoir in Highly Migratory Animals A further complicating factor is the difficulty of interpreting radiocarbon dates on migratory marine animals. For instance, many marine mammal species are highly migratory and forage far offshore throughout the Pacific Ocean meaning their marine carbon is derived from a variety of sources. Japanese researchers have extensively dated archaeological northern fur seal (Callorhinus ursinus) bones and associated human remains from Jomon period sites in Japan (Yoneda, et al. 2004). The highly migratory fur seals returned radiocarbon ages consistently older than contemporaneously deposited human remains. This same study further demonstrated that both fur seal and human remains exhibited respectively older radiocarbon ages than contemporaneously deposited terrestrial mammal bones (Yoneda, et al. 2004). On the Northwest Coast, researchers seeking to directly date whalebone or salmon or any number of highly migratory species should consider this additional uncertainty of the Delta-R estimate. – 106 –  Variable Reporting of Radiocarbon Dates and Calibration Results  Despite the long history of radiocarbon dating, the convention of reporting calibrated radiocarbon ages remains somewhat of a marginal practice in some regions and time periods even through some journals offer very specific recommendations (e.g., Canadian Journal of Archaeology10). The perceived complexity of radiocarbon calibration combined with the diversity of approaches taken by different researchers and research teams, and the quickening pace of research has contributed to considerable variation in the reported radiocarbon record along the northern Northwest Coast. Given the fact that observations have been compiled over 40 years and with different research goals and standards in mind, such variation is entirely understandable. However, this has led to a circumstance where different sets of dates have not been directly compared because they differ in their marine or terrestrial content and have not been placed on a singular calibrated timescale. More importantly, published and repetitively cited accounts of the age of certain cultural events and epochs are understood to have occurred during particular time periods when these ages are more appropriately unanchored points in radiocarbon years lacking appropriate temporal uncertainty. In contrast, all archaeologists report and regularly discuss un-calibrated radiocarbon dates (e.g., 285±50 14C yr bp) despite the fact that the actual age-range of a given set of dates may not be directly comparable or change significantly when calibrated (especially if marine influenced). While this may be seen to be necessary for disciplinary discourse, it neglects the fact that a given radiocarbon result may not be the same time scale. For instance, two books concerning archaeology on the Northwest Coast (Ames and Maschner 1999; Fedje and                                                 10  – 107 –  Mathewes 2005) do not attempt to calibrate radiocarbon dates but rather, offer readers a conversion table that lists a limited series of atmospheric radiocarbon dates in relation to their ‘true’ age in astronomical years. While this convention is welcome (it reduces translation error when discussing dates), it perpetuates a ‘floating’ chronology that is not tied to the astronomical (calendrical) timescale which becomes confusing when dealing with more than a handful of calibrated ages whose differences and similarities may be non-linearly different when calibrated. As a result, many researchers continue to report and discuss radiocarbon dates in un-calibrated radiocarbon years and do not use the full potential of current calibration methods. This is unfortunate because it decreases the accuracy and utility of radiocarbon dates used in archaeological interpretations.  Unfortunately for the archaeologist, achieving sufficient and clear knowledge of the process of radiocarbon calibration is hindered in part by different commercial or research laboratories, whose calibration estimates may not always be the most up-to-date or appropriate for a given region, time period, or context, particularly for marine or marine influenced samples. Moreover, different laboratories often have slightly differing reporting conventions and terminologies that make it difficult to interpret and compare radiocarbon results11. To complicate matters further, academic journals in archaeology have a range of radiocarbon reporting standards and some major journals appear to lack consistently applied standards. These challenges add to the general level of confusion about the process and utility of calibration.                                                  