UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Lithics and livelihood : stone tool technologies of central and southern interior B.C. 1983

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
UBC_1983_A1 M33.pdf [ 18.21MB ]
Metadata
JSON: 1.0058420.json
JSON-LD: 1.0058420+ld.json
RDF/XML (Pretty): 1.0058420.xml
RDF/JSON: 1.0058420+rdf.json
Turtle: 1.0058420+rdf-turtle.txt
N-Triples: 1.0058420+rdf-ntriples.txt
Citation
1.0058420.ris

Full Text

LITHICS AND LIVELIHOOD: STONE TOOL TECHNOLOGIES OF CENTRAL AND SOUTHERN INTERIOR B.C. by MARTIN PAUL ROBERT MAGNE B.Sc. , University of Toronto, 1976 M.A., University of Manitoba, 1978 A DISSERTATION SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Anthropology and Sociology, University of British Columbia) We accept this.thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September 1983 '© Martin Paul Robert Magne, 1983 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f the requirements f o r an advanced degree a t the U n i v e r s i t y o f B r i t i s h Columbia, I agree t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and study. I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the head o f my department o r by h i s o r her r e p r e s e n t a t i v e s . I t i s understood t h a t copying or p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be allowed without my w r i t t e n p e r m i s s i o n . Department o f Anthropology and Sociology The U n i v e r s i t y o f B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date October 3, 1983 DE-6 (3/81) ABSTRACT This study is designed to investigate patterns of l i t h i c techno- logical v a r i a b i l i t y in relation to settlement strategies that were em- ployed by late prehistoric inhabitants of central and southern regions of interior British Columbia. The research contributes to current archaeological method through an experimental program of stone tool manufacture, and also to current understanding of Interior Plateau pre- history, through a multiregiohal analysis of technological varia b i l i t y . The f i r s t stage of the study involves conducting a controlled exper- iment , to determine the degree to which l i t h i c debitage can be used to predict stages of chipped stone tool manufacture, and to devise an eff- icient means of classifying debitage into general reduction stages. The experiment is unique in providing control over the precise sequential removal of flakes, and also in examining quantitative va r i a b i l i t y in deb- itage that have been produced as the by-products of the manufacture of several tools and cores. The result of the experimental program i s the formulation of a debitage classification that classifies flakes into early, middle or late reduction stages, and also into b i f a c i a l and bipolar re- duction types. The archaeological analyses i n the second major stage of the research use the debitage reduction stage classification and the occurrence of various l i t h i c tools to examine the nature of interassemblage v a r i a b i l i t y across the 38 sites from four regions of the Interior Plateau. A total of 14,541 flakes, 164 cores and 861 tools from the Eagle Lake, Mouth of the Chilcotin, Lillooet and Hat Creek regions are analyzed, using multivariate and bivariate quantit- ative methods. Three hypotheses relevant to l i t h i c technology and hunter- i i i gatherer archaeology are evaluated in this stage of the study. The analyses f i r s t employ the experimental debitage classification to obtain interpretable patterns of inter-assemblage similarities and differences. Multivariate analysis shows that several kinds of sites defined on the basis of features can be grouped by their predominance of early/core reduction, middle/wide ranging reduction, and late/ maintenance reduction debitage. The f i r s t formal hypothesis tested is that obsidian and chert raw materials should evidence patterns of conservation and economizing behavior by virtue of their geological scarcity in relation to vitreous basalt raw material. A series of chi-square tests demonstrates that debitage frequencies by reduction stage are proportionately equal for these three raw materials in a l l but the Mouth of the Chilcotin region. In a l l regions, except Lillooet where tool sample sizes are too small for reliable testing, tool sizes and scar counts show no significant difference attributable to raw materials. A slight trend is noted for chert tools to be larger and simpler than vitreous basalt or ob- sidian tools. A set of bivariate graphs demonstrates that while l i t h i c raw materials may be reduced in highly similar manners, one raw material may have served to replace another. The second hypothesis, that tool curation and maintenance strongly affects assemblage composition, is f i r s t tested by examining tool assem- blage measures that have been suggested by recent l i t h i c technological models. Assemblages are highly variable with respect to the numbers of tools l e f t at sites in relation to the intensity of tool maintenance that occurred at sites. iv The third hypothesis tested i s that a set of site occupation purposes can be reliably predicted on the basis of debitage reduction stages and a functional tool classification. Using multiple discriminant analysis, house- pit sites are accurately predicted at an 80% rate, and l i t h i c scatters with- out features are accurately predicted at a rate of 60%. Lithi c scatters with housepits achieve 86% correct classification; l i t h i c scatters with cachepits are correctly classified at a rate of 75%; and l i t h i c scatters with f i r e - cracked rock are accurately predicted 80% of the time. The results of this analysis are further strengthened by removing an ambiguous assemblage from consideration. The most significant findings of the multiregional analyses are those of definite tool cuiration patterns as evidenced i n the raw material analysis, and the occupation span inferences of the tool maintenance analysis. Over- a l l , i t has been demonstrated that an experimentally obtained stage class- i f i c a t i o n of debitage enables the derivation of behavioral inferences that could not be currently obtained by other means. In i t s multiregional per- spective, this study has shown that processes of l i t h i c assemblage formation are largely independent of regional provenience and more dependent on settle- ment purpose. Overall, the greatest determinant of assemblage var i a b i l i t y i s inferred to be site occupation span. V ACKNOWLEDGEMENTS I owe a great deal to my advisory committee. Dr. David Pokotylo, chairman, steered my fledgling interests in l i t h i c technology to constructive paths, and his demands for origin- a l i t y and thoroughness are appreciated. Dr. R.G. Matson pro- vided the opportunity for the Eagle Lake research, gave a great deal of advice, and always had his door open. Dr. David Aberle encouraged anthropological and logical awareness, and lent an editorial hand. Dr. Richard Pearson provided an invaluable example and took the time to comment on drafts and generally assist throughout my program. Dr. Arnoud Stryd trusted the Lillooet collections to my care, read drafts, and he and his wife Melanie graciously allowed me the use of their home in Kamloops. For financial support, I am grateful to the B.C. Heritage Trust for awarding me the 1979-80 Charles Borden Scholarship in Archaeology, and I would also like to thank the University of British Columbia for the 1980-81 Charles and Alice Borden Fellowship! >in Archaeology, and also for the Norman Mackenzie Fellowship. Unlike many people in B.C., I never knew Charles Borden, yet I must acknowledge his pioneering work and dedication to the f i e l d . The research conducted at Eagle Lake was supported by a grant from the Social Sciences and Humanities Research Council of Canada to Dr. R.G. Matson, as was the Shuswap Settlement Patterns Project. The Hat Creek Project under Dr. D. Pokotylo's direction was funded by the B.C. Hydro and Power Authority. Dr. A. Stryd has directed the Lillooet v i A r c h a e o l o g i c a l P r o j e c t over the years w i t h f e d e r a l , p r o v i n c i a l and l o c a l funds. Moira I r v i n e deserves s p e c i a l mention f o r her q u a l i t y pre- pa r a t i o n s of a l l the i l l u s t r a t i o n s and photographs that are con- tained i n the f o l l o w i n g pages. The t y p e s c r i p t has been p a t i e n t l y produced by C a r l i Nixon. The experimental program would not have been p o s s i b l e without the a s s i s t a n c e of George K u r z e n s t e i n , and the students of ANTH 420, 1979: Rhonda Aceman, Cindy Bunbury, Mike Cook, Tony Laroc, Briony Penn, C o l l e e n Rudy, Terry S e i d e l , L i s a Smedman, P a t r i c i a Ward, and B r i g i t t e Westergaard. For t h e i r companionship," heated d i s c u s s i o n s , and o f t e n ram- bunctious senses of humour, thanks to Michael B r o d e r i c k , Gary Coupland, C a r o l Coupland, David F r i e s e n , Leonard Ham, Dana Lepofsky, Deanna Ludowicz, Richard Mackie, Lynda MacCaull, C a r l i Nixon, L a r r y Palmer, Briony Penn, Donald P r i c e , Linda Roberts, David Rozen and Robert Tyhurst. S p e c i a l thanks to Susan Matson at Eagle Lake. I am g r a t e f u l to a l l who maintained correspondence and exchanged manuscripts, e s p e c i a l l y E i l e e n C a m i l l i , Dr. James Cha t t e r s , Dr. Knut Fladmark, Dr. B r i a n Hayden, Dr. Joseph Jorgensen, Dr. Paul Katz,. Dr. Raymond Leblanc, Dr. David S t a h l e , and Dr. David Thomas. My g r e a t e s t a p p r e c i a t i o n s are f o r my parents Charles and Delphine, and my brothers Jean, Gerard and Luc. v i i TABLE OF CONTENTS ABSTRACT 1 1 ACKNOWLEDGMENTS . . v i LIST OF TABLES ± x LIST OF FIGURES x i i Chapter 1. INTRODUCTION 1 2. LITHIC TECHNOLOGY AND HUNTER-GATHERER MOBILITY. 5 2.1. I n t r o d u c t i o n 5 2.2. The Mousterian Problem 6 2.3. L i t h i c s , L o g i s t i c s and L i v e l i h o o d 21 3. ETHNOGRAPHIC AND ARCHAEOLOGICAL CONTEXTS 48 3.1. Regional Ethnography 48 3.2. Ethnographic References to L i t h i c Technology. 64 3.3. Regional P r e h i s t o r i c Archaeology. 69 4. THE EXPERIMENTS IN DEBITAGE CLASSIFICATION 94 4.1. I n t r o d u c t i o n 94 4.2. Experimental Controls 95 4.3. The P i l o t Study 9 9 4.4. Experimental Products 100 4.5. Stage D e f i n i t i o n 106 4.6. Debitage Variables. 1° 8 4.7. Hypothesis T e s t i n g 114 4.8. Summary of Experimental Findings 127 v i i i 5. THE ARCHAEOLOGICAL DATA BASE 130 5.1. S i t e D e s c r i p t i o n s 130 5.2. A r t i f a c t C l a s s i f i c a t i o n 159 6. A MULTIREGIONAL PERSPECTIVE ON LITHIC ASSEMBLAGE VARIABILITY 195 6.1. I n t r o d u c t i o n 195 6.2. Reduction Factors 196 6.3. Raw M a t e r i a l Factors 205 6.4. Implement Maintenance and Curation F a c t o r s . 222 6.5. Settlement Strategy Factors 231 6.6. Assemblage Formation Summaries 245 6.7. Summary 251 7. SUMMARY AND CONCLUSIONS 253 7.1. Summary.....'. 253 7.2. Conclusions 262 8. REFERENCES CITED 267 9. APPENDIX 1. 292 ix LIST OF TABLES 1. Seasonality of Interior Plateau groups as evidenced by general activities undertaken during "moons" 61 2. Frequencies of general flake classes and reduction events for each experimental core and tool 96 3. Mean number of flakes per reduction event and PRB/ Shatter ratio, in grouped reduction types by raw material 105 4. MDA classification results of a l l PRB's (N=-994) 117 5. MDA classification results of obsidian PRB's, 25% random sample (N=28) 117 6. MDA classification results, debitage produced by experienced knappers, 50% random sample (N=222) 119 7. Chi-square contingency table, PLCO by Stage, PRB's produced by experienced knappers 119 8. Chi-square contingency table, DOCO by Stage, Shatter produced by experienced knappers 121 9. Mean, median and standard deviations of weight, plat- form and scar counts, debitage produced by experienced knappers, broken down by stage of reduction 122 10. Summary data for the 38 assemblages under study 135 11. Assemblage debitage classes, raw counts, a l l raw materials 162 12. Assemblage debitage classes, percent by count, a l l raw materials 163 X 13. Tool morphology c l a s s e s 1°5 14. Tool type frequencies by s i t e 166 15. Assemblage context compared to major r e d u c t i o n f a c t o r s 204 16. Percent raw m a t e r i a l . c o m p o s i t i o n of debitage assemblages by counts 209 17. Raw m a t e r i a l composition of t o o l assemblages by ? i n percentages 18. Chi-square t e s t of Eagle Lake debitage general r e d u c t i o n stages by raw m a t e r i a l . . 211 19. Chi-square t e s t of Mouth of the C h i l c o t i n general r e d u c t i o n stages by raw m a t e r i a l 211 20. Chi-square t e s t of L i l l o o e t general r e d u c t i o n stages by raw m a t e r i a l 212 21. Chi-square t e s t of Hat Creek general r e d u c t i o n 9 19 stages by raw m a t e r i a l . . . . 22. Chi-square t e s t of Eagle Lake t o o l s i z e s by raw . -i 214 m a t e r i a l 23. Chi-square t e s t of Eagle Lake t o o l scar counts O 1 / by raw m a t e r i a l 24. Chi-square t e s t of Mouth of the C h i l c o t i n t o o l 215 s i z e s by raw m a t e r i a l 25. Chi-square t e s t of Mouth of the C h i l c o t i n t o o l 215 scar counts by raw m a t e r i a l . . . . . 26. Chi-square t e s t of Hat Creek t o o l s i z e s by raw «- • i 216 m a t e r i a l x i 27. Chi-square test of Hat .Creek tool scar counts by raw material 216 28. Total tool weights and scar counts by site 224 29. Data employed in the settlement component discriminant analysis 236 30. Result of the multiple discriminant analysis based on functional tool classes and condensed debitage classes 237 31. Chi-square test of independence, five settlement types by personal gear and bipolar cores 240 32. Results of multiple discriminant analyses with F8:1 removed 242 33. Chi-square test of independence, five settlement types by personal gear and bipolar cores with F8:l removed 244 x i i LIST OF FIGURES 1. The general model of l i t h i c r e d u c t i o n , maintenance and d i s p o s a l 24 2. Comparison of Ebert's (1979) i n f e r e n t i a l p o i n t swarms w i t h a c t u a l comparative s c a l e 38 3. Physiographic zones of B r i t i s h Columbia, showing the area of study 49 4. Ethnographic groups of I n t e r i o r B r i t i s h Columbia, showing the major bands of i n t e r e s t 50 5. Flake blanks removed from l a r g e s i n g l e - p l a t f o r m b a s a l t core. Not a l l are shown 101 6. B i p o l a r cores and derived blanks 101 7. Large b i f a c i a l t o o l products 102 8. Large u n i f a c i a l t o o l products 102 9. Large marginal t o o l products 103 10. Small marginal t o o l products 103 11. Debitage a t t r i b u t e s employed i n the experimental program 112 12. Graph of mean, median and standard d e v i a t i o n values f o r weight and p l a t f o r m scar count of PRB's produced by experienced knappers 1-23 13. Graph of mean, median and standard d e v i a t i o n values f o r weight and d o r s a l scar count of Shatter produced by experienced knappers 1-24 14. The experimental debitage c l a s s i f i c a t i o n 129 x i i i 15. Locations of the four regions under study 131 16. Eagle Lake region sites 13° 17. Eagle Lake region site EIRw 4 I 3 7 18. Mouth of the Chilcotin region sites 144 19. Lillooet region sites 20. Hat Creek region sites 1 5 3 21. 14:2 tools 1 7 0 22. 16:1 tools and cores 1 7 0 23. 19:1 tools 1 7 1 24. 19:1 tools and cores 1 7 1 25. 19:1 cores • 1 7 2 26. 22:1 tools and cores. 1 7 2 27. 26:3 tools 1 7 3 28. 32:1 tools and cores 1 7 3 29. CR28 tools 1 7 4 30. CR64 cores 1 7 4 31. CR40 tools 1 7 4 32. CR73 tools 1 7 5 33. EIRw 4 tools and cores 1 7 5 34. CR92 tools and cores 1 7 6 35. CR92 tools and cores 1 7 6 36. EkRo 18 tools 1 7 7 37. EkRo 31 tools and cores 1 7 7 178 38. EkRo 48 tools and cores 178 39. 2:3 tools and cores 179 40. 2:3 tools.. xiv 41. 2:3 t o o l s and cores 42. 4:2 t o o l s and cores 1 8 0 43. 4:5 t o o l s and cores 1 8 0 181 44. 4:1 t o o l s and cores 181 45. 5:1 t o o l s and cores ' 1 8? 46. 9:1 t o o l s and cores 47. 9:2 t o o l s and cores I Q O 48. 12:6 t o o l s and cores.. 49. EeRk 16 t o o l s 1 8 4 50. EeRl 41 t o o l s and cores 1 8 4 51. EeRk 7 t o o l s • 1 8 5 I Q C 52. EeRk 7 t o o l s and cores 1 8fi 53. EeRk 7 t o o l s and cores 54. EeRk 4:38 t o o l s and cores 1 8 6 1 87 55. EeRl 40 t o o l s < 187 56. EeRl 40 t o o l s and cores 57. G21:9 t o o l s . . 188 58. G23:l t o o l s and cores 189 59. G2:12 t o o l s and cores 189 60. G31:l t o o l s and cores 190 61. F 8 : l t o o l s 190 62. F 8 : l t o o l s and cores 191 63. Fl2:5 t o o l s and cores l ^ l 64. J22:2 t o o l s . 192 65. J38:2 t o o l s and cores X V 66. K2:l t o o l s and cores 192 67. EeRj 1 t o o l s 193 68. EeRj 1 t o o l s and cores 193 69. EeRj 1 t o o l s 194 70. The general character of assemblage debitage r e d u c t i o n stages as revealed through TSCALE and Ward's HCLUS 1 9 9 71. Graph of the percent of debitage derived from v i t r e o u s b a s a l t v s . the percent of t o o l s d erived from v i t r e o u s b a s a l t per assemblage 218 72. P l o t of the percent of debitage derived from chert or o b s i d i a n vs. percent of t o o l s f o r the same raw m a t e r i a l per assemblage 220- 73. Assemblage t o t a l t o o l weight p l o t t e d against t o t a l t o o l scar counts ^25 74. Graph of the t o t a l number of t o o l s vs. the percent 75. R-Mode a n a l y s i s of the presence or absence of 21 t o o l and f e a t u r e c l a s s e s i n the 38 assemblages.. 1 CHAPTER 1 INTRODUCTION The purpose of t h i s study i s to discover how l i t h i c technology v a r i e d w i t h i n a wide range of settlement s t r a t e g i e s that were employed by l a t e p r e h i s t o r i c i n h a b i t a n t s of c e n t r a l and southern I n t e r i o r B r i t i s h Columbia. To achieve t h i s g o a l , the research proceeds i n two major stages. The f i r s t step i n - volves conducting an experimental program i n chipped stone t o o l manufacture, to determine the degree to which t o o l manufacture stages can be i n f e r r e d by a n a l y s i s of the by-products of that process, and to devise a r e l i a b l e , yet r e l a t i v e l y simple means of c l a s s i f y i n g debitage i n t o r e d u c t i o n stages. The second major aspect of t h i s study i n v o l v e s the a p p l i c a t i o n of the experimental f i n d i n g s to a r c h a e o l o g i c a l c o l l e c t i o n s from four regions of the I n t e r i o r P l a t e a u , to evaluate a set of general p r o p o s i t i o n s con- cerning assemblage v a r i a b i l i t y . The major polemic that i s advanced i n the f o l l o w i n g pages i s t hat the va r i o u s uses of s i t e s by hunters and gatherers, r a t h e r than the a n t i q u i t y or ethnic a f f i n i t y of s i t e s are the most important determinants of l i t h i c assemblage composition. The s p e c i f i c b e h a v i o r a l inferences that are derived f o r assemblages are based on both experimental and a r c h a e o l o g i c a l c o n t r o l s as w e l l as analogs w i t h recorded ethnographic p a t t e r n s . This d i s s e r t a t i o n 2 has a strong e m p i r i c a l and methodological focus, and the i n t e r - r e g i o n a l research i s unique i n i n v e s t i g a t i n g the extent to which p r e h i s t o r i c settlement behavior apparent i n one re g i o n may be comparable to that e x h i b i t e d i n other nearby and d i s t a n t r e g i o n s . The b e h a v i o r a l viewpoint discussed i n Chapter 2, has been ev o l v i n g i n a r c h a e o l o g i c a l research f o r some two decades, but i s only r e c e n t l y being a p p l i e d i n B r i t i s h Columbia, i n s t u d i e s that do more than a l l u d e to t h i s important concept. Chapter 2 presents the c r i t i c a l o r i g i n s of b e h a v i o r a l approaches to stone t o o l s , and d e t a i l s the development of s e v e r a l approaches, as witnessed mainly i n the co n t i n u i n g arguments of Lewis B i n f o r d . The t h i r d chapter provides background d i s c u s s i o n s , focusing on the e x i s t i n g ethnographic and a r c h a e o l o g i c a l records of the I n t e r i o r P l a t e a u . The review of ethnographic knowledge serves to demonstrate that the e a r l y h i s t o r i c i n h a b i t a n t s of the I n t e r i o r Plateau had an e s s e n t i a l l y common l i f e s t y l e , that was h i g h l y seasonal and very mobile. Here are discussed p a r t i c u l a r exceptions to the general p a t t e r n , that are i n evidence w i t h respect to the groups l i v i n g w i t h i n the four regions that are i n v e s t i g a t e d . The chapter a l s o b r i e f l y reviews ethnographic references to l i t h i c technology. The development of p r e h i s t o r i c a r c h a e o l o g i c a l research i n the ce n t r a l - s o u t h e r n I n t e r i o r Plateau i s discussed i n terms of e a r l y h i s t o r i c observations and s p e c u l a t i o n s , c u l t u r e - h i s t o r y i n - v e s t i g a t i o n s , and settlement p a t t e r n s t u d i e s . The l a t t e r are im- portant i n p r o v i d i n g both the methodological and e m p i r i c a l bases 3 fo r the current study. The experimental program i n debitage a n a l y s i s and c l a s s - i f i c a t i o n i s presented i n Chapter 4. The task of rendering the d e s c r i p t i o n of chipped stone t o o l manufacturing stages i n t o a q u a n t i t a t i v e method i s discussed i n terms of i t s o r i g i n s and outstanding problems, and a s o l u t i o n to some of these problems i s developed. The experiments are prec e d e n t - s e t t i n g i n t h e i r use of s p e c i f i c c o n t r o l s and i n examining general r e d u c t i o n stages i n the manufacture of a wide range of t o o l s and cores, r a t h e r than s i n g l e t o o l types. In Chapter 5, I des c r i b e the a r c h a e o l o g i c a l data base of the research program. Each of the 38 s i t e s under study i s described. Summary q u a n t i t a t i v e data on the debitage and t o o l assemblages are provided, as are photographic i l l u s t r a t i o n s of the t o o l and core assemblages. M u l t i r e g i o n a l analyses of l i t h i c assemblage v a r i a b i l i t y are presented i n Chapter 6, where three hypotheses of importance to model-building i n hunter-gatherer archaeology are t e s t e d . The analyses d i s c l o s e patterns of assemblage formation processes w i t h respect to r e d u c t i o n stages, raw m a t e r i a l s , t o o l c u r a t i o n , and settlement s t r a t e g y f a c t o r s . As a means of summarizing the r e s u l t s of the analyses, the s i t e s are grouped i n terms of i n f e r r e d occupation spans, and kinds of c u l t u r a l features present, and c o n s i s t e n t and p r e v a i l i n g patterns of l i t h i c assemblage formation are discussed. 4 The f i n a l chapter concludes the study by summarizing i t s major contributions to current archaeological method and to Interior Plateau archaeology. The overall success of the study is evaluated here, and areas of research in need of further consideration are identified. 5 CHAPTER 2 LITHIC TECHNOLOGY AND HUNTER-GATHERER MOBILITY 2.1. I n t r o d u c t i o n The i n t e r p r e t a t i o n of l i t h i c assemblage v a r i a b i l i t y i n p r e h i s t o r i c human l o c a l e s i s one of archaeology's l e a d i n g pro- blem areas. U n t i l w e l l i n t o the 1960's, and s t i l l a v a l u a b l e paradigm, the p r e v a i l i n g approach to stone t o o l s was c u l t u r e - h i s t o r i c , and was r a r e l y based on q u a n t i f i e d explanations of meaning i n t o o l form. The major t h e o r e t i c a l and methodological innovations that i n i t i a t e d b e h a v i o r a l approaches to l i t h i c assemblages were provided by B i n f o r d and B i n f o r d (1966, 1969), and continue to be a c t i v e l y pursued by Lewis B i n f o r d , although he would probably not p r e s e n t l y l a b e l h i s approach as b e h a v i o r a l (see B i n f o r d 1981b). This study e x p l i c i t l y employs B i n f o r d ' s p e r s p e c t i v e and expectations, and the purpose of t h i s chapter i s to examine the development of b e h a v i o r a l analyses of l i t h i c assemblages. Here are f i r s t discussed the beginnings of the s h i f t i n paradigms, that focused on the i n t e r p r e t a t i o n , of s t r a t i g r a p h i c d i f f e r e n c e s i n assemblages from the Mousterian of Europe. The Mousterian problem has been at the f o r e f r o n t of a r c h a e o l o g i c a l awareness, and a thorough d i s c u s s i o n of i t s development provides a s u i t a b l e ana- log of the changes i n a r c h a e o l o g i c a l theory and method that have 6 been brought about by mainly Western a r c h a e o l o g i s t s i n the l a s t 20 years. The d i s c u s s i o n a l s o r e v e a l s some va l u a b l e e m p i r i c a l i n f o r m a t i o n , and r e i n f o r c e s the l a r g e - s c a l e p e r s p e c t i v e of t h i s study. To provide the major t h e o r e t i c a l background to t h i s study, I then review current developments i n understanding the r e l a t i o n - ships between l i t h i c technology and the m o b i l i t y of hunters and gatherers. Again, B i n f o r d ' s c o n t r i b u t i o n s are extremely r e l e v a n t , and are d e t a i l e d enough to provide p r o p o s i t i o n s that are examined i n f u t u r e chapters, as are c e r t a i n g e n e r a l i z a t i o n s provided by other researchers working along these l i n e s . 2.2. The Mousterian Problem The term Mousterian i s used to describe a r t i f a c t assemblages o c c u r r i n g during the time of the Eemian i n t e r g l a c i a l and e a r l y to middle Wurm environmental episodes i n Europe and Western A s i a . I t has a l s o been a p p l i e d to assemblages i n China (Bordes 1969: 129 - 130). The Mousterian was f i r s t defined by de M o r t i l l e t (1869) at Dordogne s h e l t e r , as a means of d i s t i n g u i s h i n g i t s f l a k e t o o l i n - dustry from the e a r l i e r handaxe assemblages of the Acheulian. Francois Bordes, the researcher who was by f a r the most f a m i l i a r w i t h the Mousterian and a l l of i t s v a r i a n t s , maintained that the Mousterian i s - d e s c r i b e d t e c h n o l o g i c a l l y as an i n d u s t r y composed of f l a k e s that may or may not have f a c e t t e d s t r i k i n g p l a t f o r m s , w i t h v a r i a b l e proportions of p o i n t s , s i d e - s c r a p e r s , d e n t i c u l a t e t o o l s , 7 bifaces, and burins (among other sub-types). In a l l , 63 tool types are used to describe Mousterian variability (Bordes 1972: 48). There are four recognized major kinds of Mousterian assem- blages, or facies, that are based on cumulative frequency graphs of the 63 tool types, when they are arranged in a specific order (Bordes 1972: 49 - 52; cf. Fish 1976). 1. The Mousterian of Acheulian Tradition (MAT) was thought by Bordes (1961) to be derived from the late Acheulian, and is divided into two subtypes, A and B. MAT subtype A contains relatively intermediate amounts of sidescrapers and denticulates (20 to 40 % ) , well-made cordiform and triangular handaxes (8 - 40%), and rare backed knives. MAT subtype B contains very low frequencies of sidescrapers, but large amounts of denticulates. Handaxes are rare. Occasionally, tool types more common in the Upper Paleolithic, such as burins and endscrapers, are present. 2. The Typical Mousterian is principally defined by the absence of tools such as handaxes, backed knives, and any tools with steep "Quina" retouch. Sidescrapers range from 20 - 65% of the total inventory of tools. While MAT subtypes A and B are thought to' occur in the Wurm I and Wurm II/III respectively, no chronological position is assigned to Typical Mousterian. Furthermore, Bordes (1972) appar^^t ently placed any assemblage that cannot be assigned to the other Moust- erian types, into the Typical Mousterian. 3. Denticulate Mousterian assemblages contain few sidescrapers -; and many denticulate tools. Notched and denticulated tools account together for some 80% of a l l tools. 8 4. In the Charente Department of France there occur t o o l s of two kinds of Mousterian assemblages that together c o n s t i t u t e the Charentian. Quina Mousterian i s r e a d i l y i d e n t i f i e d through the presence of sid e s c r a p e r s w i t h a high angle of retouch. F e r r a s s i e Mousterian i s d i f f e r e n t from Quina in';:that the Lev- a l l o i s technique of f l a k e manufacture i s much more predominant, and i t a l s o contains r e l a t i v e l y few Quina scrapers. Both of these types have r e l a t i v e l y few d e n t i c u l a t e s , handaxes, and Upper P a l e o l i t h i c t o o l types. I t i s g e n e r a l l y accepted that these four kinds of f a c i e s represent a general l e v e l of t e c h n o l o g i c a l and t y p o l o g i c a l de- velopment amongst H. sapiens neanderthalensis p o p u l a t i o n s , hut i t i s a l s o g e n e r a l l y recognized that the s p a t i a l and temporal/ s t r a t i g r a p h i c occurrence of these kinds of assemblages i s ex- tremely complex. As attempts to e x p l a i n v a r i a b i l i t y i n the Mousterian, three kinds of i n t e r p r e t a t i o n s have been o f f e r e d , and there i s a great deal of debate among authors as to the s i g - n i f i c a n c e of Mousterian v a r i a b i l i t y . The f i r s t k i n d of i n t e r p r e t a t i o n o f f e r e d i s the idea that each kind of Mousterian represents a separate but l a r g e l y con- temporaneous c u l t u r a l t r a d i t i o n . Bordes (1961) a l s o examined the p o s s i b i l i t y that each kind was an i n d u s t r i a l f a c i e s adapted to a p a r t i c u l a r microenvironment, and he a l s o considered that each type could represent s e a s o n a l l y d i f f e r e n t a c t i v i t i e s . He was able to r e j e c t both of these hypotheses. On the b a s i s of m i c r o g e o l o g i c a l 9 work at the important s i t e s of Combe Grenal and Pech de I'Aze (Bordes 1972), he demonstrated that there was no correspondence between Mousterian type and i n d i c a t e d microenvironment. A l s o , some kinds of Mousterian are widely d i s t r i b u t e d throughout France and the Levant, l e a d i n g one to doubt that the e f f e c t s of a s i n g l e environment type could account f o r assemblage type d i f f - erences. To t a c k l e the second hypothesis, Bordes w i t h the a s s i s - tance of Bouchud and Prat ( c f . Bordes and Prat 1965) analysed faunal data which to :them showed that some of the Mousterian occupations were year-round. While t h i s seemed q u i t e unusual f o r what i s commonly thought of as a hunting and gathering adaptation, Bordes found h i s o r i g i n a l hypothesis of t o o l types representing d i f f e r e n t e t h n i c groups most acceptable. However, Bouchud's f a u n a l a n a l y s i s of .caribou has been s t r o n g l y c r i t i c i s e d by B i n f o r d (.1973: 238 - 240) on the b a s i s of c e r t a i n assumptions made concerning tooth e r u p t i o n stage, and i t seems more probable that the Mousterian samples analysed a l l represent short-term occ- upations. The second major i n t e r p r e t a t i o n of Mousterian v a r i a b i l i t y i s that each kind of Mousterian occurs i n a d i s c r e t e temporal span. This p o s i t i o n has been mostly defended by M e l l a r s (1965, 1970) , who argues that there i s l i t t l e temporal overlap i n the occurrence of Quina, F e r r a s s i e , and Mousterian of Acheulian T r a d i t i o n . This hypothesis i s based on the a n a l y s i s of 12 s i t e s i n southern France, which suggest that Quina evolved from F e r r a s s i e , and that MAT occurs 10 a f t e r Quina. Doran and Hodson (1966) subscribe to e s s e n t i a l l y the same hypothesis. In t h e i r work, an e a r l y m u l t i v a r i a t e ana- l y s i s of 16 s i t e s from France, Monaco, Spain and Greece produced three c l u s t e r s that seemed to broadly agree w i t h Bordes' Mousterian f a c i e s . While Doran and Hodson 1s r e s u l t s can be dismissed as probably f o r t u i t o u s due to poor sampling c o n s i d e r a t i o n s , M e l l a r s ' research was much more c a r e f u l l y thought out and executed, but has major problems as well.' F i r s t of a l l , M e l l a r s ignores the contemporaneity of the MAT and Charentian types at Combe Grenal and Pech de l'Aze ( c f . Bordes 1972) during the e a r l y Wurm I p e r i o d . Secondly, L a v i l l e (.1973) has shown through s e d i m e n t o l o g i c a l work on the e a r l y Wurm chronology at these s i t e s and Caminade and Le Moustier that the three types of Mousterian d i d c o - e x i s t . He was a l s o able to demon- s t r a t e that Quina can precede F e r r a s s i e , and the MAT subtype B can precede F e r r a s s i e and Quina ( L a v i l l e 1973). In s h o r t , L a v i l l e ' s work seems to confirm Bordes' expectations of contemporaneity. There i s s t i l l a major flaw i n M e l l a r s ' argument, and that i s that even i f s e q u e n t i a l assemblage types were demonstrable, what would that t e l l us about why t h i s i s so? B i n f o r d has st a t e d the problem as f o l l o w s : Time and space are reference dimensions which we use f o r monitoring the operation of system dynamics. The demonstration of c l u s t e r i n g along e i t h e r of these dimensions only informs us that some systemic processes were at work. Such a demonstration does not inform us of the nature of these processes (B i n f o r d 19.73: 247 - 248). 11 The t h i r d approach to assemblage v a r i a b i l i t y i n the Mousterian i s known as the "fu n c t i o n a l argument" (Binford 1973, Mellars 1970). The o r i g i n of t h i s argument i s the now c l a s s i c a r t i c l e by Binford and Binford (1966; see also Freeman 1966) b r i e f l y presented i n Binford and Binford (1969) . The Binfords' purpose was to show that the Mousterian assem- blages' v a r i a b i l i t y could be systematically p a r t i t i o n e d according to the kinds of a c t i v i t i e s that had been undertaken at s i t e s , with the kinds of a c t i v i t y being represented by vary- ing proportions of tools i n Bordes' type l i s t . Thus, rather than assuming that the proportions of the 63 t o o l types varied with ethnic d i f f e r e n t i a t i o n , chronological ordering, or s t r i c t l y seasonal patterning, the Binfords assume that pro- portions of tools should vary according to d i s c r e t e functions.- : There are two other basic assumptions here. The f i r s t i s that "function" i s mu l t i v a r i a t e and systemic, or that there are multiple, linked "determinants of any given s i t u a t i o n " (Binford and Binford 1966: 241). The second i s "that v a r i a t i o n i n the structure and content of an archaeological assemblage i s d i r e c t l y related to form, nature and s p a t i a l arrangement of human a c t i v - i t i e s " ( I b i d ) . The reasoning behind t h i s argument has not been c r i t i c i s e d except f o r the assumption that Bordes' t o o l typology expresses function, which i s quite untested (Cowgill 1968). In t h e i r analysis of Mousterian v a r i a b i l i t y among one s i t e from France (Houpeville) and two from the Near East (Jabrud and 12 Shubbabiq) the B i n f o r d s argued that d i f f e r e n t i a l p r o p o r t i o n a l frequencies of t o o l types were the r e s u l t of d i f f e r e n t tasks being c a r r i e d out at s i t e s . The e s s e n t i a l d i s t i n c t i o n was be- tween base camps, where maintenance tasks would be c a r r i e d out, and work camps, where e x t r a c t i v e tasks were undertaken. To demonstrate t h i s , i t was necessary f o r them to f i n d f u n c t i o n a l u n i t s that could be used to compare s i t e assemblages. This was accomplished through f a c t o r analyses of the data, to d i s - cover which t o o l types tended to covary; that i s , t o o l s used together would tend to be found together, and through f a c t o r a n a l y s i s , the d i f f e r e n t t o o l s used i n any given task should c o n s t i t u t e a d i s t i n c t f a c t o r . The p r i n c i p l e upon which the B i n f o r d s ' assumption was based was s t a t e d e s s e n t i a l l y i n an e a r l y paper as f o l l o w s : The l o s s , breakage and abandonment of imple- ments and f a c i l i t i e s at d i f f e r e n t l o c a t i o n s , where groups of v a r i a b l e s t r u c t u r e performed d i f f e r e n t t a s k s , leaves a " f o s s i l " record of the a c t u a l operation of an e x t i n c t s o c i e t y ( B i n f o r d 1964: 425). This p r i n c i p l e has been s t r o n g l y c r i t i c i z e d by S c h i f f e r (1976: 11), who makes e x p l i c i t t h a t there are c u l t u r a l and n a t u r a l t r a n s - formation processes which may a l t e r the s p a t i a l , q u a n t i t a t i v e , formal and r e l a t i o n a l c h a r a c t e r i s t i c s of a r t i f a c t s subsequent ..to t h e i r primary d e p o s i t i o n . The B i n f o r d s made no attempt to study s y s t e m a t i c a l l y the p o s s i b l e e f f e c t s of n-transforms such as geo- 13 logical processes of erosion, nor such possible c-transforms as site re-occupation and tool re-use. Granted, these are d i f f i c u l t problems, but i t is clear that assumptions about the homogeneity of these processes between sites of -different kinds (i.e. sheltered and open) are:unwarranted^ even though Schiffer (1976: 57) seems to imply that using single occupation 'sites rather than divisions within stratigraphic layers as ana^ l y t i c units avoids most of such problems. Thus, the Binford's exercise was primarily methodological in its contribution (which they readily admit; 1966: 289) . It shows the kinds of explanations that could be offered about Mousterian var i a b i l i t y given a processual perspective, but there is no : claim to substantive or "factual" additions to our knowledge at the time. I mean that the tool types within the five factors isolated - 1. secondary tool manufacturing (non- l i t h i c ) , 2. hunting and butchering, 3. food processing, 4. shredding and cutting, and 5. other k i l l i n g and butchering - cannot be used as interpretive units in unrelated studies with- out computing an a l l new set of factor scores. The Binfords found a way to re-interpret Bordes' type l i s t . This essentially methodological aspect is witnessed by the fact that Freeman's (.1966) factor analysis of Mousterian materials in Spain (see also Freeman 1978) failed to produce factor load- ings on tool types similar to those in the Binfords 1 study. The scope of v a r i a b i l i t y in the Spanish sites was interpreted to re- 14 s u i t from a c t i v i t i e s ranging from " s c r a p i n g " to " c u t t i n g - chopping" (Freeman 1966: 235). A very i n t e r e s t i n g s i t u a t i o n i s that i n both sets of research, the i s o l a t e d f a c t o r s show broad s i m i l a r i t i e s to Bordes' four Mousterian v a r i a n t s ( c f . Freeman 1966: 234; B i n f o r d and B i n f o r d 1966: 259). However, t h i s does not occur i n a l l f a c t o r s , and t h i s confusion i s r e c o n c i l e d by Freeman: The two models ( i . e . e t h n i c i d e n t i t y and f u c t i o n a l s p e c i f i c i t y ) are not a l t e r n a t i v e explanations of the same kinds of v a r i a t i o n . Both may be e q u a l l y c o r r e c t i n the most gen- e r a l sense, but t h e i r v a l i d i t y r e q u i r e s con- s i d e r a t i o n of d i f f e r e n t aspects of the data (Freeman 1978: 58). In a study aimed mainly at understanding the e n t i r e Paleo- l i t h i c c o l l e c t i o n from Douara Cave i n S y r i a (Hanihara and Akazawa 1979), Akazawa (1979) f a c t o r analyzed 71 Mousterian assemblages described w i t h reference to Bordes' type l i s t . These assemblages were from the Douara Cave, Yabrud S h e l t e r 1, and included Combe-Grenal from France. The f a c t o r a n a l y s i s produced f i v e f a c t o r s , and I t h i n k i t worthwhile to e x t r a c t some c r i t i c a l f i n d i n g s that are based on the p l o t t i n g of f a c t o r scores: 15 Although the Upper French Acheulian shows concrete evidence of clustering, the other five types of assemblages classifiable as French Mousterian and Yabrudian show-a wide range of distribution and overlapping. In particular, distribution of the Typical Mousterian assemblage Jisepeculiar^acharacter- ized by overlapping with assemblages class- i f i e d as another major group (Levantine). This suggests that these assemblage types have more complex features, and therefore cannot be explained simply on the basis of typological characteristics.(Akazawa 1979: 42, emphasis added). Akazawa refrains from any functional interpretations, but adds that the groupings of assemblages are due mainly to their con- tained frequencies of sidescrapers, and Levallois and dentic- ulate tools (Akazawa 1979: 42). It seems clear in this discussion of factor analysis that the tjasic functional assumptions about the use of Bordes' type l i s t are unwarranted, do not produce consistent results, and may in fact complement culture-historic interpretations. The Binfords' logical argument that ethnic group identity cannot account for differential assemblage composition has been countered by Fish (1976: 19) who notes that the MAT is common in the Dordogne region, but practically non-existent i n Charente. In this vein, however, there is the question of whether Bordes' classification i s one that would have been re- cognized by the sites' inhabitants. In societies using stone tools today (granted, with much simpler l i t h i c technologies), 16 there has been shown an appreciable difference between the producers' and the archaeologists' classifications (Gould 1972; White, Modjeska and Hipuya 1977). Bordes' classification and general scheme have other faults. For example, the rigor of the method of assemblage type definition i s weakened by the use at times of single tool types like denticulates while MAT is distinguished by several, and Typical i s a sort of catch-all. Further, there is the problem of how to classify multiple tool types occurr- ing on single artifacts. The method is not rigorous. A recent a r t i c l e that is already a classic in method (Cahen, Keeley and Van Noten 1979: also Van Noten, Cahen and Keeley 1980) describes a site in which three Upper Paleolithic end-scraper types (eight actual tools) were fi t t e d onto a single block and had a l l been used for hide-scraping (Cahen jet al 1979: 666). This may indicate that Bordes' classification i s far too finely s p l i t for functional interpretation,.".and this has serious implications for those who use the type l i s t essen- t i a l l y unaltered for functional analyses (cf. Binford and Binford 1966: 244). One.cannot deny the importance of Bordes' work - he formulated a complex, more or less objective typology that has permitted standardization and large-scale comparison and scholarly communication. His interpretations and theory however, are rather organic and undemonstrable. The ethnic hypothesis has not been completely refuted (although the temporal succession hypoth- 17 e s i s seems destroyed), s i n c e i t s t i l l needs to be te s t e d e m p i r i c a l l y w i t h the use of ethnographic data, as indeed does the f u n c t i o n a l argument. This p o i n t i s noted by B i n f o r d and B i n f o r d , who observe that " c u l t u r a l borrowing" needs to be understood since i t i s impossible to imagine mobile c u l t u r e s , depositing a l t e r n a t i n g assemblages, who never acknowleged t h e i r neighbours (1966: 240). Indeed, the B i n f o r d s apply the type l i s t to assemblages from France and the Near East, which could be i n t e r p r e t e d as a r e c o g n i t i o n of some degree of "borrowing" of b a s i c t r a i t s . There i s a l s o good evidence that hunters and gatherers do maintain s o c i a l i d e n t i t i e s w i t h s p e c i a l membership requirements (Campbell 1968; Lee and DeVore 1976; Jorgensen 1980), and that a c t u a l e t h n i c d i f f e r e n c e s can be demonstrated i n chipped stone p r o j e c t i l e p o i n t s (Magne and Matson 1982; Greaves 1982), although there have been few attempts to r e s o l v e t h i s i s s u e w i t h preservable m a t e r i a l c u l t u r e . In an a r c h a e o l o g i c a l s i t u a t i o n where the greater p a r t of the i n f o r m a t i o n i s obtained from stone t o o l s , there has been an amazing l a c k of t e c h n o l o g i c a l p e r s p e c t i v e f o r the Mousterian. I f technology i s defined as a mechanical means of a r t i c u l a t i n g human populations a d a p t i v e l y to environments which n e c e s s i t a t e movement between h a b i t a t i o n and resource l o c a t i o n s (Munday 1976: 113, c f . B i n f o r d 1962: 328), then i t i s apparent that the above s t u d i e s have not d e a l t w i t h technology. Apart from the common usage of the L e v a l l o i s index (Bordes 1972), there has been minimal c o n s i d e r a t i o n of the manufacturing processes of stone t o o l s , how 18 these might vary i n space and time, and the c o n d i t i o n s l e a