11 For instance, Beta Analytic, a major North American Lab, reports both a ‘measured’ radiocarbon age and a ‘conventional’ (aka. ‘normalized’) radiocarbon age but does not clearly indicate that the difference is a correction for isotopic fractionation and may create confusion in deciding on which age estimates to subject to calibration.  – 108 –  Fortunately, researchers can retroactively re-calibrate adequately reported radiocarbon dates12. This can be quickly accomplished using free online software such as CALIB or OxCAL13.  These programs enable researchers to calibrate large sets of dates and produce tabular and graphical representations of the calibrated age-ranges. Indeed, many academic radiocarbon laboratories (e.g., UC Irvine) no longer conduct the additional interpretive step of calibration. Rather, they leave such decisions up to the individual researchers in recognition that critical contextual details that underlie calibration are best achieved by researchers who selected and submitted the dated material. This is a positive step and we hope will encourage the wider adoption of calibration on the Northwest Coast. Following Bronk Ramsey (2008:249) we contend that “more than ever, it is necessary for users of radiocarbon to understand and engage with the science that underlies the method.” Problems in Interpreting the ‘Calibrated Intercept’ One formerly widespread format for archaeological reporting of calibrated radiocarbon dates is to report the ‘calibrated intercept’ (Bowman 1990). This intercept is the point at which a radiocarbon age (the number before the “±”) ‘intersects’ the calibration curve. However, due to fluctuations in the width and thickness of the curve (Figure 4.1) and the change from a bell-curve from the radiocarbon age to an irregular curve in the calibrated age-range, there is rarely a ‘single’ point or number at which this intersection occurs and                                                 12 One issue that can be extremely confusing and could lead to duplicated data when examining a date, is that multiple laboratory names are occasionally published for the same samples.  For instance, TO-2352 is the laboratory number in CARD and MacDonald and Inglis 1981, whereas Coupland, et al. (2003:154) references the same date as ISOTRACE-2352.  Both are correct, as radiocarbon laboratories often subcontracted out specialized work (such as early accelerator dates), and in these cases both laboratories issued their own numbers. Identification of specific radiocarbon samples is necessary to evaluate and, where appropriate, modify calibration methods. 13 Calibration programs are available for free download or online calculation Calib: or Oxcal:  – 109 –  often the curve is intersected in two or more places and may even have relatively equal  probabilities (Figure 4.2). Thus, while the intercept number is intuitively appealing, it is no longer considered an acceptable reporting convention (Telford, et al. 2004). Rather than a single numerical age bounded by an error estimate, a more appropriate representation of a calibrated age is the probability of a date occurring within an estimated time frame. Free calibration programs readily calculate both the 1-sigma and 2-sigma calibrated age-ranges, representing the 68.3% probability and 94.5% probability respectively. While some may find that reporting and discussing an age-range is awkward, it remains the most justifiable way to succinctly characterize the uncertainty of a calibrated age estimate. Fortunately, calibration programmes also provide researchers with an effective and efficient method for representing the probability visually through graphical imagery (e.g., Figure 4.7). As several Northwest Coast researchers have shown, these graphical features readily accommodate the detailed analysis of individual dates (e.g., Ames 2012:180; Grier 2006:102) or show the levels of relative probability for a series of dates (e.g., Lepofsky, et al. 2005:274; Morin, et al. 2008/09:24). Visual tools illustrate the relative temporal distribution of a suite of dates as well as show the undulations and features of the calibration curve itself. Through such graphical analyses, it is possible to identify chronological similarities and differences that may not be readily apparent in tabular form. Problems in Summarizing Multiple Dates  A persistent problem related to the intercept concept are summary figures such as histograms commonly employed to illustrate trends from multiple dates and sites over time both on the un-calibrated (Fladmark 1975:298) and calibrated timescales (Maschner – 110 –  1991:930). For instance, a recent study from the Great Basin (Louderback, et al. 2011) presents a series of un-calibrated radiocarbon dates in histograms with 150-250-year bin categories to evaluate paleo-demographic sequences. Curiously, this study does not attempt to calibrate these compiled dates but if this had been conducted, the individual dates would likely span more than one histogram bin category and therefore blur the sharp peaks and troughs inferred to represent demographic trends. An alternative and more appropriate method of displaying the cumulative results of multiple radiocarbon dates is by calibrating a set of dates on a single linear timescale using the ‘summed probability’ method discussed below.  Summed Probability as an Analytical Method Summed probability is a Bayesian method for pooling or combining the statistical probabilities from multiple calibrated radiocarbon dates into a singular age-range on the calendrical (linear) time scale. The method provides a statistically robust way to combine the calibrated age-range for a series of dates and hence is suitable for assessing the radiocarbon chronology from multiple archaeological contexts (Figure 4.5). It is particularly useful for combining the probabilities from sets of radiocarbon dates to evaluate whether they cluster in time, are spread over a long interval, or exhibit modes and/or toughs potentially indicative of an archaeological phase or event. Archaeologists have used summed probabilities of radiocarbon datasets to refine archaeological chronologies for some time (Eighmy and LaBelle 1996; Gregory 2001) but as the use of freeware calibration tools such as Calib and OxCal have proliferated, the method has been increasingly applied to address a wide range of hypotheses such as the assessment of cultural historical phases (Ames 2012; Lepofsky, et al. 2005; Morin, et al. 2008/09), proxies for paleo-demography (Peros, et al. 2010; Williams – 111 –  2013), and indicative of population migrations and colonization events (Buchanan, et al. 2008; Collard, et al. 2010; Hamilton and Buchanan 2007).   Figure 4.5. An example of three typical radiocarbon dates on terrestrial charcoal calibrated using the 2009 atmospheric curve and underlain by the summed probability of these three dates. Model produced using Calib 6.10 (Reimer et al. 2009). A flurry of recent research has identified several interpretive and methodological concerns with summarizing radiocarbon data using summed probabilities (Bamforth and Grund 2012; Guilderson, et al. 2005; Steele 2010; Williams 2012). A recent review by Williams (2012) identifies that summed probability profiles are not direct paleo-demographic proxies but are subject to several potential biases, including small sample sizes, taphonomic loss of older sites (Surovell and Brantingham 2007), and calibration effects such as plateaus in the radiocarbon curve that constrain or exaggerate the resolution of some age-ranges. In this latter instance, a given set of dates that concentrates along a relatively ‘flat’ portion of – 112 –  the curve may cause a spuriously high summed probability that would not be as prominent if the same a cluster of dates were taken from a ‘steep’ part of the calibration curve (see Guilderson et al. 2005 for some examples). Other researchers have identified that the inclusion or exclusion of even a single dataset of dates may influence the shape of the summed probability curve and hence a demographic trends resulting from it (Steele 2010). Bamforth and Grund (2012) have demonstrated that with simulations wherein the periodicity (chronological spacing) of dates directly impacts on the shape of a summed probability profile and that this is particularly strong during the