d i n g to such v a r i a b i l i t y . The u t i l i t y of l i t h i c debitage i n b e h a v i o r a l r e c o n s t r u c t i o n i s now appreciated i n archaeology (see Chapter 4 ) , and debitage has been analysed i n an Upper P a l e o l i t h i c - Texas Ar c h a i c com- p a r a t i v e study by C o l l i n s (.1974, 1975). While C o l l i n s ' work was undertaken w i t h i n an e x p l i c i t b e h a v i o r a l framework a l a S c h i f f e r (1972) , F i s h (1979) conducted a study of Mousterian debitage w i t h the purpose of demonstrating " t e c h n o l o g i c a l " r e g u l a r i t i e s i n Bordes' four Mousterian types. F i s h ' s study, based on deb- i t a g e and scrapers from four s i t e s i n France and the Near East, has s e r i o u s methodological problems, such as the use of biased sampling techniques, small samples, very redundant measures, and no e x p l i c i t statement about the s i g n i f i c a n c e of L e v a l l o i s debitage ( i . e . what does core p r e p a r a t i o n i n d i c a t e ? ) i n even a h y p o t h e t i c a l manner. On i t s p o s i t i v e s i d e , F i s h ' s study shows that debitage v a r - i a b i l i t y does not correspond w e l l to t y p o l o g i c a l Mousterian f a c i e s . Nevertheless, the removal of t o o l s from s i t e s i s a f a c t o r not w e l l c o n t r o l l e d i n t r a d i t i o n a l and -behavioral s t u d i e s . For example.,, at Pech de l'Aze Bed 4, 43% of the debitage analysed were f l a k e s of b i f a c i a l retouch, yet only 5.8% of the t o o l s were handaxes; and i n Bed 28 at Combe-Grenal, no handaxes were found, yet 13.5% of the debitage were b i f a c i a l retouch f l a k e s ( F i s h 1979: 133). Thus, we.: can see an advantage of debitage a n a l y s i s ; i t i s a way of r e l i a b l y 19 demonstrating that the tools l e f t at a site may not represent the f u l l range of activities that were performed there. As concerns the Levallois index, or the proportion of items in the assemblage exhibiting complex scar patterns other than deliberate retouch, I find this far too narrow a defining characteristic of "technological" differences between assem- blages. Nevertheless, Fish (1979: 128 - 130) found an inter- esting correspondence between overall low Levallois occurrence at Combe-Grenal and Pech de l'Aze and the availability of pre- dominantly small cobbles as the l i t h i c resource for these loc-^ cations. This seems to make a great deal of sense: the smaller the cobbles used for tool manufacture, the more- . conservative tool-making w i l l result in complex "exhausted" debitage. Obviously this has implications for the use of standardized tool typologies. It can be expected that tools near locations where only small size raw materials are present, would be less "expedient" than those with larger cobbles available, but also that scarce l i t h i c re- source areas would tend to exhibit more tool curation, paradox- i c a l l y resulting in small, highly complex retouched items. For the Mousterian, an explicitly technological approach is that of Munday (1976). Working in the area of the Negev desert of Israel, Munday sought to explain the variable comp- sition of open-air sites, rather than cave or shelter sites that are typical of Mousterian studies (but see Binford and Binford 1966: Houpeville). Jelinek (1976) notes that the "value of ±n situ deposits in open sites (with few exceptions) 20 i s i n the c l e a r f u n c t i o n a l a s s o c i a t i o n of elements of s i n g l e or t r a d i t i o n a l l y l i n k e d occupations" ( J e l i n e k 1976: 23; c f . Wobst 1979)!. Among 11 s i t e s , Munday examined q u a n t i t a t i v e r e l a t i o n s h i p s among cores and debitage, and f l i n t and water resources. Using m u l t i p l e r e g r e s s i o n a n a l y s i s , i t was found that debitage s i z e and core weight are h i g h l y r e l a t e d to s i x independent v a r i a b l e s that e s s e n t i a l l y c o n t r o l the amount of work inv o l v e d i n moving raw m a t e r i a l s between s i t e s : raw mat- e r i a l d i s t a n c e , slope of s i t e to raw m a t e r i a l , distance to water, slope to water, a l t i t u d e to water and a l t i t u d e to raw m a t e r i a l . These f a c t o r s accounted f o r 90% of the v a r i a b i l i t y i n debitage, and 80% of the v a r i a b i l i t y i n core weight (Munday 1976: 139). In r e l a t i o n to F i s h ' s (.1979) f i n d i n g s regarding L e v a l l o i s technique v a r i a b i l i t y , Munday (.1976: 139) found that at s i t e s f a r from raw m a t e r i a l "more i n t e n s i v e core p r e p a r a t i o n took p l a c e , as e x h i b i t e d by the more complex q u a l i t a t i v e t e c h n o l o g i c a l v a r i a b l e s ( p l a t f o r m types, d o r s a l scar p a t t e r n i n g and d o r s a l scar count) found w i t h the r e s u l t a n t debitage" (1976: 139). This i s explained w i t h reference to the p r i n c i p l e of l e a s t e f f o r t ( Z i p f 1949), or simply, i t seems that assemblage composition i s s t r o n g l y a f f e c t e d by economizing behavior. We thus see i n the Mousterian example that a r c h a e o l o g i c a l method has developed to seeking ways of r e c o n s t r u c t i n g b e h a v i o r a l s i t u a t i o n s that lead to inter-assemblage v a r i a t i o n s i n l i t h i c contents. The c u l t u r e - h i s t o r i c and e t h n i c paradigms of archaeo- 21 l o g i c a l remains are not completely r e f u t e d i n the theo- r e t i c a l sense and are e s s e n t i a l c o n s t r u c t s i n both the Old and New Worlds, even i f only as convenient ways of d e s c r i b i n g l a r g e - s c a l e e v o l u t i o n a r y trends. The t r a d i t i o n a l paradigms are b a s i c a l l y complementary to the b e h a v i o r a l viewpoints, but are a l s o i n need of methodological improvement. 2.3. L i t h i c s , L o g i s t i c s and L i v e l i h o o d 2.3.1. I n t r o d u c t i o n In many respects the a r c h a e o l o g i c a l i n t e r e s t s i n stone t o o l s .as i n d i c a t o r s of subsistence and s o c i a l a c t i v i t i e s i n the 1960's were unable to answer w i t h much c e r t a i n t y the questions they posed. Much of B i n f o r d ' s w r i t i n g through the 1970's was t h e o r e t i c a l , and the advent of systems theory a p p l i - c a t i o n s , as w e l l as a g e n e r a l l y greater p h i l o s o p h i c a l awareness co n t r i b u t e d a great d e a l to the q u a l i t y of the questions being asked. B i n f o r d ' s c o n t i n u a l r e - e v a l u a t i o n s of the s t a t e of a r c h a e o l o g i c a l conduct have r e c e n t l y been focused on p r e c i s e l y methodological problems and ways of " b r i d g i n g " e m p i r i c a l f a c t s , w i t h t h e o r e t i c a l demands. This process he c a l l s "middle-range theory b u i l d i n g " , and he considers i t to/represent a major s h i f t i n s c i e n t i f i c a r c h a e o l o g i c a l terms of reference, models and paradigms ( B i n f o r d 1972, 1977; Kuhn 1962). 22 Throughout the development of b e h a v i o r a l research w i t h stone t o o l s , there i s an i n t e r e s t i n g i n t e r p l a y of i n d u c t i v e and deductive reasoning, that i s ; between questions they ask, i n the extreme: What can t h i s stone t o o l t e l l us about be- h a v i o r , and; What are the i m p l i c a t i o n s of behavior f o r stone t o o l s ? However, I choose not to review s e p a r a t e l y here the philosophy of s c i e n t i f i c reasoning and the many t h e o r e t i c a l i n t e r f a c e s of analogy, experimentation and ethnoarchaeology. Recent •comprehensive d i s c u s s i o n s by Charlton (1981) and Salmon (1982) r e v e a l that the issues are complex and beyond the scope of t h i s study. 2.3.2. Experimental l i t h i c research R e p l i c a t i o n and s i m u l a t i o n experiments i n stone t o o l manufacture g e n e r a l l y seek to r e l a t e q u a n t i t a t i v e and q u a l - i t a t i v e v a r i a b i l i t y i n t o o l s and manufacturing by-products (debitage) to processes of production, use and d i s p o s a l . While the designs of such experiments, and the methods of a n a l y s i s vary a great d e a l , most aim to increase the r e l i a b i l i t y of be- h a v i o r a l inferences that can be made on the ba s i s of archaeo- l o g i c a l m a t e r i a l p a t t e r n i n g (see Tringham 1978: Charlton 1981: 146 - 147). C u r r e n t l y , b e h a v i o r a l i n s i g h t s provided by system- a t i c stone t o o l r e p l i c a t i o n (e.g. Muto 1971a; Crabtree 1972) are a s i g n i f i c a n t part of many r e c o n s t r u c t i o n s or r e g i o n a l r e l a t i o n - ships between l i t h i c technology and settlement patterns (e.g. Knuds 23 1973; Gardener 1976; Katz 1976; Phagan 1976; Chapman 1977; Pokotylo 1978a; Kimball 1980). The general model of stone tool technology employed in this research was roughly developed by Holmes (1890) and has been refined by several researchers (Collins 1974, 1975; Bradley 1975; Gunn 1975; Sheets 1975; Katz 1976; Schiffer 1976; Pokotylo 1978). Flow charts are used to model the various steps involved in stone tool manufacture, use, modification and dis- posal, and are linear in nature because output products cannot resume a previous state. Perhaps the clearest and most useful such model is that proposed by Collins (1974, 1975), here summarized in Figure 1. In Collins' (1975) model, the f i r s t step in making a stone tool is aquiring the raw material. As Binford (1979) has pointed out, this activity can be embedded in other subsistence tasks, or i t can be a direct, special purpose task such as in v i s i t s to quarry locations. The next step in the model is to prepare cores and reduce them. Here, the cores themselves may be desired pro- ducts, or flakes removed from them can be used as tools, or as blanks for further reduction. Following core or flake blank pro- duction, primary trimming may produce useful tools and/or preforms ("unfinished" , tools). The next step is secondary trimming, to pro- duce complex tools, hafting provisions, serations, aesthetic flake scar patterns, and so on. Collins' model recognizes the use of tools as a distinct step in their modification, and following 24 Figure 1. The general model of l i t h i c r e d u c t i o n , main- tenance and d i s p o s a l . Revised a f t e r C o l l i n s (1975) . 25 use, t o o l s may be resharpened, or s u b s t a n t i a l l y r e f u r b i s h e d . The f i n a l p o s s i b l e step i n t h i s model i s s p e c i a l i z e d d i s p o s a l of a r t i f a c t s , i n caches, as grave goods, and;the l i k e . In g e neral, a great d e a l of progress has been made i n the l a s t 20 years of a century of l i t h i c experimentation (see Johnson 1978), but s e v e r a l problems p e r s i s t i n experimental stone t o o l s t u d i e s : 1. There i s a se r i o u s l a c k of adequate experimental c o n t r o l s (see Chandler and Ware 1976: 25; Dincauze 1978) . B a s i c a l l y t h i s problem stems from the h i s t o r y of viewing stone f l a k i n g , or f l i n t k n a p p i n g as an " a r t " r a t h e r than a s c i e n t i f i c endeavor. L i t h i c use-wear a n a l y s t s have c l e a r l y recognized the value of e x p l i c i t experimental c o n t r o l s (e.g. Keeley 1980; Tringham et a l 1974; O d e l l 1977) , and only r e c e n t l y have l i t h i c r e d u c t i o n ex- periments been conducted w i t h f i r m c o n t r o l s (Speth 1972; Raab et a l 1979; Burton 1980; Stahle and Dunn 1981). 2. B i f a c e s and p r o j e c t i l e p o i n t s are the u s u a l subjects of i n - v e s t i g a t i o n s (e.g. Newcomer 1971; Callahan 1977; Flenniken 1978), and there i s a l a c k of experimentation aimed at s y s t e m a t i c a l l y understanding the manufacture of many other t o o l forms, or the f u l l range of r e d u c t i o n processes. Again, t h i s i s r e l a t e d to the h i s t o r i c a l problem of d e a l i n g w i t h items that are perhaps best s u i t e d to t y p o l o g i c a l i s s u e s , yet even some of the b e t t e r c o n t r o l l e d experiments such as Stahle and Dunn (1981) propose to somehow c h a r a c t e r i z e e n t i r e debitage assemblages only" by d e a l i n g with, b i - face production r e s i d u e s . 26 3. There i s a strong tendency to use redundant a n a l y t i c v a r i a b l e s , e s p e c i a l l y s i z e v a r i a b l e s (e.g. F i s h 1976, 1979, 1981), or morphological v a r i a b l e s w i t h l i t t l e e x p l i c i t ana- l y t i c value (e.g. Patterson and S o l l b e r g e r 1978; Burton 1980) to d e s c r i b e debitage v a r i a b i l i t y ; This problem i s by no means l i m i t e d to l i t h i c a n a l y s i s , s i n c e some a r c h a e o l o g i s t s study as many v a r i a b l e s as p o s s i b l e i n the hope of d e r i v i n g meaningful p a t t e r n s , and o f t e n a s s i g n meaning to v a r i a b l e s only a f t e r patterns have been detected. At the l e a s t , expectations of v a r i a b l e p a t t e r n i n g should be proposed p r i o r to completing a set of experiments. 4. Many experiments do not i n c l u d e s t a t i s t i c a l e v a l u a t i o n s of research f i n d i n g s i n t h e i r design, and i n t e r p r e t a t i o n s are o f t e n s u b j e c t i v e (e.g. Muto 1971a; Kobayashi 1975; Pat t e r s o n and S o l l - berger 1976, 1978; Flenniken 1980). This i s a s e r i o u s problem but one that i s being resolved as a r c h a e o l o g i s t s gain greater f a m i l i a r i t y w i t h q u a n t i t a t i v e methods (e.g. Chandler and Ware 1976; S t a f f o r d and S t a f f o r d 1979; Stabl e and Dunn 1981). 5. The f i e l d i s very p a r t i c u l a r i s t i c , and experimental r e s u l t s are u s u a l l y a p p l i e d to s m a l l - s c a l e a r c h a e o l o g i c a l samples such as s i n g l e s i t e s , or are used with, l i t t l e other purpose than to demonstrate that c e r t a i n techniques of t o o l manufacture were used i n the past (e.g. Crabtree 1966, 1968; Callahan 1977; Flenniken 1978). This kind of p a r t i c u l a r i s m i s necessary at a bas i c l e v e l , but the f i e l d has to generate higher l e v e l s of methodological and 27 t h e o r e t i c a l awareness i f i t i s ever to c o n t r i b u t e to archaeology i n a s c i e n t i f i c manner. Major c o n t r i b u t i o n s to l i t h i c assemblage i n t e r p r e t a t i o n s during the 1970's were made i n the area of use-wear a n a l y s i s , p a r t i c u l a r l y the work of Keeley (.1980), O d e l l (1977, 1980) and Tringham et a l (1974),.and as discussed above, the innovating work of Cahen et^ a l (1979). The l i t e r a t u r e on use-wear i s very ext e n s i v e , f a s c i n a t i n g , contentious and almost completely s i t e - p a r t i c u l a r . For these reasons, and because t h i s study i s con- cerned w i t h inter-assemblage l i t h i c manufacturing p a r t i c u l a r s and p a t t e r n s , the scope of use-wear research (see Hayden 1979) i s beyond d e t a i l e d e l u c i d a t i o n here. L i t h i c use-wear a n a l y s i s c o n t r i b u t e s g r e a t l y to the kinds; of models that a r c h a e o l o g i s t s use, s i n c e t h e o r e t i c a l l y , i f we can observe s p e c i f i c t o o l f u n c t i o n s then we should be able to measure time and energy expenditures and r e t u r n s a n d seek to model and understand ;the operations of populations i n r e l a t i o n to stone t o o l s . However, the f i e l d i s not p r e s e n t l y able to r e - solve such i s s u e s , due to d i f f i c u l t i e s i n method, and i n t e r p r e - t a t i o n of the e m p i r i c a l evidence. I t i s d i f f i c u l t to i d e n t i f y the type of m a t e r i a l worked by stone t o o l s , the motion i n v o l v e d , and l e s s so, the general hardness of worked m a t e r i a l and a c t u a l presence of wear. The experimental foundations of use-wear a n a l - y s i s are r a p i d l y growing, and have always been s c r u p u l o u s l y r e - examined (e.g. H o l l y and Del Bene 1981; Keeley 1981). I n i n t r o - 28 ducing a volume on the then-current s t a t e of l i t h i c a n a l y s i s and p r e h i s t o r i c behavior, Davis noted that "while the major issues i n the a n a l y s i s of l i t h i c assemblage v a r i a b i l i t y are t h e o r e t i c a l , the major d i f f i c u l t i e s are methodologicalV (Davis 1978: i i i ) . I t seems there are general schools of agreement about what we want to know, we j u s t do not know how to gather the necessary i n f o r m a t i o n , nor how to i n t e r p r e t what we have. 2.3.3. Current models of l i t h i c technology and settlement p a t t e r n s A community's settlement system can be defined as f o l l o w s : ...a s o l u t i o n to the problem of l o c a t i n g s i t e s so as to minimize the amount of energy that must be expended to procure necessary resources, be t h i s by j u d i c i a l choice of a s i n g l e s i t e , l o c a t i o n of s e v e r a l s i t e s at d i f f e r e n t times i n d i f f e r e n t ^ s i t u a t i o n s , development of s t o - rage and/or p r e s e r v a t i o n techniques, or a com- b i n a t i o n s t r a t e g y (Roper 1979: 16). Hunter-gatherers are c l a s s i f i e d by B i n f o r d (1980) i n t o two b a s i c kinds of s o c i e t i e s : f oragers and c o l l e c t o r s . Foragers procure resources on a day-to-day b a s i s , do not p r a c t i c e extended food storage, and move residences o f t e n as l o c a l resources are de- p l e t e d . Examples of foragers i n c l u d e the K a l a h a r i Bushmen, (Lee and DeVore 1968, 1976; Y e l l e n 1977), and the A u s t r a l i a n Western Desert Aborigines (Gould 1969, 1980; Hayden 1976, 1977), at l e a s t those that remain and s t i l l engage i n hunting and gathering as the 29 p r i n c i p l e mainstay of t h e i r e x i s t e n c e . C o l l e c t o r s maintain r e s i d e n c e s , yet move o f t e n f o r extended periods tfo procure i n d i v i d u a l or sets of resources, r e t u r n i n g to the residences. Food storage p r a c t i c e s are v a r i e d and common, and more seasonal extremes i n subsistence a c t i v i t i e s are e x h i b i t e d . For example, the Nunamiut Eskimo ( B i n f o r d 1977, 1978a, 1978b), the B o r e a l Forest Cree (Bishop 1974, Leacock 1973, Rogers 1973) and most temperate hunters and gatherers, i n c l u d i n g the groups i n h a b i t i n g the B.C. I n t e r i o r P l a t e a u , may be considered resource c o l l e c t o r s . C o l l e c t o r s employ at l e a s t f i v e kinds of s i t e s : 1. r e s i - d e n t i a l bases; 2. l o c a t i o n s (of k i l l s or g a t h e r i n g s ) ; 3. f i e l d camps, f o r task groups; 4. s t a t i o n s , where informa t i o n i s gathered and scheduling d e c i s i o n s are made; and 5. caches ( B i n f o r d 1980: 10). B i n f o r d does not discuss r i t u a l i s t i c s i t e s such as petroglyphs or b u r i a l s , although the nature arid d i s t r i b u t i o n of these i s to a c e r t a i n extent conditioned by m o b i l i t y . P r a c t i c a l l y every kind of s i t e can be re-occupied f o r another purpose, andcthis i s a po- t e n t i a l way to measure settlement m o b i l i t y , s i n c e greater i n t e r - assemblage v a r i a b i l i t y can be expected, the greater the number of "generic f u n c t i o n s " a s i t e undergoes ( B i n f o r d 1980: 12). Thus, the a r c h a e o l o g i c a l problem i s to "develop a means of i d e n t i f y i n g generic types of f u n c t i o n a l d i f f e r e n t i a t i o n when they are encountered i n the a r c h a e o l o g i c a l r e c o r d " ( B i n f o r d 1979: 271). The problem i s to observe the outputs of human behavior i n c o n t r o l l e d c o n d i t i o n s , where the systemic context i s known. 30 Binford's method is ethnoarchaeology, the conduct of ethno- graphic research to solve archaeological, anthropological and even sociological issues (see Gould and Schiffer 1982; Gould 1978; Kramer 1979). Although Binford does not seem to recog- nize experimental archaeology as a means of approaching the same problems, there are several common theoretical grounds to ethnoarchaeology and experimental archaeology (Tringham 1978). Tringham views both as "experimental" research, but considers that "behavioral experimentation"' is riskier, since variables are more d i f f i c u l t to control. Armed with the experience gained in his intensive Nunamiut ethnbarchaeological research with faunal assemblages (Binford •1978a; 1981) and with the success of that research in providing empirical evidence of subistence patterns operations, Binford has again focused on l i t h i c assemblages (Binford 1979). The im- plications of the l o g i s t i c and highly mobile Nunamiut settlement pattern for l i t h i c technological inference are highly relevant to this and other explicitly technological studies of chipped stone, and are worth citing at length here. In his reconstruction of how the Nunamuit used stone tools, Binford (1979) relies heavily on interviews with elders, and practically n i l on observation, since the Nunamiut have long since abandoned stone tools, except for large hammers and anvils, bed warmers, and occasional instances of "survival gear". The 31 informants agreed that three basic kinds of gear are used, past or present. Personal gear and site furniture are anticipatory items, and situational gear is responsive in nature. As far as lithi.cs- are concerned, the following are con- sidered to have been personal gear: side-bladed tools to cut bone, cores used as sources of flakes for butchering or for manufacture into scrapers, axes, bows and arrows, stone points for bears (bone, antler or wood otherwise), pressure flaking tools with hafted scrapers on the end opposite the flaking end, and single flake knives. Personal gear is curated, being recycled, reused, and maintained. It is always brought into the f i e l d in good condition, and Binford (1979: 263) deduces from this that personal gear should be largely discarded in residential camps, and not at the locus of use. It seems to me however, that some personal gear probably included items intended only for use at a distant locus, and that this material is meant to be l e f t behind, and only i f not used, would be.returned to res- idential camp. Binford's expectation also does not include the breakage of personal gear beyond repair, or the disposal at least of fragments. Keeley (1982), for example, would main- tain that the haft portion only of a complex item of personal gear would be returned to replace a stone piece. 