dramatic fluctuations in the radiocarbon curve during the Pleistocene-Holocene transition. While these researchers caution against uncritical and overly detailed interpretation of the undulations within summed probability profiles, they by no means dismiss the analytical potential of the approach. Rather, the bulk of caution is directed at using summed probabilities as proxies for trends in human demography without examining the many complex factors underlying this method. For instance, the shape and modality of the summed probability can be influenced by but also evaluated against major perturbations in the radiocarbon curve (Bamforth and Grund 2012; Buchanan, et al. 2011; Guilderson, et al. 2005; Williams 2012). The taphonomic fact that older sites are less likely to survive to the present (Surovell and Brantingham 2007) can be considered and even modelled for a given area (Williams 2012). Sample sizes and sample contexts can be evaluated and the strength of the claim can be adjusted accordingly. These studies critically highlight the continued importance of archaeological context as well as an understanding of the constraints of the radiocarbon method, particularly when interpreting what is often a partial and imperfectly sampled archaeological record. The summed probability method remains the most suitable – 113 –  method for examining the relative distribution of dates from well-sampled archaeological landscapes over broad periods of time.  Unlike the studies described above, our use of summed probability is not to evaluate demographic trends using a the largest number of available dates, but to apply this method to systematically collected dates from the terminal deposits of individual village sites and individual dates on individual human burials. These targeted dating criteria are therefore much less subject to the vagaries of conventional archaeological chronologies (e.g., date the deepest deposit, etc.) and offer a chance to evaluate these interpretively signficant radiocarbon datasets on a single calibrated time scale. Recalibrating the Radiocarbon Record for Prince Rupert Harbour  In light of the insights and challenges presented above, the remainder of the paper focuses on the radiocarbon record for the northern Northwest Coast. We use marine reservoir estimates and the summed probability method to re-evaluate the regional archaeological chronology with the aim of integrating interpretively significant datasets that derive from marine, marine-influenced, and terrestrial radiocarbon dates. In particular, we examine 1) dates on village abandonment in Prince Rupert Harbour and 2) dates on human remains from Prince Rupert Harbour. We focus on this area as it has been subject to intensive and sustained archaeological research for over 40 years and analyse these particular datasets as they represent an important set of radiocarbon observations that underpin the cultural chronology of the northern Northwest Coast during the past 3,000 years.  – 114 –  Methods for Estimating Marine Reservoir on the Northern Northwest Coast Despite the prevalence of radiocarbon dates on marine and marine influenced samples in archaeological and paleoclimate records, the northern Northwest Coast lacks a time series estimate of Delta-R. Previous research by Southon et al. (1990) and Southon and Fedje (2003) presented an extensive, data-rich study of the total [R(t)] marine reservoir effects for three areas of the British Columbia Coast, but do not provide a specific estimate of Delta-R. Rather they provide only a total marine reservoir correction value, which is not useful for calibration because particular Delta-R values are required for each dated sample when using calibration programs14. Another paper by Deo et al. (2004) constructed a Delta-R curve for the southern Northwest Coast, but this only covers the past 3,000 years and is applicable to a very specific region of the Salish Sea. The lack of a published Delta-R estimate or a time series of estimates as well as the perceived complexity of calculating Delta-R values has led to several differing conventions for researchers calibrating marine influenced samples. For instance, many researchers simply subtract an average of the total reservoir age offset (often noted as 600 or 650 years) from normalized marine shell dates and then calibrate the date on the