32 Site furniture "belongs" to a site, and is available for use by any occupants (Binford 1978a: 339). Such items as large bone-cracking rocks, anvils, hearthstones, heavy marrow scrapers, and tent weights are common'.items of site furniture. These objects enter the archaeological record.when a site f a l l s into disservice, or as natural processes remove them from the active system. Site furniture, in Schiffer's (1972, 1976: 14) terms, is de facto refuse, usable gear that is abandoned. Situational gear is task-specific. For example, Binford (1979: 266) relates an anecdote of two hunters, hunting caribou and needing knives to butcher the animals, but lacking a good steel knife. One of them found suitable rock, broke i t , and they used the resulting flakes. No great deal of effort was expended, but i t provided gear suitable for the task at hand. In certain situations, personal gear or gear that has been cached can be modified for the required purpose. What of the interrelationships of these kinds of gear? Bin- ford writes: ...we can expect assemblages which are "curated" in the broad sense to exhibit patterns of inter- assemblage v a r i a b i l i t y depending upon the organ- ization of the technology as seen in the propor- tion of situational to more curated types of gear (.1979: 269) . The notion of proportional relationships i s important, since sites can be re-occupied or abandoned independent of l i t h i c 33 technology. The mere presence of any p a r t i c u l a r gear i s i n s u f f i c i e n t evidence of the purpose of the occupation. The general a r c h a e o l o g i c a l goal then, i s to re c o n s t r u c t the archaeology of s p e c i f i c " p l a c e s " , bytstudying the i n t e r - r e l a t i o n s h i p s of l i t h i c s , fauna, e t c . and t h e i r s p a t i a l d i s t r i b - u t i o n ( B i n f o r d 1982). I s B i n f o r d r e - i n v e n t i n g settlement p a t t e r n archaeology ( c f . Gummerman 1971; Euler and Gummerman 1978), by asking a r c h a e o l o g i s t s to consider the i n t e r - r e l a t i o n - ships of s i t e s ? B i n f o r d o f f e r s s e v e r a l d e t a i l e d , p a r t i c u l a r l i t h i c tech- n o l o g i c a l expectations or "probable consequences" of va r i o u s "systems c o n d i t i o n s " of the Nunamiut settlement system: 1. Items of personal and household gear are apt to be both produced and maintained w i t h i n r e s i d e n t i a l s i t e s , r e s u l t i n g i n an a s s o c i a t i o n at such l o c a t i o n s of deb r i s from manufacture, r e p a i r , and f i n a l d i s c a r d of worn-out items. 2. Items that have r e l a t i v e l y long use l i v e s are not l i k e l y to be "worn out" at s p e c i a l pur- pose l o c a t i o n s , s i n c e p r e t r i p gearing-up oper- a t i o n s would r e s u l t i n the replacement of hea- v i l y worn items before l e a v i n g the r e s i d e n t i a l l o c a t i o n . 3. Manufacturing d e b r i s from l i t h i c processing i s apt to vary i n content seasonally (represent- ing d i f f e r e n t p r o p o r t i o n s of d i f f e r e n t sources), since there i s l i k e l y to be sea s o n a l l y v a r i a b l e e x p l o i t a t i o n of d i f f e r e n t geographical areas and l i t h i c raw m a t e r i a l s would g e n e r a l l y be obtained w i t h i n the context of normal subsistence procure- ment schedules. Given r e s i d e n t i a l m o b i l i t y , l i t h i c source v a r i a b i l i t y as i n d i c a t e d i n primary debris, should he c o r r e l a t e d with, the geographical pos- i t i o n of the r e s i d e n t i a l s i t e . 34 4. Manufacturing d e b r i s o c c u r r i n g on s p e c i a l purpose s i t e s which, are intermediate between r e s i d e n t i a l s i t e s and procurement s i t e s (such as hunting stands or camps) may w e l l e x h i b i t considerable l i t h i c d e b r i s from work on p a r t - i a l l y f i n i s h e d or "staged" items. Flakes or (.sic) b i f a c i a l retouch, core r e d u c t i o n , or the use of a " d i s p r o p o r t i o n a t e " number of t o o l s designed f o r the m o d i f i c a t i o n of other raw m a t e r i a l s such as wood, a n t l e r , bone or f i b e r might w e l l be a n t i c i p a t e d . On such "intermed- i a t e " l o c a t i o n s , work scheduling would gener- a l l y be c a r r i e d out i n "dead time" on items introduced i n a n t i c i p a t i o n of t h i s a c t i v i t y (see B i n f o r d 1978a). This means that many " i n - complete" items would be f u r t h e r modified on such l o c a t i o n s , r e s u l t i n g i n " d i s j u n c t i v e " de- b r i s to t o o l r e l a t i o n s h i p s . 5. The highest incidence of r e c y c l i n g and r e - use of items of personal gear i s most l i k e l y to occur i n s p e c i a l purpose l o c a t i o n s . This f o l - lows from the observation that personal gear i s f r e q u e n t l y " d r a f t e d " f o r use" as the source of m a t e r i a l f o r s i t u a t i o n a l gear. 6. High incidences of f l a k e s from b i f a c i a l "cores" are apt to c h a r a c t e r i z e s p e c i a l pur- pose s i t e s . Such f l a k e s can be expected to show r e l a t i v e l y high use r a t i o s , that i s , the number evidencing use should be high. 7. We might expect a general inverse r e l a t i o n - ship between the proportions of reuse and r e - c y c l i n g of personal gear and the abundance of s i t u a t i o n a l l y produced gear from immediately a v a i l a b l e raw m a t e r i a l s ( B i n f o r d 1979: 269 - 270). These expectations are admittedly not exhaustive, and some are of vague u t i l i t y . For example, i t i s no problem to recognize b i f a c i a l d e b r i s , but what e x a c t l y served as personal gear, s i t e f u r n i t u r e or s i t u a t i o n a l gear i s not c l e a r from B i n f o r d ' s arguments. 35 Ethnoarchaeological reasoning has also been used by Ebert (.1979) to suggest that indices of tool sizes and com- plexity are better indicators of group mobility, tool cur- ation behavior, and specific activities than traditional typological means of analysis. This research, conducted among the Botswana Bushmen, was aimed at providing some generalizations about stone tool use, discard, and loss by observing situations in which steel tools are employed today. The two major "bridging assumptions" between group mobility and l i t h i c assemblages that Ebert addresses are: 1. Tools manufactured with the object of being carried out are expected to be smaller than tools intended to be used in one place.* 2. Tools intended for multiple episodes of use are expected to be the result of greater input of energy during manufacture and main- tenance than tools used once and then dis- carded (Ebert 1979: 68). Thus, gear that is analogous to Binford's personal gear should be small and complex, while expedient tools and habitation site maintenance tools should be larger and simpler. Site furniture should also be relatively complex. The practical ways to measure such differences are: for complexity, the frequencies of tools' flake scars produced dur- ing manufacture and maintenance, and for size, the product of length, width and thickness measures. It is suggested later in 36 t h i s study that a more appropriate measure of a t o o l ' s s i z e i s i t s weight, and that scar counting r e q u i r e s r i g i d c u t - o f f s i n s i z e and c o n t i n u i t y . L i k e B i n f o r d ' s argument, Ebert's (.1979) makes the point that i t i s o v e r a l l assemblage v a r i a t i o n that i s important and that i n d i v i d u a l t o o l measures are secondary. The p r e d i c t i o n s that Ebert makes about the r e l a t i o n s be- tween l i t h i c assemblages and settlement m o b i l i t y are l e s s de- pendent on a b s t r a c t c o n s t r u c t i o n s , and may be summarized as f o l l o w s : 1. Assemblages w i t h small t o o l s e x h i b i t i n g high manufacture/maintenance energy inputs are essent- i a l l y composed of "curated, s m a l l , s p e c i f i c - u s e t o o l s , p o s s i b l y pieces of a mobile t o o l k i t . Used i n jobs or tasks i n which a s p e c i f i c set of oper- a t i o n s i s c a r r i e d out" (Ebert 1979: 68). 2. Large t o o l s w i t h complex r e d u c t i o n p a t t e r n s are " s p e c i f i c use or s p e c i f i c job t o o l s probably not transported as f a r as those £that are small and complex^ , but curated" (Ebert 19.79: 69). 3. Small t o o l s w i t h low scar counts are "expedient, s i n g l e - u s e , immediately discarded t o o l s " (Ebert 1979: 69). Ebert suggests that small s i z e here may i n d i c a t e "raw m a t e r i a l s t r e s s " , but i n such a case, complexity would be expected to r e s u l t from extended maintenance. Thus, t h i s : e x p e c t a t i o n i s ambiguous. 37 4. Ebert's. f i n a l e x p e c t a t i o n i s r a t h e r weak, a l s o , s t a t i n g that l a r g e , simple t o o l s "should be manufactured e x p e d i e n t l y , used only once, and not tr a n s p o r t e d " (J.979: 69); however, I t h i n k that m u l t i p l e uses of l a r g e items seem l i k e l y , over extended periods of time. The method re q u i r e s each t o o l to be p l o t t e d , w i t h respect to s i z e and "energy" axes, observing the trends f o r each assemblage, p a r t i c u l a r l y predominant extreme trends, and i n - f e r r i n g the r e l a t i v e d u r a t i o n and i n t e n s i t y of the a c t i v i t i e s that produced them. This seems s t r a i g h t f o r w a r d enough, but a c l o s e look at Ebert's model re v e a l s a se r i o u s flaw i n h i s i n t e r p r e t a t i o n s of two Botswana Middle Stone Age s i t e s . In t h i s case, the scales of examination of t o o l s i z e s are d i f f e r e n t by a f a c t o r of s i x (1979: 70). This e r r o r i s i l l u s t r a t e d i n Figure 2. Ebert's s c a l e of comparison f o r s i t e KP47 encompasses that f o r s i t e KP48. Thus, the i n t e r p r e t a t i o n s that KP47 r e s u l t e d from energy invested i n mobile t o o l s , and .that KP48 i n d i c a t e s a minimum amount of energy invested i n "medium-sized" and non-portable t o o l s (Ebert 1979: 70), should be reversed. KP47 seems to be a longer term occupation, or re-occupation, kind of s i t e , whereas KP48 i s more r e s t r i c t e d i n v a r i a b i l i t y , and probably In " f u n c t i o n " . Ebert's model-building i s complementary to 38 Ebert's Graphs 200 400 Size I xwx 1 KP48 a ' 0 20 40 60 80 100 Z Size I x w x t T3 C O) CL K w Actual Comparison Scale Figure 2. Comparison of Ebert's inferential point swarms with actual comparative scale. 39 B i n f o r d ' s , but w i t h a method of " a n a l y t i c a l convention", or " o b s e r v a t i o n a l language" that enables us to d i f f e r e n t i a t e one kind of gear from another, and thus add p r e c i s i o n to our a b i l i t y to d i f f e r e n t i a t e s i t e s one from another ( B i n f o r d 1982) . Ebert's " c u r v e - f i t t i n g " approach i s the basic method of Bi n f o r d ' s (1978, 1981) f a u n a l a n a l y s i s techniques of recon-:: s t r u c t i n g s i t e purpose, and l i k e them (see Gould 1979), has a few problems w i t h confidence l e v e l s . However, approaches l i k e Ebert's are necessary to provide l i t h i c assemblages w i t h a "generic" taxonomy comparable to i d e n t i f y i n g s k e l e t a l e l e - ments w i t h bone fragments. These kinds of f a c t o r s are exam- ined l a t e r i n t h i s study, i n I n t e r i o r Plateau assemblages. C o l l i n s (1975) has presented a model of l i t h i c tech- nology as a subsystem of c u l t u r a l ecology that i s remarkable i n i t s g e n e r a l i t y and i n i t s b a s i s i n b e h a v i o r a l archaeology. I have reviewed the model e x t e n s i v e l y elsewhere (Magne 1978; see a l s o Pokotylo 1978), and d i s c u s s i t f u r t h e r and attempt to o p e r a t i o n a l i z e i t i n Chapter 4. Thus the present review i s b r i e f . C o l l i n s (1975: 16 - 19) argues that ongoing c u l t u r a l systems using stone t o o l s (the systemic context; S c h i f f e r 1972, 1976) produce d i s t i n c t product groups through f i v e major l i t h i c t e c h n o l o g i c a l steps: 1. a c q u i s i t i o n : of the raw m a t e r i a l ; 2. core p r e p a r a t i o n and i n i t i a l r e d u c t i o n ; 3. o p t i o n a l primary trimming; 4. o p t i o n a l secondary trimming; 5. o p t i o n a l maintenance 40 and modification. Collins supports the model in a unique cross-cultural comparison: Archaic period materials from Arenosa Shelter in Texas, and Solutrean deposits at Laugerie Haute Ouest in France (.Collins 1974) . General patterns of reduction through time are traced at both sites, but d i f - f i c u l t i e s are encountered in identifying any but the earl- iest and b i f a c i a l reduction strategies. Substantive findings include^ that even in the Solutrean, known for i t s fine b i - faces, less than 20% of the tool kits were produced by secon- dary trimming. Furthermore, the Arenosa Archaic assemblages average some 20% more secondary trimming debitage than Laugerie Haute Ouest (Collins 1974). The value of l i t h i c debitage in revealing basic stone tool manufacturing patterns is made explicit: If isolated, product groups can be described in terms of their technological attributes and inferences can be drawn concerning the specific activities by which the particular manufacturing step was accomplished. The waste, or debitage is particularly amenable to this technological analysis (Collins 1975: 17). Without ethnographic analogy, the "specific a c t i v i t i e s " are purely l i t h i c technological, and indeed such findings as that b i f a c i a l manufacture in the Archaic levels was far more efficient than in the Solutrean occupations (.27 bifaces per 100 secondary stage flakes versus 2 per 100) and the observation 41 that Solutrean debitage is more variable than the Archaic material, are valuable. The further meaning of this pattern in terms of human evolution is not developed by Collins, but with a larger number of such analyses, is quite possible, given such discussions of Old and New World similarities in cultural '.evolution as that of Hayden (1981) . The assemblages analysed by Collins seem to indicate that Solutrean groups were less l o g i s t i c a l l y organized, and more residentially- based than Archaic groups, and more expedient and less cur- ative with their gear. However, bifaces may have been more specialized, and more apt to removal from sites in the Solutrean, resulting in the small number of such items in comparison to late stage debitage. Collins' application of his model was fraught with tech- nical d i f f i c u l t i e s , particularly in data reduction, attribute selection and stage inferences. Successful refinement in method and outlook, but maintaining the behavioral approach and the general ^reduction model provided by Collins (1975), was achieved in Pokotylo's (1978) studies of the Upper Hat Creek Valley of British Columbia. Pokotylo's concern was with ex- plaining the "dispersal or aggregation of l i t h i c reduction steps at different site locations" (.1978: 163) within the valley. Since this meant having in hand some means of measuring reduction steps, a sample of archaeological debitage (in contrast to exper- imental materials as in Collins 1974) was factor analysed to yield 42 a reduced number of variables l i k e l y to yield stage data. The tool data, as morphological types within raw material classes, were analysed separately to provide information on basic use- related patterns. Overall, the Hat Creek data exhibit high v a r i a b i l i t y . Five separate debitage site groupings exist, ranging from single event, situational kinds of assemblages, to quarry-like, to late-stage maintenance and occupation patterns (Pokotylo 1978). The tool data also revealed five patterns of deposition, and again these are highly variable. Several sites of the 42 are expedient, while many are abundant, high diversity assemblages, and microblade assemblages occur relatively frequently. Pokotylo's (1978) "experimental" method of defining appro- priate debitage attributes was similar to that of Katz (1976), who studied reduction stages of several Kansas City Hopewell assemblages. Both researchers solved the problem of providing behavioral analogs for chipped stone processes by interpretation of patterns within a small archaeological sample. This kind of method has the advantage of limiting extraneous, knapper-specific bias, but lacks generality to other assemblages. Thus, within each of Pokotylo's and Katz' f i n a l assemblage groupings, ambig- uities occur (see Chapter 4) which are d i f f i c u l t to explain. Nevertheless these studies improved significantly on approaches of Fish (1976: 1981) and Collins (.1974), which are typified by analysis of redundant variables. 43 The c o n t r i b u t i o n of Pokotylo's r e s e a r c h i s that the operation of two major processes of. assemblage formation, has been demonstrated: d e p o s i t i o n of manufacture d e b r i s , and post-use d e p o s i t i o n of t o o l s . I t i s apparent that these pro- cesses a r e - p a r t i a l l y independent, but when combined, y i e l d p atterns i n t e r p r e t a b l e as " s i t e u t i l i z a t i o n " (Pokotylo 1978: 321 - 322). The patte r n s are a l s o f a r more u s e f u l than those obtained using e i t h e r process by i t s e l f . The l i t h i c sub-system as a whole i s dependent on the settlement system, but the i n t e r - p r e t a t i o n s of s i t e occupation purposes are not as " f i n e - g r a i n e d " as p r e c i s e i d e n t i f i c a t i o n of l a r g e mammal or f l o r a l resource a c q u i s i t i o n , p rocessing, s t o r i n g , consuming and d i s p o s a l would a l l o w , s i n c e use-wear, f a u n a l a n a l y s i s and other n o n - l i t h i c e v i - dence are not part of the argument. However, i n the "generic" sense (basecamp, staging camp, hunting/butchering, s p e c i a l pur- pose), Pokotylo's (1978) r e c o n s t r u c t i o n of the Hat Creek s e t t l e - ment system was h i g h l y s u c c e s s f u l . The e n t i r e combination of t e c h n o l o g i c a l d e t a i l , at a r e g i o n a l l e v e l , w i t h an environmentally- s t r a t i f i e d set of abundant l i t h i c assemblages, w i t h a f a i r degree of ethnographic analogy, i s an a r c h a e o l o g i c a l precedent i n Canada comparable to e a r l i e r systemic Great Basin s t u d i e s (Matson 1971; Thomas 1973) . I t i s i n t e r e s t i n g to observe the s i m i l a r i t y i n s t r u c t u r e of the research being undertaken b y . C o l l i n s , Pokotylo, Katz, Fish, and the current study. A l l f i r s t make c l e a r they are operating under 44 the assumptions and l i m i t a t i o n s of a general l i t h i c r e d u c t i o n model, then propose means to measure the d i s t r i b u t i o n of v a r i o u s "stages" or s t a t e s of complexity. F i n a l l y , the analyses performed are m u l t i v a r i a t e and m u l t i d i m e n s i o n a l , o f f e r i n g comparisons from the l e v e l s of i n t e r - f e a t u r e , i n t r a - s i t e , to i n t r a - r e g i o n a l , and i n t h i s study, i n t e r - r e g i o n a l . S e v e r a l researchers acknowledge that l i t h i c technology, when a p p r o p r i a t e l y described, not only o f f e r s clues as to the operation of the l a r g e r settlement system, but i s a l s o a resource procurement and processing sub-system that i t s e l f poses c o n s t r a i n t s on the l a r g e r system, and at i t s s c a l e , i s worthy of a n a l y s i s as an economy (Singer and E r i c s o n 1977; Goodyear 1979; Gardener 1976). The p r i n c i p a l argument i s that patterns of m o b i l i t y can be t i e d to constant l i t h i c sources that are g e o l o g i c a l l y d i s t i n c t , and that succeeding r e d u c t i o n stages should be h i g h l y determined by conservation and d i s t a n c e to sources. Goodyear (1979) presents the general case f o r Paleo-Indian uses of v a r i o u s raw m a t e r i a l s , where the s i t u a t i o n was that l i t h i c resources e x h i b i t e d "some severe s p a t i a l incongruencies" w i t h l o - c a t i o n s where the stone was a c t u a l l y used. Note the c o n t r a s t be- tween t h i s and B i n f o r d ' s (1979) "embedded" argument f o r the Nuna- miut, i n that the Paleo-Indian a c q u i s i t i o n of stone i s a s p e c i a l purpose task. Yet Goodyear's pe r s p e c t i v e i s s i m i l a r to B i n f o r d ' s i n a l s o being i n t e r e s t e d i n the o r g a n i z a t i o n of curated or " c a r r y - i n g " t e chnologies. Furthermore, B i n f o r d (1979), Goodyear (.1979) , 45 and Pokotylo (1978) and others see the need to model l i t h i c technology as a flexible, situationally responsive^ means of solving other resource-related problems, yet as made clear by Collins (.1975), the operation of the technology is in many ways independent of - the subsistence and/or settlement model under investigation. An important substantive implication of Goodyear's con- siderations is that on a continental scale, Paleo-Indians ex- hibit ranges (i.e. diameters) of mobility of 100 to 200 miles (.160 to 320 km), and that during the following Archaic periods, raw material use becomes increasingly local, indicating de- creased mobility (Goodyear 1979: 9 - 10). Goodyear uses his arguments to propose the hypothesis that: Among mobile hunter-gatherers, the use of cryptocrystalline raw materials is a strat- egy for creating portable and flexible technologies to offset geographic incon- gruencies between resources and consumers (Goodyear 1979: 12) . This kind of "economic" model is explicit in Gardener's work on the Flint Run Complex (Gardener 1976) as well as in the research of Kimball (1980), Raab, Cande and Stahle (1979), Chapman (1977), and Singer and Ericson (.1977) , who investigate changes in patterns of reduction, mainly of bifaces, through time and space. The recent surge of technological awareness in l i t h i c studies, in contrast to the technical emphasis of replication studies (see 46 Chapter 4), poses methodological problems for prehistoric archaeologists wishing to reconstruct hunter-gatherer patterns of movement, and the evolution of such patterns. With respect to this study, the technological expectations of Nunamiut settlement and subsistence patterns can be contrasted with those of Interior Plateau ethnographic observations. Since Binford is aware of the biases that can be introduced by "extreme cases" such as the Nunamiut (1979: 255), i t can be suggested that Interior Plateau peoples have been less mobile or differing from the Nunamiut in certain respects, and appropriate variations in l i t h i c technology can be tested for in this study. This assumes that Binford's (1979) information i s accurate and not induced to the informants, and demonstrating common features i n the two systems may be rendered d i f f i c u l t by a lack of direct ethnographic evidence from contemporary Plateau peoples. Cross-cultural research in l i t h i c technology has reached a level of awareness such that "traditional" problem assemblage complexes appear to exhibit patterns comparable to independently derived deductions. One such pattern or problem i s that revealed in the Mousterian discussion above and reinforced by Goodyear (1979), where raw material availability strongly determines the character of l i t h i c assemblages, perhaps even masking settlement factors. Binford concludes that major rethinking of current approaches to l i t h i c technology is required, especially in the areas of ...'cost/benefit' analysis of l i t h i c source reduction strategies, raw materials, tool design, recycling, reuse j.̂ and the relative contributions of each to 'assemblage var i a b i l i t y ' . We should expect different designs and reduction strategies for functionally similar tools, depending upon their intended technological roles, given variable situations of tool demand and adequate gear provisions (Binford 1979: 271). 47 The Nunamiut settlement system i s h i g h l y l o g i s t i c . Men t r a v e l l i n g long distances on caribou hunts i n a severe environment r e q u i r e caches and good knowledge of t h e i r l o - c a t i o n s , to insure t h e i r hunting endeavors against a c c i d e n t s , breakage, and to l i g h t e n , loads. B i n f o r d (.1979: 258) estimates that at any p o i n t i n time, about 60% to 70% of the gear known to a Nunamiut man i s p a s s i v e , or not i n use. Thus, p r o v i d i n g that the I n t e r i o r S a l i s h and C h i l c o t i n were somewhat l e s s " l o g i s t i c " than the Nunamiut, the amount of passive gear would be expected to be lower, s i n c e fewer types of s i t u a t i o n s may e x i s t where separate gear i s r e q u i r e d . To continue e x p l o r i n g and e v a l u a t i n g the a r c h a e o l o g i c a l foundations of t h i s study, the f o l l o w i n g chapter presents the ethnographic and a r c h a e o l o g i c a l contexts of research i n the c e n t r a l and southern I n t e r i o r of B r i t i s h Columbia. I t h i n k that the ethnographic l i t e r a t u r e , p a r t i c u l a r l y of the P l a t e a u , has not been s u f f i c i e n t l y recognized i n the general archaeo-. l o g i c a l l i t e r a t u r e f o r t h e / d e t a i l i t provides of a f a s c i n a t i n g mobile, salmon p r o c u r i n g , hunting and gathering c u l t u r e complex. The i n f o r m a t i o n reviewed provides e m p i r i c a l evidence that bears d i r e c t l y on the t h e o r e t i c a l and methodological issues discussed above, completing the major terms of reference w i t h i n which the ensuing analyses were undertaken. 48 CHAPTER 3 ETHNOGRAPHIC AND ARCHAEOLOGICAL CONTEXTS 3.1. Regional Ethnography The archaeological assemblages examined in this study were obtained from four regions of the Interior Plateau (Figure 3) that were historically occupied by the following groups: 1. Chilcotin; 2. Canyon Shuswap; 3. Upper Lillooet; and 4. Upper or Spences Bridge Thompson (Figure 4). The Chilcotin are an interior Athapaskan speaking group and are currently the most southerly Athapaskans in Canada. The Nicola, now extinct, were the most southerly. The remaining three groups represent linguistic and t e r r i t o r i a l divisions of Interior Salish peoples. This section describes the subsistence practices and settlement patterns of these people as recorded mainly in the late 19th century, and indicates basic similarities as well as important differences in their l i f e s t y l e s . Sub-sections discuss the results of recent cross-cultural analyses and briefly review the interior ethnographic record of l i t h i c technology. The Thompson, Lillooet and Shuswap are relatively well known, mainly through the observations of James Teit (1900, 1906, 1909a), who gathered information for the Jesup North Pacific Expedition under the general direction of Franz Boas. Dawson (1891) also 49 Figure 3. Physiographic zones of Bri t i s h Columbia, showing the area of study. After Holland (1964) 5 0 Figure 4. Ethnographic groups of B r i t i s h Columbia, the major bands of interest. After Duff showing (1964) . 51 contributed information on the Shuswap, obtained while under- taking reconnaissance for the Geological Survey of Canada. Teit's work is especially valuable, for i t often contains comparisons of material culture, beliefs, shelter and food acquisition. The Chilcotin are reasonably well described by Lane (1953, 1981), but his research was conducted quite late in time (1951). Teit's (1909b) Chilcotin writings contain only minimal reference to subsistence and settlement, being mostly concerned with basket- ry and motifs, and Farrand's (1898, 1900) accounts of myths and legends provide l i t t l e substantive data. Ray (1942) interviewed only one Chilcotin in his Plateau culture traits study, and men- tions that he considers his Chilcotin data to be the least reliable of his sample. The Reverend A.G. Morice compiled detailed accounts of the Carrier and Chilcotin during his missionary work; however, his references to Chilcotin are few and often offered in compar- ison to Carrier (Morice 1893, 1906). Morice's writing also has a strong antiquarian and ethnocentric tone, at times leading one to suspect the accuracy of his statements. A recent study of Chil- cotin ethnohistory by Tyhurst (n.d.) presents an in-depth exam- ination of economic circumstances that have led to current eastern Chilcotin culture. A l l the classic authors take pains to point out that the cul- tures described were observed after large-scale decimations in population had occurred, mainly due to smallpox epidemics of the 52 1860's (see Duff 1964). Yet i t i s notable that s i g n i f i c a n t aspects of the a b o r i g i n a l c u l t u r e s remain today, w i t h respect to the use of p l a n t , f i s h and mammal resources, and c e r t a i n o r a l t r a d i t i o n s (Bouchard and Kennedy 1979; Kennedy and Bouchard 1978; Turner, Kennedy and Bouchard 1980; Turner 1977, 1978, 1979). 3.1.1. C h i l c o t i n The C h i l c o t i n occupied the western edge of the I n t e r i o r P lateau and were western neighbours of the Canyon Shuswap and northern neighbours to the Upper L i l l o o e t . The C h i l c o t i n had access to salmon', both sockeye and kokanee, but Lane (1953: 42) observed that t r o u t , w h i t e f i s h and suckers were o v e r a l l of more importance than were e i t h e r the annual r i v e r run or land-locked salmon. T e i t (1909b: 779) wrote that the m a j o r i t y of the salmon used by the C h i l c o t i n were obtained through trade w i t h the B e l l a Coola and Shuswap. Jorgensen (1980) observed that i n a sample of 172 western.'Indian T r i b e s "...only the C h i l c o t i n aquired more than 10 percent of t h e i r t o t a l d i e t from f i s h gained from t h e i r neighbours " (1980: 125). According to Lane (1953: 172 - 173) the months of J u l y to September were the period of greatest aggregation f o r the C h i l - c o t i n . While engaged i n root gathering i n the mountains and salmon f i s h i n g at favored l o c a t i o n s along the C h i l k o and C h i l c o t i n r i v e r s , s e v e r a l f a m i l i e s would camp together and cooperate i n 53 food aquisition, processing and storage. In October, and part of November, individual families would hunt game, and from November through to February, encampments of one or two families would winter together. Individual families would disperse to fishing sites from March to April and from then into July es- pecially productive fishing and berrying locales would be fre- quented by "semi-bands" comprising several families. House structures of the Chilcotin.are reported originally to have been gabled plank houses, with rectangular and oval outlines. Round semisubterranean pithouses were later copied from the Shuswap (Lane 1953: 146; 1981: 403). Lane's informants claimed that their ancestors built their houses in isolation near lakes, and denied that housepits found near fivers were Chilcotin in origin. Lane photographed an abandoned but stand- ing "bark house" near Puntzi Lake in 1951 (Lane 1981: 403). In the summer, brush shelters were erected, but Lane notes that ". . . i n both summer and the winter, people often camped in the open with no shelter " (1953: 46). To store salmon and other foods for winter months, caches were constructed that consisted of low log structures that Morice (1893: 179) says were placed on the ground. Morice also noted that these were constructed at some distance from regular villages, but Lane (1953: 46) claims caches were put up at planned future campsites. In neither case is i t mentioned whether or not caches were located near winter camps. Proximity of storage f a c i l i t i e s 54 to long term camps i s not a b s o l u t e l y necessary. For example, Honigmann (1954) observed that Kaska of northern B r i t i s h Columbia would t r a v e l as f a r as 35 miles to r e t r i e v e cached food during winter shortages. The area of Eagle or Choelquoit Lake i s not mentioned s p e c i f i c a l l y i n the ethnographic r e c o r d , and the word "Choelquoit" i s not known to have any meaning i n C h i l c o t i n or C a r r i e r (Tyhurst 1982, personal communication). Lane (1953) p l o t s a C h i l c o t i n housepit s i t e on the south edge of the l a k e , but the s c a l e i s so innaccurate as to be of no use i n d i f f e r - e n t i a t i n g that one s i t e from s e v e r a l others c u r r e n t l y known near the l a k e (Matson et a l 1980; Germann 1979). 3.1.2. Thompson Upper Hat Creek V a l l e y was l a r g e l y a p a r t of the t e r r i t o r y claimed by the Spences Bridge d i v i s i o n of the Upper Thompson Indians ( T e i t 1900: 170), but the extreme northern part of the v a l l e y and i t s lower reaches to the Bonaparte River were w i t h i n Bonaparte Shuswap t e r r i t o r y ( T e i t 1909a: 456). The v a l l e y i s s p e c i f i c a l l y mentioned by T e i t (1900: 170) as being an area near the western l i m i t of the Spences Bridge band. T e i t c l e a r l y regarded the Upper Thompson and the Bonaparte Shuswap as h i g h l y s i m i l a r i n manufactures, subsistence p r a c t i c e s and s o c i a l organ- i z a t i o n , and i n h i s d e s c r i p t i o n of the Shuswap (1909a), makes c o n t i n u a l reference to h i s volume on the Thompson (1900). Strong s i m i l a r i t i e s were a l s o perceived by Jorgensen (1969), whose ieompre- 55 hensive q u a n t i t a t i v e study of S a l i s h c u l t u r e grouped both i n t o a "Thompson Cult u r e C l u s t e r " , w i t h i n which 70% of t e c h n o l o g i c a l , s o c i a l o r g a n i z a t i o n a l and i d e o l o g i c a l charac- t e r i s t i c s were shared. Contemporary I n t e r i o r S a l i s h i n f o r - mants a l s o consider themselves to be a p a r t of a common c u l - t u r a l p a t t e r n (see Brow 1972). The Upper Thompson and Bonaparte Shuswap both wintered i n s h e l t e r e d major r i v e r v a l l e y s , w i t h most f a m i l i e s occupying pithouses, although mat-covered lodges p a r t l y banked w i t h e a r t h were a l s o constructed ( T e i t 1909a: 493; Boas 1890: 634). T e i t saw only a few pithouses s t i l l i n use by the time he undertook h i s s t u d i e s , but he was able to gather much v a l u a b l e i n f o r m a t i o n . For example, f o r the Thompson, T e i t observed: The e x i s t e n c e of numerous r u i n s of underground houses might be considered as s u f f i c i e n t proof of the decrease of the t r i b e , were i t not that the same f a m i l y sometimes constructed s e v e r a l of these houses...(1900: 175). Working together, a group of 20 or 30 people could c o n s t r u c t a pithouse i n a s i n g l e day ( T e i t 1900: 192), and the d w e l l i n g s were u s u a l l y i n h a b i t e d from December u n t i l February or March, which f o r the Thompson and Shuswap was the period of greatest p o p u l a t i o n aggregation ( T e i t 1900: 194, 238). People r e l i e d h e a v i l y on foods that had been stored from the summer salmon runs and root and berry crops, but on occasion or i n periods of duress, they 56 would hunt l a r g e game and trap smaller mammals. Many kinds of snares, d e a d f a l l s and traps were used f o r both kinds of game, i n c l u d i n g deer fences and p i t t r a p s . By A p r i l , the pithouse v i l l a g e groups had dispersed to lake and stream f i s h i n g l o c a t i o n s and were engaged i n the gath- e r i n g of r o o t s , new shoots and cambium. The composition of such task groups i s not d e t a i l e d , but i n a l l l i k e l i h o o d s i n g l e f a m i l i e s set out at f i r s t , and r e s o u r c e - r i c h areas of the summer were the scenes of band-level aggregation on the order of 20 to 30 people among two to four f a m i l i e s . Hunting and trapping were c a r r i e d out by men, w h i l e women undertook the c o l l e c t i o n and processing of pla n t foods (Dawson 1891: 19; T e i t 1900: 230). Root resources were e s p e c i a l l y important during the e a r l y summer. These were dug from the rocky s o i l s they favour w i t h the help of dig g i n g s t i c k s and processed f o r immediate consumption and f o r storage. Root baking or steaming was accomplished by the c o n s t r u c t i o n of earth ovens. These were b u i l t by both sexes, and were a l s o used to cook mammals- (Dawson 1891:9; Ray 1942). At summer camps where an extended stay was planned, temporary s h e l t e r s of mats, bark and s k i n s were constructed ( T e i t 1900: 195 - 197, 1909a: 493). In l a t e summer, about August, people congregated along the major r i v e r s i n a n t i c i p a t i o n of the annual salmon runs. Large camps were set up on the banks of the Thompson and Fraser R i v e r s , favoured l o c a l e s being n a t u r a l narrowings i n the waterways. 57 The ascending salmon were caught with spears, nets and weirs, and dried by air and smoke to be preserved for the winter months. Salmon were also traded among Indian groups, as were o i l and other by-products, as well as dried roots. The dried salmon were stored in underground pits that were lined with bark and were usually located close to winter habitation sites (Teit 1900: 198 - 199). The remainder of the year prior to the winter's accumulation of snow, was spent hunting, trapping and gathering late season foods such as white-bark pine and ponderosa pine nutlets (Dawson 1891: 22). 3.1.3. Shuswap Concerning the general pattern of Shuswap subsistence and settlement, Teit wrote: The Shuswap may be classed as a hunting and fishing tribe; the former occupation, on the whole, predominating. The Fraser River and Canon bands were the most sedentary, the latter being almost entirely so; while the North Thompson bands were the most nomadic (1909a: 513). The Mouth of the Chilcotin region assemblages that are analysed in Chapter 6 were recovered from the Canyon Shuswap territory at the confluence of the Fraser and Chilcotin rivers. Teit clearly considered them to partake of a l i f e s t y l e somewhat different from other Shuswap and neighbouring Chilcotin: 58 They controlled part of the Chilcotin salmon supply, and the Chilcotin traded extensively with them...they...did very l i t t l e travelling or hunting (1909a: 535). Yet i t was apparent that overall, Shuswap band composition was f l u i d , in part because of a very mobile pattern of settlement: ...the small wintering places were frequently changed, and even the main locality of village of a band would have more families one winter and less another. Some families were more nomadic than others, and each band would have people from neighbouring villages li v i n g with them every winter (Teit 1909a: 457). Teit (1909a: 457) was of the opinion that 50 years prior to'Uhis time (i.e. about 1850) there were more, and smaller villages in existence. Before the smallpox epidemics of 1860 - 1863, the Canyon division was estimated to number about 700 people in four bands (.100 of these in the band at the Mouth of the Chilcotin), and the Bonaparte division was estimated at 700. people in three bands (Teit 1909a: 464 - 465) . As for structures, Teit (1909a: 493 - 495) notes that the following were in use among the Shuswap: conical mat lodges and semi-subterranean lodges for winter dwellings, long double lodges for several families at fishing resorts, trapping lodges built near deer fences, menstrual huts for young women, and sweat houses. It is explicit throughout the Shuswap descriptions 59 that the Thompson used much the same kinds of s h e l t e r s . F i s h i n g i n lakes and streams was g e n e r a l l y of greater importance to a l l Shuswap than to Thompson ( T e i t 1909a: 513), and gathering may have been. T e i t (1909a: 513 - 514) l i s t s 15 mammals, 18 v a r i e t i e s of r o o t s , 18 kinds of b e r r i e s , as w e l l as mosses, l i c h e n s , c a c t i , nuts and the cambium of 8 t r e e species that were r e g u l a r l y used by the Shuswap. I t i s l i k e l y that more p l a n t s than those enumerated were used r e g u l a r l y . D e t a i l e d d e s c r i p t i o n s of f l o r a l resource a q u i s i t i o n and pro- cessing are provided by Turner (1977). 3.1.4. L i l l o o e t The Upper or Fraser R i v e r band of the L i l l o o e t t r i b e occ- upied the east and west s i d e s of the Fraser River from Seton Lake and the present town of L i l l o o e t n o r t h to . P a v i l i o n Creek and the Fraser R i v e r ( T e i t 1906). T e i t (1906: 223) notes that the Upper L i l l o o e t made two kinds of food c e l l a r s . One kind was very care- f u l l y b u i l t , and was employed to s t o r e food u n t i l s p r i n g ; the other kind was used f o r the w i n t e r ' s food supply only, and was l e s s c a r e f u l l y b u i l t , near the w i n t e r house. O v e r a l l , L i l l o o e t c u l t u r e was much l i k e that of the Thompson and Shuswap, e s p e c i a l l y the Upper L i l l o o e t , s i n c e the Lower L i l l o o e t i n t e r a c t e d c o n s i d e r - ably w i t h the Coast S a l i s h groups ( T e i t 1906). The L i l l o o e t were known to hunt caribou i n the extreme northwest of t h e i r hunting grounds, along w i t h mule deer, mountain goat, mountain sheep, hoary marmot and b l a c k bear ( T e i t 1906: 223), a p r a c t i c e uncommon among 60 Shuswap or Thompson. It should be noted here that Hill-Tout's (1905) description of the Lillooet does not offer much detail about subsistence or settlement practices of the Fraser River band. Boas (1906) con- sidered Hill-Tout's account to contain inaccuracies in content, relative to Teit's (1906) record of Lillooet culture. Kennedy and Bouchard (1978) have added to the accounts of pithouses, by interviewing contemporary Lillooet. While most information agrees with Teit's description of the Thompson, i t was noted that abandoned pithouses were at times used as work- shops for the manufacture of implements (Kennedy and Bouchard 1978: 37), and also that at "potlatches" deer or horses were tossed into pithouses, to be butchered by guests. Elderly people are reported to have resided in pithouses during the summer months. To keep snakes from frequenting the houses, ants' nests were placed about them (Kennedy and Bouchard 1978: 37). Apparently ants secrete a substance that repels snakes. 3.1.5. Cross-Cultural Discussion From a cross-cultural perspective, i t is apparent that the four groups under consideration were much alike in technological and economic adaptations. A comparison of the general round of seasonal activities conducted during the "moons" or months of the year for each of the groups, as e l i c i t e d by Teit (1900, 1906, 1909a, 1909b) and Morice (1893) , is a convenient manner to demonstrate their subsistence and settlement patterns (see Table 1). A l l four 61 CHILCOTIN SHUSWAP THOMPSON LILLOOET - (Morice 1893) ( T e i t 1909a) ( T e i t 1900) ( T e i t 1906) JANUARY sun turns deer bucks shed a n t l e r s does le a n c o l d e s t weather FEBRUARY chinook winds s p r i n g winds some people leave houses people come out of houses MARCH come out of subterranean huts leave p i t - houses, d i g roots a l l people come out of houses some f i s h i n g and hunting APRIL suckers f i s h e d snow gone from high ground, people d i g r o o t s f i s h t r o u t w i t h dip nets trap lake f i s h MAY people f i s h t r o u t at lakes root d i g g i n g f i r s t salmon small f i s h JJUNE s e r v i c e b e r r i e s r i p e n young deer born, b e r r i e s r i p e n b e r r i e s r i p e n JULY kokanee f i s h e d salmon a r r i v e b e r r i e s r i p e n some people hunt berry p i c k i n g AUGUST salmon f i s h a l l month sockeye run salmon run SEPTEMBER cache f i s h hunt cohos come b o i l salmon, make o i l OCTOBER - hunt and trap i n mountains t r a p , hunt hunt and trap NOVEMBER enter sub-terranean huts going i n time, deer r u t deer r u t going i n time DECEMBER i c e f i r s t r e a l c o l d i n t o winter houses sun turns TABLE 1. S e a s o n a l i t y of I n t e r i o r Plateau groups as evidenced by general a c t i v i t i e s undertaken during "moons". Monthly equivalents pro- vided i n the ethnographies c i t e d . 