atmospheric curve as if it was terrestrial source material (e.g., Fedje and Mathewes 2005; McLaren 2008; Orchard 2009). Others, including ourselves (Eldridge, et al. 2008), have previously used various estimates for Delta-R based on generalized estimates for the whole Northwest Coast (following Ames 2005 citing Stuiver and Braziunas 1993:156) or specific unpublished estimates recommended by specialists (Moss 1989: citing a personal communication from                                                 14 Delta-R values should not be calculated simply by subtracting the average global reservoir (405 years) from the total reservoir as the global reservoir value also fluctuates through time (Reimer et al. 2009). – 115 –  Stuiver). Some researchers (e.g., Coupland et al. 2006; 2010) have avoided the issue of marine reservoir by exclusively dating charcoal, which is subject to its own possible error from the ‘old wood’ effect15 (Moss 2011a:6; Schiffer 1986). Estimating Delta-R for Prince Rupert Harbour and the Dundas Islands Given the considerable numbers of marine and marine influenced radiocarbon dates obtained from sites in and around Prince Rupert Harbour and the lack of a regionally specific reservoir correction (Delta-R), we created a local record of marine reservoir by compiling all available information into a regional time series (Table 1, Figure 4.6). This record contains 10 dates from 5 shell-wood pairs spanning the past 7,000 years, and which were reported in McLaren (2008) and Southon and Fedje (2003) from the Dundas Islands and Prince Rupert respectively. Each of these marine – terrestrial pairs has been appropriately normalized for isotopic fractionation and dated using accelerator mass spectrometry (AMS) at either Lawrence Livermore National Laboratories or the Keck Carbon Cycle AMS facility at the University of California Irvine. Terrestrial charcoal from these samples was specially obtained using a microscope and selecting identifiably deciduous or recent growth plant remains while the outermost growth margin from shells was submitted for dating (personal communications with Fedje and McLaren, 2008). The Dundas Islands data were obtained from archaeological deposits at Far West Point (GcTr-6) and the Connell Island Site (GcTr-7) and date within three broad intervals in the late, middle, and early Holocene (McLaren 2008; McLaren, et al. 2011). The Prince Rupert data were obtained from archaeological deposits on Tugwell Island just off the northern entrance to the Harbour and date to within                                                 15 This phenomena represents the potential for wood or charcoal to come from old-growth trees or driftwood, the inner parts of which have the potential to be centuries older than the age at which it was deposited.  – 116 –  the late Holocene (Southon and Fedje 2003:107). There are additional shell wood-pair data from Kitimat, 120 km southeast of Prince Rupert (ibid.), but these data were excluded as they date exclusively to the late Pleistocene and were obtained on burrowing shellfish from freshwater-influenced estuary subject to freshwater reservoir effects (e.g., Ingram and Southon 1996; Rick et al. 2012). For each terrestrial date in this modest database, we determined the global marine reservoir for that particular point in radiocarbon time [R(t)] using the raw observations that comprise the Intcal04/09 calibration curve, kindly provided to us by Paula Reimer (Personal Communication, 2008, 201316). These data show the finest existing temporal record for the terrestrial ages that collective make up the published Intcal04/09 calibration curve (Reimer, et al. 2004; Reimer, et al. 2009). The difference in age between coeval Marine04 and IntCal04 dates reveals the known global difference between ocean circulation and atmospheric circulation and thus represents the most specific estimate of global R(t) values through time. After assigning a total reservoir estimate [R(t)] and its error based on the terrestrial radiocarbon age of each of the five shell-wood pairs in the sample, we then subtracted the local shell-wood pair difference from the total (global) reservoir estimate to derive a single Delta-R estimate for each datapoint (± 1 standard deviation). The resulting Delta-R values are then plotted over time (Figure 4.6) and overlaid by polynomial curve to infer temporal trends. The resulting figure represents the current temporal record of marine reservoir for Prince Rupert and the Dundas Islands.                                                 