62 "entered pithouses about the month of November, the Thompson p o s s i b l y w a i t i n g u n t i l a f t e r the deer r u t was over. Root d i g - ging was a p r i o r i t y a c t i v i t y during the months of March to May, w i t h the Thompson p o s s i b l y spending more time at t h i s , perhaps because of t h e i r access to good root grounds such as Hat Creek. P r i o r to the a r r i v a l of the salmon i n August, they spent June and J u l y undertaking a wide range of f o r a g i n g a c t i v i t i e s , but focusing on b e r r i e s , p a r t i c u l a r l y s e r v i c e berry. August and September was the time to c a t c h , process and s t o r e salmon, and the month or two p r i o r to the commencement of winter l i f e was the time to hunt l a r g e mammals w h i l e they were r u t t i n g and descending to lower e l e v a t i o n s . Jorgensen's (1980) m u l t i v a r i a t e study of 172 western I n d i a n t r i b e s includes the most recent and perhaps the most o b j e c t i v e comparison of the C h i l c o t i n , Shuswap, Upper L i l l o o e t and Upper Thompson. In r e l a t i o n to the broad range of environments occupied by the Indians of western North America, Jorgensen shows that these four groups had s i m i l a r resource types and c l i m a t i c c o n d i t i o n s . They had h i g h l y s i m i l a r technologies, u.as w e l l as r e l a t i v e l y strong resemblances i n economic and s o c i a l o r g a n i z a t i o n . The subsistence economy of the C h i l c o t i n and Shuswap was somewhat d i f f e r e n t from that of the Upper L i l l o o e t and Upper Thompson, and i t i s notable that the L i l l o o e t and Thompson f a l l i n t o two completely separate c l u s t e r s i n the subsistence economy a n a l y s i s . The Shuswap, Upper L i l l o o e t and Upper Thompson are s i m i l a r to each other w i t h respect 63 to ceremonialism and spiritualism, but the Chilcotin are different from them in both these aspects. Yet the Shuswap and Chilcotin are alike in p o l i t i c a l organization, and the Upper Lillooet and Upper Thompson diverge from these two groups as well as from each other in this analysis. For the purposes of this study, these are appealing results of an exhaustive research programme, but Jorgensen's style of the quantitative study renders conclusions d i f f i c u l t . Overall, the four Plateau cultures under consideration show a communality in culture that seems to be more attributable to environment than to language or ideology, and this is not surprising for hunting and gathering societies. The four are loosely grouped in Jorgensen's C1980) analysis of economic and social organization, but this is perhaps where the Interior Plateau ethnographies are weak. The Chilcotin here are somewhat more loosely linked to the three Salish tribes. In a l l fairness, i t must be recognized that the attributes coded for the groups in Jorgensen's analyses had to be "averaged out" for several bands and in the face of sometimes conflicting evidence, yet some problems do exist with the data codings. For example, Jorgensen (.1980: 356) classes the Chilcotin as employing double lean-tos as winter habitations, the Shuswap, Upper Lillooet and Upper Thompson as employing pithouses. This is perhaps acceptable as far as the prehistoric use of such structures is concerned, but i t is clear from Morice and Teit as discussed above 64 that the Chilcotin did live in pithouses, and eventhough these were "borrowed" from the Shuswap, the Chilcotin clearly had their own patterns of using them, preferring isolated rather than grouped villages, and lake locations rather than rivers. This problem appears to stem from Jorgensen's reliance on Ray's (1942) Chilcotin evidence rather than that of Morice (1893). Perhaps also misleading is the classification of Chilcotin and Upper Lillooet as lacking conical and subconical dwellings, both att- ributed to the Shuswap and Upper Thompson, with 4-pole foundations. Furthermore the Shuswap are classed as obtaining aquatic animals only as a tertiary contribution to diet, and these are coded as secondary contributors to Chilcotin and Upper Lillooet diets, and as the dominant food source among the Upper Thompson. These cod- ifications are perhaps valid for each "tribe" as a whole, but are not the case for each band, particularly the Canyon Shuswap. In general, Jorgensen's (1980) analyses provide a panorama depict- ing the groups of interest, and the u t i l i t y of the volume is only slightly hampered by the few inconsistencies with known occurrences of specific material culture. It is clear that the four groups examined are closely related in most aspects of material culture, technology and economy. 3.2. Ethnographic References to Lithic Technology This dissertation is concerned with prehistoric stone tools, and i t is appropriate to review what has been recorded about l i t h i c technology by both the classic and the more recent ethnographers. 65 No previous c o m p i l a t i o n of the I n t e r i o r Plateau l i t h i c t e c h - nology references e x i s t s , and I b e l i e v e i t i s u s e f u l here as a source of f u t u r e r e f e r e n c e , to lend i n s i g h t to more complex patterns to be discussed, and f o r documentation i n i t s own r i g h t of t h i s p r a c t i c a l l y e x t i n c t set of tool-making techniques. The d e s c r i p t i o n s of the l a t e 19th. century are o f t e n more i n f o r - mative and l u c i d than some of the c l a s s i c cases i n the current a r c h a e o l o g i c a l l i t e r a t u r e such as the Western Desert Aborigines (Gould et a l 1971; Gould 1980, Hayden 1978). Most of the informa t i o n can be taken as accurate, but some i s thought to be m a r g i n a l l y so. The f o l l o w i n g excerpt from a r e c e n t l y c o l l e c t e d Shuswap i n t e r v i e w about hunting techniques (.Willard 1979) i s reasonably w e l l informed, yet a l s o u n r e a l i s t i c . Before the non-Indians came to t h i s country, the Shuswap people used to go to Ta-Ta-CAIL-in, a mountain near Kamloops, to c o l l e c t f l i n t r ocks. The rocks that were rounded on one end were used when they tanned h i d e s , and the t h i n rocks were used fo r arrowheads. To make the arrowheads t h i n and sharp, the rocks were placed i n the f i r e u n t i l they were red hot and then they were dipped i n t o some co l d water. The chips that broke o f f when the rock h i t the water were very sharp and good f o r arrow- heads . The piece of f l i n t was then fastened to a j u n i p e r s t i c k which had been s p l i t and w h i t t l e d . The f i n i s h e d arrow i s about three f e e t long (.Willard 1979: 139; emphasis added). H i l l - T o u t ' s informants i n the Lytton, and L i l l o o e t area appear not to have been f a m i l i a r w i t h stone t o o l s as subsistence implements, but claimed that they were used i n personal s c a r i f i c a t i o n ( H i l l — 66 Tout 1905: 64). Teit provides descriptions of quarries, observations on functional specificity of tool types, and a good description of bipolar core reduction: Arrowheads were made of glassy basalt which was obtained at a certain place north of Thompson River... Many were made out of large chipped heads, which are found in great numbers in the valleys (1900: 241). ...spearheads were similar in shape and material to the arrowheads except that they were larger (1900: 236). The Indians are s t i l l familiar with the art of making arrowheads. When these were to be made from a boulder, the following method was employ- ed. The boulder was s p l i t by being laid on a stone and struck with a hand-hammer, generally a pebble of handy size. When a suitable piece had been obtained, i t s edges were trimmed off with a hard stone. Then i t was wrapped in grass or hay, placed on edge on a stone, and large flakes were s p l i t off with a hand-hammer. After a suitable piece had been obtained, i t was placed on a pad in the l e f t hand and held in position with the fingers. It was given i t s f i n a l shape by means of a flaker made of antler...which was used with a forward and downward pressure (1900: 182). Teit was aware of the concept of stages in stone tool manufacture, and compared what he witnessed among the Thompson to stone work- ing described by Mor ice (.1893): The blunt point served for flaking off larger chips, while the smaller one was used for the fi n a l stages of the work. In later times, iron flakers were used. The method of holding the flake was the same as that of the Carrier Indians of northern British Columbia (Teit 1900: 182). 67 Morice's (1893: 51) d e s c r i p t i o n of b i p o l a r r e d u c t i o n , which he maintained was used "almost i n v a r i a b l y " , i s compar- able to T e i t ' s but l a c k i n g i n d e t a i l . There i s a l s o evidence that h a f t s were not the most valued part of a l l composite stone t o o l s , c o ntrary to Keeley (1982), who argues that the e f f o r t s of r e h a f t i n g blunt t o o l s are l a r g e enough to warrant extensive resharpening p r i o r to d i s c a r d : This h a f t i n g i s temporary as the stone part only of the implement i s u s u a l l y kept among the f a m i l y c h a t t e l s (Morice 1893: 51). Morice i s here d e s c r i b i n g cobble s p a l l hide s c r a p e r s , f o r which contemporary c u r a t i o n of t h i s s o r t has been r e c e n t l y recorded. A l b r i g h t (1982) has observed a Tahltan woman i n Telegraph Creek using stone hide scrapers and searching f o r s u i t a b l e stone w h i l e on t r a p l i n e s . The woman keeps her s p a l l t o o l s and has her mo^ ther's as w e l l , numbering some f i v e to ten i n a l l . I should note here that a C h i l c o t i n woman of the Nemiah band at C h i l k o Lake i s . reported to make and use s p a l l hide scrapers a l s o (D. Lulua , personal communication 1979; see Matson et a l 1980: 230). There are references to t r a d i t i o n a l names f o r stone raw mat- e r i a l s such as / p i s / , which i s a "black resonant rock" i d e n t i f i e d by Dawson as a u g i t e - p o r p h y r i t e (Morice 1893: 53). Morice mentions that the C a r r i e r had s i x words f o r s u i t a b l e chipping stone, i n c l u d - ing / n a l r e / f o r o b s i d i a n (1.893: 53), which i s a l s o known as /bez/ 68 by Anahim Lake Chilcotin (Wilmeth 1978), and Nemiah Chilcotin (personal observation) and / t s e - l k r a i / for chalcedony. There is no real contradiction in the use of /p i s / for dark basaltic rock and /bez/ for obsidian, since the Athapaskan word applies to black rock in general, viz the Baezeko RiXjier, Beece Creek, places where quantities of glassy basalt and other volcanics can be found (Ty- hurst, personal communication 1982); i . e . these two words are cognates. Teit (1909a: 473) noted that Shuswap and Thompson stone work- ing techniques were ident ica l , and that while rough spal l scrapers were usually employed to scrape hides, occasionally fine basalt was used. This is evidenced archaeolbgically at the Mouth of the Chi lcot in , where a very heavily worn scraper is a fine basalt b i - face, the broad blunt end being the locus of considerable round- ing (Matson, Ham and Bunyan 1979; Ham 1975: 160). Morice, how- ever maintained that such scrapers "receive no polish whatsoever" (1893: 50). F ina l ly , Morice presents evidence that there was some owner- ship attached to specific quarries: The material chosen in preference to fashion arrow or spear heads with was loose, broken pieces of rock such as were found on the sur- face. Of course, these were confined to a few loca l i t i es only wherein were situated sorts of quarries which were very jealously guarded ag- ainst any person, even of the same tr ibe , whose right to a share in their contents was not fu l ly established. A violat ion of this tradit ional law was often considered a casus b e l l i between the co-clansmen of the trespassers and those of the proprieters of the quarry (1893: 65). 69 This d i s c u s s i o n has not attempted to compile a l l the known references to the c r a f t of stone working as p r a c t i c e d by the ethnographic i n h a b i t a n t s of the I n t e r i o r P l a t e a u . For example, i t o f f e r s no d e s c r i p t i o n - o f stone g r i n d i n g techniques, which to judge by the w r i t i n g s of T e i t and Morice, one gets the impression were more i n common p r a c t i c e than was stone f l a k i n g . The e v i - dence i s only s l i g h t l y a n a l y t i c a l l y r e l e v a n t to the remainder of t h i s study, but the d e t a i l s of q u a r r i e s , core r e d u c t i o n , f l a k i n g and pressure retouch are provided to i l l u s t r a t e the nature of the a v a i l a b l e i n f o r m a t i o n . 3.3. Regional P r e h i s t o r i c Archaeology This s e c t i o n reviews the development of p r e h i s t o r i c research i n areas of immediate relevance to the present study. The d i s c u s s i o n focuses on s t u d i e s undertaken on the Fraser and Nechako Plateaux, and excludes research reported from the Okanagan and . Kootenay r e g i o n s , as w e l l as work done i n the Rocky Mountain and northern I n t e r i o r areas. The f o l l o w i n g des- c r i p t i o n of the growth of p r o f e s s i o n a l research i s s t r u c t u r e d i n terms of e a r l y s t u d i e s , c u l t u r e h i s t o r y i n v e s t i g a t i o n s , and settlement p a t t e r n research. 3.3.1. E a r l y Studies The f i r s t observations on p r e h i s t o r i c settlement on the I n t e r i o r P l a t e a u were recorded by George Dawson i n 1877 as part of a r e p o r t on Shuswap ethnography (Dawson 1891). Harlan I . Smith 70 conducted excavations of several burial sites in the southern Interior near Lytton, Spences Bridge and Kamloops, and he also undertook a limited survey of the Nicola Valley (Smith 1899, 1900). Smith interrelated the burial remains he un- covered with local Indian legends, and recognized the contin- uity of the prehistoric remains with the culture of the Thomp- son Indians as described in James Teit's ethnographic research (Teit 1900). Farther to the north on the central Plateau, the Rev. A.G. Morice (1893) described selected aspects of Carrier prehistory, including stone tools and cultural depressions. Morice disputed any claims for significant antiquity of archaeological materials, citing as evidence the similarity of abandoned sites and a r t i - facts to those in use by the Athapaskans whom he was converting to Christianity (Morice 1893: 39 - 43). 3.3.2. Culture History and Classification Studies Central Plateau No archaeological research was conducted in the central and southern Interior of the province un t i l Borden's surveys of Tweedsmuir Park and the Nechako River system in the early 1950's (Borden 1952a,b). Borden's work was carried out to partially off- set environmental impacts caused by the construction of the Kenney Dam by the Aluminum Company of Canada, and can be seen as being inspired by the extensive river basin surveys and salvage archae- ology projects that were being carried out by American archaeologists 71 of the time. Borden's surveys on the central Plateau also prompted him to devise a uniform site recording scheme (Borden 1952c), now known as the Borden system. The most significant results of Borden's research came from the excavations at Chinlac village (GaRv 1) and Natalkuz Lake (FiSi 19). Chinlac was recognized as a site occupied during the protohistoric and historic periods, and Natalkuz Lake revealed two periods of occupation. The lower levels of FiSi 19 con- tained micro- and macroblades, and were dated to 2415 + 160 BP. Borden classified the lower part of this site, actually a large hearth feature, as the remains of a non-Carrier or Chilcotin "Natalkuz Lake Culture", and considered the uppermost remains to represent a late prehistoric Carrier occupation (Borden 1952b). Chinlac was the site of a historically recorded battle, ca. 1745 (see Wilmeth 1978: 6) between Carrier and Chilcotin (Morice 1906: 14 - 15). The village may have been visited by Simon Fraser sometime around 1806 or 1807 (Lamb 1960; Nechako Valley Historical Society 1979). Wilson Duff's fieldnotes on Carrier Indians (Duff 1951) contain interviews with informants who claim that the site was not reoccupied following the massacre. Duff's (1951) Carrier informants claim that the site was a summer fishing site located near a large weir on the Stewart River. Chinlac presently consists of ten shallow, large, rectang- ular depressions in a clearing about an acre in size, with well over 100 cachepit depressions located in the forest west of the 72 c l e a r i n g . In h i s excavation of one of the l a r g e r depressions (House I I I ) at C h i n l a c , Borden recovered items of i r o n and cop- per, g l a s s beads, bark r o l l s , f a u nal remains and items of stone and bone manufacture. The m a t e r i a l s from t h i s s i t e have never been f u l l y described or analyzed, although a 25 percent sample of the debitage from f i e l d bags was examined i n a debitage study of s e v e r a l Plateau assemblages by the present author (Magne 1980). A l s o , 14 of the hundred-odd p r o j e c t i l e p o i n t s from Chinlac were used as "known Athapaskan" items i n a study of ethnic homogeneity i n small side-notched po i n t s t y l e s (Magne and Matson 1982). The e n t i r e assemblage i s c u r r e n t l y undergoing study by Cranny (1982) , who t h i n k s that the s i t e i s multi-component. I t seems r a t h e r c l e a r that the House I I I depression excavated by Borden was b u i l t and used by a s i n g l e C a r r i e r band, although hearth fe a t u r e s appear to have been used repeatedly, p o s s i b l y w i t h seasonal lapse s . As concerns the 130 or so s i t e s that he l o c a t e d i n h i s 1951 survey of some 400 miles of r i v e r and lake shores, Borden notes: Most of the s i t e s . . . are hunting, f i s h i n g , b e r r y - p i c k i n g and cambium-gathering camps w i t h - out i n d i c a t i o n s of permanent h a b i t a t i o n . S i t e s are o f t e n l o c a t e d at the head of o u t l e t of l a k e s , near marshes or game c r o s s i n g s , i n s h e l t e r e d bays or coves w i t h sandy beaches, and near head lands a f f o r d i n g a sweeping view of the l a k e . Most s i t e s are found on the n o r t h s i d e of the l a k e s , i n d i c a t - ing that a southern exposure was a desirable f a c - t o r CBorden 1952b: 34). 73 The Punchaw Lake site (FiRs 1), located 55 km southwest of Prince George, consists of A3 house platforms, 57 storage pits, and a historic t r a i l segment. Two house platforms were excavated at this site (Fladmark 1976; Montgomery 1978). Area A, reported by Fladmark (1976) contained a burial, below which deposits were dated to 3980 + 100 BP, and "the last major occ- upation" of the site i s thought to have taken place between AD 1700 and AD 1800 (Fladmark 1976: 31). Montgomery's (1978) ana- lysis of the stone tool assemblage from Area C demonstrated that a l l stages of tool manufacture were present within the deposits. At the Tezli site (FkSd 1), several of A6 visible cultural depression features were test excavated by Donahue (1977). Donahue posited that the site was f i r s t occupied about 2500 BC by people using pithouses as winter habitations. The artifacts from Tezli were classified into many morphological types and compared visually with other collections from western Canada and the U.S. The results led Donahue to assert that no major popu- lation displacements have occurred on the Interior Plateau within the last A500 years, and that continuous cultural evolution has occurred throughout the region (Donahue 1977) . It is clear that Donahue also recognized certain "influences" and may have glossed over "diagnostic" artifacts in the Tezli assemblage. In particular, IA microblades were found at Tezli, 12 of these being from the same stratigraphic layer that yielded a 3850 +160 BP date, but "for a l l intents and purposes" (Donahue 1977: 259) a microlithic technology 74 is not present. Since Donahue discontinued screening of the site matrix early in the excavation schedule (1977: 119), a bias towards large artifact recovery is not surprising. A significant collection of microblades may yet exist at the site. Donahue's investigations of Carrier prehistory through the direct historic approach (Steward 1973) were initiated at Ulkatcho, an early historic trading centre for Carrier. Ulkatcho was visited by Mackenzie in 1793 and by Dawson in 1876 (Donahue 1973) , and the people of this village were the subjects of Goldman's (1940) ethno- graphic research. Wilmeth (1969, 1970, 1971, 1975, 1977, 1978) has investigated several sites in the Anahim Lake area. He has attempted to date the arrival of Athapaskan Chilcotin in the region, and to compile definitive traits of prehistoric Chilcotin material culture also via the direct historical approach. Using evidence obtained from five house remains at the Potlatch site (FcSi 2), two houses from the Goose Point site (FdSi 3), and another from the Daniktco site (FdSi 3), Wilmeth's (1978) current interpretation is that five principal phases, or "component clusters" of human occupation are evident in the area. The earliest of these spans a period of AD 1 to AD 400, and is characterized by microblades, and the second, dating from AD 700 to AD 850, also contains microblades but is distinct by virtue of an apparent temporal hiatus. The third phase dates around AD 1200 to AD 1800. The White River Ash f a l l in the Yukon that is estimated to have occurred at about AD 700 is said to be the major factor precipitating Chilcotin mi- 75 g r a t i o n to the area (Wilmeth 1978: 173). Elsewhere on the c e n t r a l P l a t e a u , M i t c h e l l (.1969, 1970) excavated three s i t e s , a s s i g n i n g each to a d i f f e r e n t phase of the Nesikep T r a d i t i o n that i s discussed below. S t r i c t l y on the b a s i s of r a t h e r questionable t y p o l o g i c a l comparisons, M i t c h e l l (1969) placed the Poplar Grove s i t e (FaRx 1) i n the Lower Middle period (5000 to 3500 BP), the Horn Lake Southwest s i t e (EkSc 1) i n the Upper Middle period (3500 to 2000 BP), and the N a t s a d a l i a Crossing s i t e (FdSi 2) i n the Late Nesikep period (2000 BP to h i s t o r i c ) . Wilmeth (1978) considers FdSi 2 to be a C h i l c o t i n occupation. Prompted by h i s research at T e z l i and Ulkatcho, Donahue (1975) examined c o l l e c t i o n s of surface c o l l e c t e d items from 40 l o c a t i o n s that had been donated to the N a t i o n a l Museum of Man i n Ottawa. With l a c k of good provenience, the r e s u l t i n g catalogue cannot serve as a base f o r f i r m c o n c l u s i o n s , but the d e s c r i p t i o n of a S c o t t s b l u f f - E d e n p o i n t found near Vanderhoof on the Nechako River suggests a p o t e n t i a l occupation of the area s t a r t i n g as e a r l y as ca. 9000 BP. Wilmeth (.1978: 143) notes that an A l b e r t a point made of o b s i d i a n was found near Anahim Lake, and i n excavations at the P o t l a t c h s i t e he recovered a broken b i f a c e that he types as a Pryor Stemmed p o i n t , d a t i n g to ca. 5610 to 6550 BC i n P l a i n s r e - gions. This l a t t e r f i n d i s questionable given that no radiocarbon dates from the P o t l a t c h s i t e o l d e r than AD 80 were obtained (Wilmeth 1978: 154). The p o s s i b i l i t y of Paleo-Indian peoples i n the c e n t r a l 76 Plateau i s also reinforced by the discovery of Pleistocene mammoth remains at Babine Lake (flarington et al. 1974), a l - though no human a r t i f a c