16 These data are also available from the Radiocarbon journal website: – 117 –  This marine reservoir record spans 7,000 years and indicates only moderate fluctuation in regionally specific marine reservoir (Delta-R) during the mid-to-late Holocene (Figure 4.6). Notably, the two similar elevated estimates for the Dundas Islands show virtually no difference between the early and mid-Holocene (Delta-R values of 215, 225 respectively). The two late-Holocene marine–terrestrial pairs from Tugwell Island just outside Prince Rupert Harbour are slightly elevated relative to the two mid and early-Holocene pairs from Dundas (Figure 4.6). This may be due to the proximity of Prince Rupert to the estuarine Skeena River mouth (~25km southeast). However, this may also reflect a climatic influence during this period which overlaps with the ‘neoglacial’ expansion between 3,500-1,000 calendar years ago, when regional climate was colder relative to earlier periods (Menounos, et al. 2009:2063). Similarly, the most recent shell-wood pair (from late period deposits in the Dundas Islands) is slightly elevated relative to the middle and early Holocene shell-wood pairs which occurs during the onset of the ‘little ice age’ (ibid). Overall, the observed fluctuation in reservoir age over time is low to moderate but the temporal trends appears broadly similar to the more variable pattern identified for Haida Gwaii by Southon and Fedje (2003:100).  – 118 –   Figure 4.6. Local marine reservoir (Delta-R) estimates over the Holocene for marine–terrestrial pairs from Prince Rupert Harbour and the Dundas Islands (error bars represent ±1 standard deviation).  Source data obtained from Southon and Fedje (2003) and McLaren (2008). The Delta-R estimate during the last 3,000 years is of the greatest relevance to the datasets presented in this paper.  – 119 –  Table 4.1 Paired marine and terrestrial radiocarbon dates from Prince Rupert Harbour and the Dundas Islands and estimates of the local marine reservoir value depicted in Figure 4.6. Lab #  Marine Material 14C Date ± Lab # Terrestrial  Material  14C Date ± Total Age Offset ± R(t) Poly Local Delta-R poly ± Reference CAMS-49626 Saxidomus 2370 50 CAMS-49625 charcoal  1560 40 810 60 355 455 60 Southon and Fedje 2003:107 CAMS-49624 Protothaca 2780 50 CAMS-49623 charcoal  2040 50 740 70 340 400 70 Southon and Fedje 2003:107 UCIAMS-21882 Shell 4200 15 UCIAMS-21985 charcoal  3645 25 555 30 330 225 30 McLaren 2008:222 UCIAMS-21881 Mytilus 7510 20 UCIAMS-21984 charcoal  6925 50 585 55 370 215 55 McLaren 2008:224 UCIAMS-21980 Mytilus 1395 15 UCIAMS-21983 charcoal  640 60 755 60 385 370 60 McLaren 2008:240  Table 4.2 Recalibrated radiocarbon dates on shell from the uppermost shell midden layers in village sites in Prince Rupert Harbour reported by Archer (1992, 2001). Lab Number Site Number 14C Age 14C Age ±SD Delta-R Value Delta-R ±SD 2-sigma Cal BP Dated Material Reference WSU-4366 GbTo-2  1705 90 390 60 650-1090 marine shell Archer 1992:4 WSU-4367 GbTo-4 1980 90 360 60 940-1370 marine shell Archer 1992:4 WSU-4368 GbTo-4 2980 70 300 60 2150-2670 marine shell Archer 1992:4 WSU-4369 GbTo-7 2720 100 320 60 1770-2320 marine shell Archer 1992:4 WSU-4370 GbTo-7 2360 95 350 60 1320-1830 marine shell Archer 1992:4 WSU-4371 GbTo-8 2130 90 360 60 1083-1550 marine shell Archer 1992:4 WSU-4372 GbTo-8 2550 70 340 60 1570-2030 marine shell Archer 1992:4 WSU-4373 GbTo-9  2490 90 340 60 1480-2000 marine shell Archer 1992:4; Archer 2001:212 WSU-4374 GbTo-9  2360 90 350 60 1330-1820 marine shell Archer 1992:4; Archer 2001:212 WSU-4375 GbTo-28  1900 90 360 60 880-1300 marine shell Archer 1992:4 – 120 –  Lab Number Site Number 14C Age 14C Age ±SD Delta-R Value Delta-R ±SD 2-sigma Cal BP Dated Material Reference WSU-4376 GbTo-28  2600 70 330 60 1650-2120 marine shell Archer 1992:4 WSU-4377 GbTo-32  2435 100 350 60 1380-1930 marine shell Archer 1992:4; Archer 2001:212 WSU-4378 GbTo-32  2220 70 350 60 1250-1640 marine shell Archer 1992:4; Archer 2001:212 WSU-4379 GbTo-46  2530 95 340 60 1510-2060 marine shell Archer 1992:5; Archer 2001:212 WSU-4380 GbTo-46  2590 95 340 60 1570-2130 marine shell Archer 1992:5; Archer 2001:212 WSU-4381 GbTo-57 2470 90 340 60 1460-1980 marine shell Archer 1992:5; Archer 2001:212 WSU-4382 GbTo-57  2220 65 350 60 1250-1620 marine shell Archer 1992:5; Archer 2001:212 WSU-4383 GbTo-59 2870 70 310 60 1980-2470 marine shell Archer 1992:5 WSU-4384 GbTo-59 3200 80 290 60 2360-2880 marine shell Archer 1992:5 WSU-4385 GbTo-64 2370 60 350 60 1380-1800 marine shell Archer 1992:5 WSU-4386 GbTo-64 3840 60 250 60 3290-3700 marine shell Archer 1992:5 WSU-4387 GbTo-66  2500 60 340 60 1530-1940 marine shell Archer 1992:5; Archer 2001:212 WSU-4388 GbTo-66  2590 90 340 60 1590-2120 marine shell Archer 1992:5; Archer 2001:212 WSU-4389 GbTo-70  2445 90 340 60 1410-1930 marine shell Archer 1992:5; Archer 2001:212 WSU-4390 GbTo-70  2510 100 340 60 1490-2050 marine shell Archer 1992:5; Archer 2001:212 WSU-4391 GbTo-77 3210 100 290 60 2350-2940 marine shell Archer 1992:5; Archer 2001:212 WSU-4392 GbTo-77 2925 100 300 60 2030-2670 marine shell Archer 1992:5; Archer 2001:212 – 121 –  Lab Number Site Number 14C Age 14C Age ±SD Delta-R Value Delta-R ±SD 2-sigma Cal BP Dated Material Reference WSU-4393 GbTo-78  2760 90 320 60 1830-2330 marine shell Archer 1992:6; Archer 2001:212 WSU-4394 GbTo-78  2425 80 350 60 1400-1880 marine shell Archer 1992:6; Archer 2001:212 WSU-4395 GbTo-89  2450 70 340 60 1470-1920 marine shell Archer 1992:6; Archer 2001:212 WSU-4396 GbTo-89  2490 90 340 60 1480-2000 marine shell Archer 1992:6; Archer 2001:212 WSU-4397 GbTn-9  2710 90 320 60 1780-2300 marine shell Archer 1992:6 WSU-4398 GbTn-9  4090 70 240 40 3600-4050 marine shell Archer 1992:6 WSU-4399 GcTo-6  2310 70 350 60 1300-1730 marine shell Archer 1992:6; Archer 2001:212 WSU-4400 GcTo-6  2140 90 360 60 1100-1570 marine shell Archer 1992:6; Archer 2001:212 WSU-4401 GcTo-27 2280 95 350 60 1270-1750 marine shell Archer 1992:6 WSU-4402 GcTo-27 2695 70 330 60 1790-2270 marine shell Archer 1992:6 WSU-4403 GcTo-28 2440 100 340 60 1400-1950 marine shell Archer 1992:6 WSU-4404 GcTo-28 2290 90 350 60 1280-1750 marine shell Archer 1992:6 WSU-4405 GcTo-39 2280 90 350 60 1270-1730 marine shell Archer 1992:6 WSU-4406 GcTo-39 2430 100 340 60 1390-1930 marine shell Archer 1992:6 WSU-4407 GcTo-51 2580 100 340 60 1550-2130 marine shell Archer 1992:7 WSU-4408 GcTo-51 2070 90 360 60 1030-1500 marine shell Archer 1992:7 WSU-4409 GcTo-52  2360 75 350 60 1350-1810 marine shell Archer 1992:7; Archer 2001:212 WSU-4410 GcTo-52  2530 90 340 60 1520-2050 marine shell Archer 1992:7; Archer 2001:212   – 122 –  Recalibrating Marine-Influenced Dates from Prince Rupert Harbour Using this marine reservoir time series, it is possible to re-calibrate marine and marine-influenced radiocarbon dates from Prince Rupert Harbour and the Dundas Islands on a single integrated timescale. Thus, we reviewed and compiled all available information on each archaeological date from the region, including the type of dated material (e.g., charcoal, shell), both the standard and normalized radiocarbon ages, and we compiled the measured estimates for the percentage of marine carbon in the diet of dated human remains discussed below. For each marine or marine-influenced date, we derived an age-specific Delta-R value, an uncertainty for Delta-R and then re-calibrated all dates using the CALIB calibration programme (Calib 6.1.1) using the appropriate calibration curve for each individual date (i.e., atmospheric, marine, or mixed marine-terrestrial). Outputs for these re-calibrations are presented in 2-sigma calibrated age-ranges (Table 2) as well as in summarized summed probability plots presented in Figure 4.7 and Figure 4.8.  We re-evaluate the chronology of three series of radiocarbon dates from Prince Rupert Harbour: 1) dates on settlement history and village abandonment by David Archer (1992, 2001) and 2) dates obtained on human remains obtained during the North Coast Prehistory Project (Ames 2005; MacDonald and Inglis 1981).  Calibration Issues in Village Abandonment Dates from Shell The first radiocarbon dataset we consider is from the extensive settlement history study of Prince Rupert conducted by David Archer (1992, 2001) who systematically dated shells collected from the upper layers from 23 village sites to evaluate the hypothesis that there was a period of widespread village abandonment in Prince Rupert Harbour.  – 123 –  Archer’s initial study was based on 45 da