t s are associated. Whitlam (1976) analysed materials that were recovered from three s i t e s excavated as part of a highway salvage pro- gram, near Williams Lake. A l l s i t e s were occupied during the Late Nesikep T r a d i t i o n , and housepit s i t e s FaRn 3 and ElRn 3 each appear to have been occupied twice, at times averaging 1762 + 58 BP and 1180 + 58 BP. Whitlam (1976) applied SYMAP. programs to the d i s t r i b u t i o n of a r t i f a c t s obtained from "mounds" at s i t e FaRm 8 i n an attempt to d i s c e r n occupation, stone work- ing and storage a c t i v i t y areas. In sum, the ce n t r a l Plateau has been the locus of several studies i n culture h i s t o r y and a r t i f a c t typology, but s t i l l lacks a cohesive regional scheme with f i r m horizon markers, except perhaps for the l a s t 1000 years. This, i n e f f e c t , means that p r e h i s t o r i c cultures that are d i r e c t l y and unquestionably ancestral to ethnographically documented cultures are the only "phase" that can be i d e n t i f i e d . The u t i l i t y of temporal horizon markers such as microblades, corner-notched points and small side-notched points i s uncertain for several reasons. Perhaps the most c r i t i c a l , maybe even i n c o r r i g i b l e reason i s a lack of s t r a t - i f i e d , non-housepit s i t e s such as caves, rockshelters, middens, or i n t a c t f l u v i a l / d e l t a i c s i t e s . This problem i s discussed i n more d e t a i l below. ' ~ 77 Southern Plat e a u W i t h i n the scope of t h i s d i s c u s s i o n , contemporary archaeology i n the southern Plat e a u was i n i t i a t e d by Borden. In 1954 and 1956 he excavated a b u r i a l s i t e i n the v i c i n i t y of Cache Creek (Sanger 1968a: 140). David Sanger's i n v o l v e - ment w i t h an I n t e r i o r P l a t e a u p r e h i s t o r y s t a r t e d w i t h a b u r i a l survey i n the L i l l o o e t and L y t t o n areas of the Fraser River (Sanger 1963) and excavation of a b u r i a l s i t e near Chase (Sanger 1968a). Sanger's research, i n the Lochnore-Nesikep l o c a l i t y r e s u l t e d i n the best documented c h r o n o l o g i c a l scheme p r e s e n t l y a v a i l a b l e f o r the e n t i r e I n t e r i o r P l a t e a u (Sanger 1963, 1966, 1969, 1970). This scheme was based on excavations at two deeply s t r a t i f i e d housepit s i t e s , Lochnore Creek (EdRk 7) and Nesikep Creek (EdRk 4 ) , as w e l l as two other s i t e s : Cow Springs (EdRk 5) and Lehman (EdRk 8 ) . Sanger (1970) concluded that two major c u l t u r a l episodes, the Lochnore Complex and the Nesikep T r a d i t i o n , are represented i n the deposits at these s i t e s . The Lochnore Complex (5000 BC - 3000 BC) i s thought by Sanger to represent an i n i t i a l p o p u l a t i o n moving northward i n n e a r l y immediate p o s t - g l a c i a l times. Sanger proposed that the Lochnore Complex was derived from the Old C o r d i l l e r a n Culture as described i n the U.S. northwest by B u t l e r (1961). Borden (1969, 1979) r e f e r s to t h i s complex as the Protowestern T r a d i t i o n . The predominant t r a i t s of the Lochnore Complex are leaf-shaped b i f a c e s and cobble t o o l s that are at times found w i t h other components, but Lochnore 78 Complex assemblages are d i s t i n c t i n that other, more rec e n t , complex t o o l forms are l a c k i n g . Such assemblages have been r e c e n t l y reported by E l d r i d g e (1974) and Richards (1978). Near L i l l o o e t , the Terrace s i t e i s dated at 4145 + 205 BP, and contains an assemblage l a c k i n g microblades, and e x h i b i t s l a r g e cobble cores and leaf-shaped p o i n t s (Richards 1978). S i m i l a r evidence i s to be found at the Moulton Creek s i t e ( E l d r i d g e 1974) on the South Thompson R i v e r , where the assemblage was l o c a t e d below Mt. S t . Helen's "Y" tephra, d a t i n g to about 4000 BP. I t should be noted here t h a t E l d r i d g e i n i t i a l l y l o c a t e d the c u r r e n t l y o l d e s t archaeo- l o g i c a l s i t e i n the I n t e r i o r P l a t e a u , the Gore Creek s k e l e t o n ( C y b u l s k i et a l 1981). The p o s t c r a n i a l remains of a young a d u l t male, apparently caught i n a mudslide, were dated at 8250 + 115 BP. The Nesikep T r a d i t i o n i s thought to represent a southward movement of people who employed a microblade technology. In Borden's (1969, 1979) terminology, these people are known as c a r r i e r s of the E a r l y B o r e a l T r a d i t i o n . The E a r l y Nesikep T r a d i t i o n includes as t r a i t s d i s t i n c t p r o j e c t i l e p o i n t s that are t h i n , r e l a t i v e l y l a r g e and f i n e l y pressure f l a k e d . Sanger (1970) considers these p o i n t s to be derived from Piano c u l t u r e s , although exact t y p o l o g i c a l comparisons are not p o s s i b l e . The Lower and Upper periods of the Middle Nesikep T r a d i t i o n e x h i b i t an abundance of corner-notched p o i n t s , some w i t h concave bases or shoulder tangs that resemble s e v e r a l Middle P r e h i s t o r i c or Archaic p o i n t s from P l a i n s r e g i o n s . The Plateau Microblade t r a d i t i o n continues to e x i s t through the 79 Middle p e r i o d , and i s suggested to terminate about 2000 BP. In the Late Nesikep T r a d i t i o n (.2000 BP to AD 1800) , Sanger r e - cognizes c h a r a c t e r i s t i c s of p r o t o h i s t o r i c and ethnographic i n - h a b i t a n t s of the southern I n t e r i o r , such as l a r g e , numerous pithouse v i l l a g e s , s m a ll side-notched (Kamloops) p r o j e c t i l e p o i n t s , and a v i s i b l e bone and a n t l e r i n d u s t r y . In the Kamloops Phase of the Late Nesikep T r a d i t i o n (ca. 1000 BP to 1800), corner- notched p o i n t s are v i r t u a l l y absent and p o i n t s w i t h m u l t i p l e notches on the blades are q u i t e common (Sanger 1970: 122). In t o t a l , the 7000 year long development of the Nesikep T r a d i t i o n i s thought to represent the e v o l u t i o n of S a l i s h - s p e a k i n g c u l t u r e s of the P l a t e a u . Research undertaken by S t r y d (1970, 1971a,b, 1972, 1973a, 1973b, 1973c, 1978, 1980) i n the Fraser River V a l l e y near L i l l o o e t was aimed at c l a r i f y i n g the p r e h i s t o r i c sequence of c u l t u r e s during the Late Nesikep p e r i o d . L i k e Sanger, Stryd focused on excavating housepits and compiling l i s t s of " d i a g n o s t i c " t r a i t s f o r e x c l u s i v e c u l t u r a l phases, supported i n part by radiocarbon dates. S t r y d (1973a) defined three major components i n the housepits of the L i l l - ooet r e g i o n . The N i c o l a Phase (2750 BP to 1750 BP) i s the e a r l i e s t of these, the L i l l o o e t Phase (.1750 BP to 1150 BP) i s intermediate, and the Kamloops Phase (.1150 BP to 200 BP) i s the l a t e s t . According to Stryd (1973a), the N i c o l a Phase i s c h a r a c t e r i z e d by a l a c k of a microblade technology and s m a l l arrow p o i n t s , and contains corner- notched a t l a t l p o i n t s ( l a r g e w i t h wide necks). Ih the L i l l o o e t Phase, 80 the bow and arrow was introduced, leading to an abundance of small projectile points, both corner and side-notched, and a lack of large corner-notched points. The Kamloops Phase con- tains abundant Kamloops projectile points, which are relatively- thin and well-made, and a f a i r number of zoomorphic figures in bone and stone are also present. Stryd later revised this scheme in an unpublished paper (1973b) , by deleting the Nicola and L i l l - ooet Phases, and preferring to place greater emphasis simply on the introduction of the bow and arrow at ca. 2400 to 1800 BP. This thus extended the Kamloops Phase to 1800 BP, although Stryd may presently include the Lillooet Phase (Matson, personal communica- tion 1983). This latter scheme is perhaps the most defensible, partially because Stryd no longer stresses the bone and antler i n - dustry, and since i t is clear that more of such artifacts were pre- sent in later assemblages, perhaps preservation factors were being reflected more than cultural ones. Thus, in the f i n a l analysis, Stryd's research added detail to the Late Nesikep Tradition as de- fined by Sanger in terms of material culture, but did l i t t l e to an- swer questions pertaining to internal site structure, housepit contemporaneity, or social implications of housepit arrangement within complex sites, that were posed prior to the major portion of his research (Stryd 1971b). Stryd's latest assessment of the Lillooet region sequence is that microblade technology occurs as late as 1250 BP (1973c: 8), and that housepit structures have two basic forms: small ones with 81 c o n i c a l r o o f s , and l a r g e r ones with, a d i f f e r e n t , but undetermined o v e r s t r u c t u r e (1973c: 8). C e r t a i n observations by Stryd concern- i n g f i v e of the L i l l o o e t r e g i o n s i t e s examined i n the present study are presented i n the s i t e d e s c r i p t i o n s e c t i o n of Chapter TV. In the Kamloops l o c a l i t y , Wilson (.1980) defined two pre- h i s t o r i c c u l t u r a l phases s t a r t i n g at ca. 2500 BP. The Thompson Phase (2500 to 1400 BP) represents the f i r s t occupation of the l o c a l area. While t h i s phase inc l u d e s the t r a i t s of Stryd's (1973a) N i c o l a and L i l l o o e t Phases, i t a l s o includes macro- and microblades, l e a f shaped and stemmed p r o j e c t i l e p o i n t s , and some arrow p o i n t s (Wilson 1980: 8 ) . Housepits are s a i d to be t y p i c a l l y s m a l l , round and l a c k i n g r i d g e s . The Kamloops Phase i s conceptualized by Wilson (.1980) as s t a r t i n g ca. 1400 BP, even though the e a r l i e s t absolute date obtained was 1140 + 100 BP (1980: 9 ). The phase i s otherwise as defined by Stryd (1973a,b,c), i n c l u d i n g the presence of l a r g e c i r c u l a r and ov a l housepits w i t h r i d g e s , and ca c h e p i t s . Wilson maintains that the Kamloops Phase was i n i t i a t e d l a t e r i n the Kam- loops l o c a l i t y "...because i n i t i a l i n t e n s i v e r i v e r i n e e x p l o i t a t i o n of the anadramous salmon occurred much l a t e r . . . " (1980:9). C. Carlson (1980) takes Wilson (.1980) to task w i t h respect to the d i f f e r e n c e s between the Thompson and Kamloops phases, based on her excavations of two s i t e s (EdRa 22 and EdRa 4) a l s o i n the Kam- loops l o c a l i t y . Carlson argues that there i s no good evidence f o r a s h i f t from hunting to f i s h i n g emphases i n the l o c a l economy, or popul a t i o n i n c r e a s e s . Carlson concludes that the only observable 82 trend from early to late in prehistory i s an increase i n frequency of small triangular side-notched points (Carlson 1980: 120). Thus the current Thompson-Kamloops Phase concept may be reflecting changes in a rather small part of material culture (mammal hunting technology) but does not lik e l y represent large scale changes in settlement and subsistence practices. In many respects this i s a defensible argument. No case for change in the diet or seasonality of people represented in the Nesikep Tradition has ever been firmly presented. Perhaps the major reason for this is poor preservation of faunal remains, a problem noted by Ham (n.d.) i n an analysis of faunal remains from several Lillooet region sites excavated by Stryd. Ham (n.d.) found deer and salmon to be the major species represented in prehistoric assemblages, with deer being replaced by horse in historic period remains (see Stryd 1980). Whitlam (1980) radiocarbon dated a l l u v i a l deposits at Lopez Creek (EeRh 3) near the town of Cache Creek, obtaining an age of 3920 + 65 BP (1980: 34), corrected to solar years to yield a date of 4448 + 144 BP. Unfortunately, i t is not possible to ascertain whether or not any artifacts are associated with the date. Other serious methodological problems render Whitlam's conclusion that the site exhibits time-transgressive occupation in discrete areas, highly questionable (see Magne 1982). Culture history in the southern Plateau i s currently tenuous before 3000 BP and only reasonably controlled in local areas for components dating since that time. Perhaps the most important 83 reason for this i s the continuing emphasis of housepit ex- cavation. Unlike the Columbia Plateau south of Wisconsinan glaci- atiori, this emphasis is due.to natural conditions of the Plateau of British Columbia where few, i f any rockshelters or caves suit- able for human habitation are available, and where soils nearly everywhere are thinly developed since glacial times, alternatives have rarely been considered. There have been no concerted system- atic attempts to discover aeolian sites, for example, nor has a research design to investigate cache pit variability in age, form and location ever been implemented. As'Wilmeth (1978b) has pointed out, the re-occupation of pithouse. depressions one.,or more times can lead to severe dis- ruptions of cultural stratigraphy and this can impede culture- historical methods. Fladmark (1982a) and Von Krogh (1980) also offer thoughts on d i f f i c u l t i e s associated with such sites, includ- ing: „the f i l t e r i n g of materials from the roof to the interior; the occurrence of this process once prior f i l l materials are used as roof insulation; differential decomposition of the structure, with intermittent partial i n f i l l i n g by aeolian, a l l u v i a l or f l u v i a l processes; and use of the house or resulting depression for non- habitation purposes such as tool manufacturing or garbage disposal (see also Kennedy and Bouchard 1978). Also, as Fladmark (1982a) points out, housepit excavations w i l l l i k e l y never yield data be- yond the 4000 years or so within which they are known to exist. Although Sanger (1970) did his best to isolate some general strat- 84 i g r a p h i c zones through, a r b i t r a r y l e v e l recovery, i t i s probable that Nesikep T r a d i t i o n m a t e r i a l s of e a r l i e r and l a t e r ages are mixed, the same i s probably true of most other multi-component housepit s i t e s that have been excavated. In my o p i n i o n , the mere existence of t h i s p r o b l e m — t h e continued re-use of s i t e areas by va r i o u s phases of p r e h i s t o r i c i n h a b i t a n t s of the Plateau- speaks l o u d l y f o r some degree of c o n t i n u i t y i n settlement and subsistence p a t t e r n s , r e g a r d l e s s of h a b i t a t i o n s t y l e or c u l t u r e "type". A determined e f f o r t , to f u l l y excavate a time-progressive s e r i e s of single-component housepits i s u r g e n t l y r e q u i r e d . 3.3.3. Settlement P a t t e r n Studies C u r r e n t l y there are only three p r o j e c t s that have c o n t r i b u t e d s u b s t a n t i a l data on the e n t i r e range of settlement-subsistence patterns of the I n t e r i o r P l a t e a u . The Shuswap Settlement P a t t e r n s p r o j e c t (.Matson et a l . 1979; Ham 1975), the Hat Creek p r o j e c t (Pokotylo 1978a; Pokotylo and Beirne 1978; Beirne and Pokotylo 1979) and the Eagle Lake p r o j e c t (Matson et a l . 1980) a l l employed r e g i o n a l sampling schemes to provide estimates of the range of s i t e types o c c u r r i n g i n f a i r l y l a r g e areas. A l l three of these s t u d i e s are of d i r e c t relevance to the present study s i n c e some s i t e s from each of these p r o j e c t s are analysed here. The purpose of the Shuswap Settlement Patterns p r o j e c t i n the southwest area of the confluence of the C h i l c o t i n and Fraser r i v e r s (the Mouth of the C h i l c o t i n ) was to study the environmental char- a c t e r i s t i c s of s i t e l o c a t i o n s and to use t h i s data and the m a t e r i a l 85 c u l t u r e evidence to t e s t the a p p l i c a b i l i t y of "sedentary" and "mobile" models of Canyon Shuswap settlement as provided i n the e x i s t i n g ethnographic record ( T e i t 1909a). As discussed i n the previous ethnographic review s e c t i o n , the Canyon Shuswap may have partaken of a d i f f e r e n t p a t t e r n than other Shuswap, main- t a i n i n g a prime salmon a c q u i s i t i o n t e r r i t o r y , a c t i n g a s . t r a d i n g middlemen between C h i l c o t i n and other Shuswap and p o s s i b l e a l s o between these and L i l l o o e t and Thompson. The a n a l y s i s of s i t e context and content by Matson et a l . (1979) demonstrated that s i x s i t e c l a s s e s are present i n the r e g i o n : r a v i n e cachepit s i t e s , ecotone cachepit s i t e s , housepit s i t e s , r i v e r s i d e s i t e s /Call but. one w i t h cachepits) , chert deb- i t a g e s i t e s , and unique s i t e s w i t h low a r t i f a c t frequencies. The chert debitage s i t e s are argued to pre-date the other s i t e s , which are s a i d to be Kamloops Phase, because the a r t i f a c t analyses show these to be d i f f e r e n t i n most r e s p e c t s , e s p e c i a l l y i n c o n t a i n i n g l a r g e corner-notched p o i n t s , and a l s o because Sanger (.1970) s t a t e d that chert i s most abundant as t o t a l debitage m a t e r i a l i n pre- Kamloops Phase components of the Nesikep T r a d i t i o n . Matson et a l . (1979) maintain; that the d i s t r i b u t i o n and composition of the other assemblages does not f i t T e i t ' s (1909a) observation that the Canyon Shuswap l i v e d i n four l a r g e pithouse v i l l a g e s , and propose that the more g e n e r a l i z e d , mobile model of Shuswap settlement i s a p p l i c - able to the Kamloops Phase occupations of the r e g i o n . 86 The information gathered by the Shuswap Settlement Pattern project was also used by Ham (.1975) in an M.A. thesis. Ham (.1975: 220 - 222) concluded that during the Kamloops Phase, two major settlement types prevailed: winter pithouse villages located on the upper benches of the Fraser River, and summer fishing camps next to the Fraser River. Ham (.1975: 210) also postulated that cachepit storage sites were the scenes of limited activities s t r i c t l y focused on salmon procurement, processing and storage. While Matson e_t al. (.1979) and Ham (1975) present reason- able evidence that the Canyon Shuswap were not entirely sedentary, I think that evidence indicates a kind of settlement pattern that . has been overlooked. It i s established that there are several kinds of subsistence orientations in evidence within an area of 2 some 40 km . Thus, given a maximum foraging radius of some 6 km, and a centralized radius of about 3 km, rather intensive use of a small area is indicated. This i s even more in evidence i f the estimated total of 247 sites (Matson jit al. 1979) within the grass- land zone is taken into consideration. It is also interesting to extrapolate these figures even fur- ther. Subtracting the estimate of 19.5 chert debitage or "pre- Kamloops" phase sites in the grassland zone from the estimated total, i t can be estimated that approximately 200 sites were formed in the laat 2000 years, or about one site every ten years. Approximately 65 housepit sites were constructed, inhabited and abandoned during this period or one housepit site every 33 years, representing 266 housepits, 87 or 1.3 houses every ten years. Further, 1046 cachepIts are. estimated in the grasslands population, representing approx- imately the use of one every two years, about four cachepits for each housepit. I suggest that these figures are f a i r estimates of housepit occupation spans and cachepit use spans for the region, and are indicative of repeated use of the re- gion. Unfortunately, these estimates cannot be compared to other areas, let alone other Shuswap occupation regions, but i f about 100 people were using the Mouth of the Chilcotin re- gion in pre-smallpox times, as is estimated by Teit (1909a), then perhaps Teit was describing intensive exploitation of a relatively small area by a relatively large group of people in early historic times, rather than purely "sedentary" people. Clearly, the answer to this problem requires a firm idea at least of housepit contemporaneity. Radiocarbon dating does not seem to be the complete answer, since wide standard devia- tions in dates and conflicting mean estimates of charcoal ages are the norm for materials within the last 1000 years (Stuiver 19.78). A more precise way to deal with the issue is dendro- chronology, and as Matson et a l . (1980), Stryd (1980), and Matson (personal communication 19.82) indicate, the present state of this method looks promising for future research on the Interior Plat- eau. Settlement pattern studies In the Upper Hat Creek Valley were initiated as a cultural resource management project designed to systematically recover archaeological data from a region planned 88 for development as an open pit coal mine and thermal generation plant. The region had not been the focus of any previous professional archaeological research, and was pract ical ly arch- aeologically unknown, yet was noted by Teit (1906) as being near the western edge of Spences Bridge Thompson terr i tory . Current- ly a great deal of data exists where none existed only seven years ago. Two impact assessment reports (Pokotylo and Beirne 1978; Beirne and Pokotylo 1979) and a Ph.D. dissertation (Poko- tylo 1978a) as well as shorter papers (Pokotylo 1978b, 1979a, 1981, Pokotylo and Beirne 1983) have been written that discuss the significance of the 200-odd sites presently known, in a re- gion that has only seen ca. 15% areal sampling. This discussion is limited to the results that bear expl ic i t ly on the re lat ion- ships observed between settlement patterns and l i t h i c technology. Pokotylo's research goals were stated as follows: 1) describe patterns of settlement u t i l i za t ion reflected by byproducts of l i t h i c technology in Upper Hat Creek Valley, and 2) compare this with patterns of stone tool dep- osit ion (1978a: 2) As is discussed in previous and following chapters, the analytic methods were pioneering in several ways, part icularly in exp l i c i t ly relating l i t h i c technological processes to the formation of surface l i t h i c scatters. The results indicated that the site classes f a l l within the range of var iab i l i t y expected from regional ethnographic 89 accounts, and that the s i t e s are r e p r e s e n t a t i v e of base camps, hunting and butchering of l a r g e game and more general a c t i v i t i e s probably r e l a t e d to root crop a q u i s i t i o n and processing. I t was a l s o found that s i t e s w i t h a wide range of t o o l manufacturing steps are found i n areas w i t h high l o c a l environmental d i v e r s i t y as measured by nearby v e g e t a t i o n community and drainage c h a r a c t - e r i s t i c s , w h i l e s i t e s w i t h more l i m i t e d t o o l manufacturing assem- blages are found i n areas w i t h low l e v e l s of environmental v a r i a - b i l i t y . S i t e s that were i n f e r r e d to represent long term occu- pations tend to be s i t u a t e d c l o s e to permanent sources of water (Pokotylo 1978a: 323). Low sample s i z e s of s i t e types o c c u r r i n g w i t h i n d i s c r e t e environmental zones prevented p r o b a b i l i s t i c e v a l - u a t i o n of these trends, yet o v e r a l l the study demonstrated q u i t e s u c c e s s f u l l y that subsistence and settlement p r a c t i c e s w i t h i n upper and middle e l e v a t i o n areas of the southern P l a t e a u produce patterns of l i t h i c assemblage v a r i a b i l i t y that can be detected w i t h a combination of t e c h n o l o g i c a l and t y p o l o g i c a l approaches. I t should be noted i n t h i s review that s e v e r a l of the Hat Creek s i t e s analysed contained microblades and formed u n i f a c e s (endscrapers), considered r e p r e s e n t a t i v e of E a r l y Nesikep period assemblages. Pokotylo (1978b) considered that 66% of the Hat Creek assemblages c o l l e c t e d i n 1976 belonged to the E a r l y Nesikep T r a d i t i o n . Pokotylo's (1978a) analyses showed that u n i f a c e s and microblades tend to be mutually e x c l u s i v e i n the Upper Hat Creek V a l l e y . A l s o , while i t i s apparent that the debitage from these E a r l y Nesikep 90 s i t e s i n d i c a t e s e i t h e r i n i t i a l t o o l manufacturing steps, or a wide range of stages, the s i t e s are not a s s o c i a t e d w i t h a d l s ^ Crete range of environmental v a r i a b l e s (Pokotylo 1978a: 328 - 329)• Again, i t i s v a l u a b l e to consider the broader i m p l i c a t i o n s of the temporal patterns of r e g i o n a l trends. Assuming that the microblade and formed u n i f a c e s i t e s are indeed E a r l y Nesikep i n age (7000 to 5000 BP), then i t appears that two kinds of e a r l y s i t e s are present, given the near-mutual exclusiveness of the two a r t i f a c t types: s i t e s r e p resenting the need f o r c u t t i n g t o o l s , and others where scraping (or c h i s e l l i n g and adzing) were r e q u i r e d . In the Late Nesikep periods of time, a much wider range of a c t i v - i t i e s were undertaken. This included the establishment of base camps w i t h a l a r g e amount of "maintenance" a c t i v i t i e s , and s e v e r a l kinds of s a t e l l i t e camps at which r o o t s were processed or l a r g e game butchered, and q u i t e short term occupation l o c i used to s t a l k game, r e p a i r t o o l s , or simply manufacture t o o l s from l o c a l raw mat- e r i a l s . This perhaps r e f l e c t s a b a s i c d i f f e r e n c e i n settlement p a t t e r n s , where E a r l y Nesikep populations used Hat Creek V a l l e y as an important but marginal area, and were centered more i n major water- shed areas such as Lochnore-Nesikep, where microblades and formed un i f a c e s are found together and i n a s s o c i a t i o n w i t h a wide range of other t o o l s . In Middle and Late Nesikep p e r i o d s , the Hat Creek V a l l e y was a more important range of the settlement p a t t e r n , where l a r g e r groups of people s e t t l e d at l e a s t t e m p o r a r i l y , and organ- i z e d a complex set of subsistence tasks r e q u i r i n g a greater degree 91 of technological specificity. The Eagle Lake project (Matson et̂  al. 1980) was aimed at describing the material culture, settlement and subsistence pat- terns of Athapaskan-speaking Chilcotin peoples in the southern area of their historically-reported territory. A major aspect of the project was an examination of ways to define archaeologically- observable differences between Chilcotin and Interior Salish pat- terns, and thus a region that had environmental similarities to those of the Shuswap Settlement Pattern project and the Upper Hat Creek Valley was chosen. In this manner, the open grassland and dry pine forest environments could be used as a kind of constant, implying a limited range of potential subsistence and settlement practices, to enable ethnic differences in material culture to appear more clearly. An area representing about 7 percent of the region around Eagle or Choelquoit Lake was surveyed using 400 m X 400 m quadrats, randomly sampled with replacement. A total of 35 quadrats yielded 46 sites. Comparison of the numbers of pit features and artifacts recorded within the quadrats showed the Eagle Lake region to be much more similar to the Mouth of the Chilcotin region than to Upper Hat Creek. Cachepits and housepits are relatively common, and l i t h i c scatters usually do not contain a great many tools or debitage. At Hat'Creek, cultural depressions other than roasting pits are quite rare, and surface scatters often contain hundreds of items. At Eagle Lake, sites average 46 artifacts, and at the Mouth of the 92 Chilcotin each site oh the average contains 90 artifacts, while in the Hat Creek Valley, sites contain an average of 1450 items (Matson e_t a l 1980: 208). Since i t was important also to be able to date the arrival of the Chilcotin in the area, a survey was conducted along 30 km of the Chilko River, in the eastern end of the study area, in an attempt to locate non-housepit stratified sites that could be reliab- ly ., dated. No such site was found in the 105 sites recorded. The finding of no microblades and only a few large, or a t l a t l , projectile points, and late radiocarbon dates from three sites (.280 + 80 BP; 360 + 80 BP; 800 + 80 BP) a l l appear to indicate that the last half of the Late Nesikep period was the only time of major occupation of the region. The small size of most assemblages and the lack of good chronological control were limiting factors in terms of project goals, yet despite . these drawbacks, significant contributions were made and ethnic differences were perceived. In a multi- variate analysis of projectile points, small, triangular side- notched points were shown to be highly discrete with respect to Salish (Mouth of the Chilcotin and Hat Creek) and Athapaskan (Chinlac and Punchaw Lake) provenience, and Eagle Lake points occur in both these kinds of groups (Magne and Matson 1982). Through this information and other data, specific sites were identified as Athapaskan occupations, including an isolated, shallow, rect- angular depression site near a small lake (also with a Kavik-style 93 point and a blue trade bead), a l i t h i c scatter site, and a small isolated circular housepit site with a single com- ponent that has been tentatively dendro-dated to AD 1561 w (outside very variable; Matson e_t al 1980; Matson personal communication 1982). Currently proposed research w i l l focus on excavations at the two putative Athapaskan dwellings noted above and at another more typically Kamloops Phase housepit site. Some of the research undertaken in the Eagle Lake project initiated the present study. The experimental debitage program described in the following chapter was piloted by a biface re- duction experiment f i r s t reported in the Eagle Lake project (Magne and Pokotylo 1980, 1981), and an analysis of the debitage from 24 Interior Plateau assemblages (Magne 1980) was a t r i a l investigation of large scale technological patterns within settle- ment types that is more f u l l y developed in Chapter 6. 94 CHAPTER 4 THE EXPERIMENTS IN DEBITAGE CLASSIFICATION 4.1. Introduction The objective of the experiments described in this chapter is to determine the degree to which general chipped stone tool manufacturing stages can be inferred from l i t h i c debitage. The specific goal of the program is to devise an efficient debitage classification of manufacturing stages that can be applied to archaeological collections. A secure debitage classification of reduction stages is re- quired to enable intersite comparisons of a multi-regional set of l i t h i c assemblages in expl i c i t l y technological terms. Such a classification has relevance well beyond this study. The general- ized approach to the experiments is in contrast to the particular- i s t i c and precise "replication" concerns that characterize most other l i t h i c experiments (see Johnson 1978). This study, in con- trast to others, concentrates on debitage, rather than on specific tool forms, where debitage is defined as non-utilized products of stone tool manufacturing and maintenance. The lack of attention that has been paid to debitage in l i t h i c technological research is surprising, since i t has several qualities that are desirable for reconstructing past processes of l i t h i c re- duction and settlement technology, including: 1. Debitage is not 95 transported, or otherwise curated to the same extent that tools are; 2. Since i t results from reductive, rather than additive processes, debitage retain evidence of previous stages of manu- facture; and 3. Debitage is very abundant and is thus suited to sampling and s t a t i s t i c a l procedures (Leach 1969; Collins 1975; 17, 19; Sheets 1975; Fish 1976). 4.2. Experimental Controls The faults of previous experimental work, as outlined in Chapter 2, and the above issues were kept in mind when designing the following controls for the present experiment: 1. The most important control factor is that flakes removed from cores, blanks, or preforms were gathered in the precise or- der of their removal, in contrast to studies that have gathered groups of flakes derived from estimated stages of reduction (e.g. Collins 1974; Burton 1980; Stahle and Dunn 1981). Burton (1980: 132) mentions that he numbered flakes consecutively, but apparently this information was not used in his study. This control factor enables reduction stages to be precisely defined in a uniform man- ner, regardless of the tool form being made (see below). Each blow that produced flakes is termed an "event" of the reduction sequence, and as can be seen in Table 2, reduction events often produce several flakes. Following each event, a l l flakes greater than 5 mm in their largest dimension were gathered by a flake retrieval person, placed in order of removal on card- 96 TOOL TYPE RAW # OF FLAKE # OF # OF # OF # OF MATERIAL EVENTS TOTAL PRB'S SHATTER BRF'S BPO'I Large Core Basalt 42 317 144 173 2 Bipolar Core Basalt 3 36 1 33 7 Bipolar Core Basalt 2,8 62 12 43 2 Bipolar Core Basalt 7 54 5 47 4 Bipolar Core Obsidian 4 43 3 36 1 Bipolar Core Obsidian 6 54 5 48 7 Bipolar Core Obsidian 12 137 10 118 2 Large Biface Basalt 186 399 116 246 37 Large Biface Basalt 44 87 33 35 19 Large Biface Basalt 97 590 150 425 15 Large Biface Obsidian 80 207 45 143 19 Ovoid Biface Basalt 60 105 44 52 9 Ovoid Biface Obsidian 67 82 37 39 6 Bimarginal Basalt 36 43 25 12 6 Bimarginal Basalt 35 58 25 23 10 Ovoid Uniface Basalt 45 64 29 35 Ovoid Uniface Basalt 47 70 43 27 Ovoid Uniface Basalt 29 43 11 32 Endscraper Basalt 13 21 9 12 Endscraper Obsidian 17 21 12 9 Endscraper Chert 16 18 11 7 Unimarginal Basalt 9 10 6 4 Unimarginal Basalt 25 26 22 4 Unimarginal Basalt 9 14 9 5 Unimarginal Basalt 44 55 32 23 Unimarginal Basalt 11 16 6 10 Unimarginal Chert 17 25 11 14 TOTALS 2657 856 1655 123 23 PERCENTAGES OF TOTAL DEBITAGE 32.2 62.3 4.6 0 PRB'S = Platform remnant bearing flakes BRF'S = Biface reduction flakes BPO'S = Bipolar reduction flakes TABLE 2. Frequencies of General Flake Classes and Reduction Events for Each Experimental Core and Tool. 97 board trays, and later catalogued. Since an archaeological sample of several thousand flakes was expected, a size cut-off of 5 mm was maintained in.the analysis of the experimental debitage, and this meant that pressure flaking could not be investigated. 2. A l l procedures involved in knapping were recorded on a standardized reduction form, requiring the knapper to note.the event at which he or she prepared platforms, changed technique, and so on (Appendix 1)'. Only stone hammers and antler b i l l e t s were used. Knappers also recorded the event at which they f e l t to have moved on to a sequent reduction stage, and were asked to note any d i f f i c u l t i e s experienced. This information was gathered mainly as back-up data, in the event that the objective reduction stages did not produce useful results. The knappers were also asked to provide measurements and scale drawings of cores and blanks prior to reduction as well as of finished products, although not a l l did so. 3. A total of 13 knappers of widely-ranging expertise produced the materials rather than a single expert. Nine of these were stu- dents in a course on archaeological laboratory methods (ANTH 406), instructed by Dr. R.G. Matson, and only the remaining four knappers can be considered truly experienced in the craft. These include R.G. Matson, David Pokotylo, George Kurzenstein, and myself. This may be seen as a "randomizing" process rather than a true control factor, but is desirable to eliminate the potential in systematic error that could occur in trying to apply the experimental results from a single 98 knapper to a r c h a e o l o g i c a l m a t e r i a l that must have been produced by many knappers. 4. The knappers were shown p h y s i c a l models and were given w r i t t e n d e s c r i p t i o n s of the t o o l s they were to attempt to r e p l i - c ate. These included s e v e r a l kinds of b i f a c e s , p r o j e c t i l e p o i n t s , scrapers, and cores from v a r i o u s s i t e s of the I n t e r i o r P l a t e a u . Several products were q u i t e inadequate r e p l i c a s and were prompt- l y removed from the a n a l y s i s . 5. The raw m a t e r i a l s employed were those that were used by p r e h i s t o r i c peoples of the regions of i n t e r e s t . The most common m a t e r i a l of the I n t e r i o r P l a t e a u l i t h i c technologies and i n the experiments i s high q u a l i t y b a s a l t that ranges i n t e x t u r e from v i t - reous to gra n u l a r , and that i s a v a i l a b l e as g l a c i a l t i l l cobbles or stream bed cobbles i n many areas. The b a s a l t used i n t h i s study was obtained from the Upper Hat Creek V a l l e y and from Cache Creek. Obsidian cobbles from Obsidian Creek i n the Anahim Lake area, and stream cobble chert derived from the Cache Creek Form- a t i o n were a l s o used. 6. Debitage that were thought to be of a s i z e s u i t a b l e f o r f u r t h e r r e d u c t i o n were removed from the a n a l y s i s . This p r a c t i c e i s meant to c o n t r o l f o r the e f f i c i e n t use of stone i n an a c t i v e l i t h i c technology. A l l previous experiments that have used l a r g e debitage i n t h e i r analyses have not considered that such l a r g e f l a k e s could be formed i n t o a wide range of items. The c u t - o f f used here i s u s u a l l y 30 grams, but t h i s was not s t r i c t l y maintained. 99 4.3. The Pilot Study An important preliminary step of the experiment was to under- take a pi l o t study, to enable more complete appreciation of the controls required, to refine the hypotheses to be tested, and to explore the range of v a r i a b i l i t y in debitage in a preliminary fashion. The pilot study als served to familiarize the present researcher with multivariate data analyses, and i n some ways can be seen as the sort of exploratory data analysis that is ad- vocated by Clark (1982). A f u l l description of the preliminary study i s in Magne and Pokotylo (1981). Briefly, debitage that resulted from the manufacture of flake blanks and a single biface waseanalysed. The major factors of quan- titat i v e v a r i a b i l i t y were derived from a judgemental, visual comparison of the data for eight flake variables. Instead of this search for major "alignments" with the reduction sequence, a more appropriate solution would have been to regress the raw data against the individual factor scores. On the basis of the preliminary methods, a debitage class- i f i c a t i o n was formulated using c r i t e r i a of flake weight and platform presence or absence, and scar counts and cortex cover were used as secondary c r i t e r i a . The classification is now understood to contain logical faults such as non-exclusiveness of certain variables, btit i t was used to examine seven sites of the 44 that were previously analysed by Pokotylo (1978). The interpretations were very close to those of Pokotylo's original study, and patterns such as b i f a c i a l tool production were revealed, which were not previously evident. Overall, the pilot study was moderately successful!, given correspon- 100 dence with Pokotylo's (1978) interpretations, and also independently supported the findings of other studies, such as Burton (1980) that found flake size to be a highly significant factor in debitage var i a b i l i t y . However, I thought that the sample size needed en- larging, that bipolar flaking would need to be investigated, and that analytic .methods leading to a reliable stage classification of debitage would need to be refined. 4.4. Experimental Products Seven cores and 20 "tools" were the retained products of the l i t h i c reduction sessions. These include one single platform core, six bipolar cores, six large bifaces, two bi-marginally re- touched flakes, three large unifaces, three endscrapers, and six uni-marginally retouched flakes. These items are shown in Figures 5 to 10; unfortunately, the single platform core was accidentally reduced by an anonymous person and was not photographed. After flakes that were thought to be suitable as blanks for further reduction were removed from the debitage, the f i n a l actual experimental sample comprised 2657 flakes greater than five m i l l i - metres in their largest dimension. Of these, 856 are platform rem-̂ nant bearing flakes or PRB's (see Knudson 1973); 1655 items are shatter, that i s , flakes lacking striking platforms. Another 1-23 are biface reduction flakes, or BRF's, that are recognized by extensively facetted, narrow angle and often "lipped" platforms (see Crabtree 1972), and 23 are bipolar reduction flakes (BPO's) having evidence of simul- taneous percussion from opposite directions, often with crushing. F i g u r e 5. F l a k e b l a n k s removed f r o m l a r g e s i n g l e - p l a t f o r m b a s a l t c o r e . Not a l l a r e shown. F i g u r e 6. B i p o l a r c o r e s and d e r i v e d b l a n k s a,b,d,f: Cache Creek b a s a l t c o r e s . c,e: O b s i d i a n Creek o b s i d i a n c o r e s , 1-5; b l a n k s Figure 7. Large b i f a c i a l t o o l products. a,b: Obsidian; c,d,e,f: V i t r e o u s b a s a l t ; d i s p i l o t study product. a b Figure 8. Large u n i f a c i a l t o o l products. a,b: V i t r e o u s b a s a l t c: Granular b a s a l t . Figure 9. Large marginal t o o l products a: v i t r e o u s b a s a l t , b i m a r g i n a l b,c: granular b a s a l t , unimarginal d: Cache Creek c h e r t , unimarginal c f 9 Figure 10. Small marginal t o o l products a,c,d,e,f: v i t r e o u s b a s a l t b: Obsidian; g: Cache Creek chert 104 The frequencies of these general f l a k e types are provided i n Table 2. I t can be seen t h a t i n the r e d u c t i o n of one of the b i p o l a r cores, two f l a k e s were cl a s s e d as BRF's. This i s not e n t i r e l y s u r p r i s i n g s i n c e b i p o l a r r e d u c t i o n , c a r r i e d to f i n a l stages, i s o f t e n b i f a c i a l i n nature. Only 32.2% of the f l a k e s e x h i b i t remnant pla t f o r m s , and i t was observed that j u s t 48% of these PRB's have fea t h e r t e r m i n a t i o n s . Given that the f l a k e s were produced i n l a b o r a t o r y c o n d i t i o n s , i t i s l i k e l y that r e l a t i v e l y more are "complete" than i n most a r c h a e o l o g i c a l s i t u a t i o n s , where trampling may break a l a r g e number of f l a k e s . I t i s evident at any r a t e , that s t u d i e s that have s e l e c t e d only "complete" f l a k e s (e.g. C o l l i n s 1974; F i s h 1976; S t a h l e and Dunn 1981) f o r a n a l y s i s of r e d u c t i o n s t r a t e g i e s have l i k e l y ignored a great deal of d e b i - tage v a r i a b i l i t y . Table 3 i l l u s t r a t e s another i n t e r e s t i n g p a t t e r n . Here the average number of a l l f l a k e s removed per r e d u c t i o n event and PRB/ s h a t t e r r a t i o s ^ a r e t a b u l a t e d by general r e d u c t i o n type and raw mat- e r i a l . I t appears that core r e d u c t i o n of any k i n d produces more f l a k e s per blow than b i f a c i a l r e d u c t i o n , which i n t u r n produces more f l a k e s per blow than u n i f a c i a l r e d u c t i o n . PRB/'shatter r a t i o s do not support the same trend observed i n the f l a k e s per event tab- u l a t i o n s , and may i n d i c a t e that the c o n t r o l exerted i n f l a k i n g pro- cedures may not be e a s i l y accounted f o r i n a r c h a e o l o g i c a l c o l l e c t i o n s . However, i t i s apparent that b i p o l a r r e d u c t i o n produces very few PRB's i n r e l a t i o n to s h a t t e r (on the order of 10 s h a t t e r per PRB), and that 105 MEAN # OF ALL FLAKES PRB/SHATTER REDUCTION TYPE/ PER EVENT RATIO RAW MATERIAL 10.46 .09 Bipolar Obsidian 7.55 .83 Core Reduction Basalt 4.00 .15 Bipolar Basalt 2.80 .45 Bifa c i a l Basalt 1.97 1.16 Unifacial/marginal Basalt 1.30 1.05 Unifacial/marginal Chert 1.24 1.33 Unifacial/marginal Obsidian TABLE 3. Mean Number of Flakes Per Reduction Event and PRB/Shatter Ratio, In Grouped Reduction Types by Raw Material. 106 s i n g l e p l a t f o r m core r e d u c t i o n , when a p p l i e d w i t h the purpose of d e r i v i n g l a r g e blanks, can y i e l d about twice as many PRB's per Shatter as b i f a c i a l r e d u c t i o n . U n i f a c i a l - and marginal f l a k i n g produce about equal numbers of PRB's and Shatter. These are obviously not r e l i a b l e trends, s i n c e only one u n i f a c i a l , and no marginal o b s i d i a n t o o l s are represented;, chert f l a k e s were produced only by u n i d i r e c t i o n a l knapping. . S i n g l e p l a t f o r m core r e d u c t i o n i s only represented by one set of b a s a l t r e d u c t i o n events. These p a r t i c u l a r data were not f u r t h e r analysed due to these l i m i - t a t i o n s of the sample, and because such trends are aside from the main d i r e c t i o n of the experiments. 4.5. Stage D e f i n i t i o n One of the purposes of t h i s study i s to construct a debitage c l a s s i f i c a t i o n that a c c u r a t e l y r e f l e c t s r e d u c t i o n stages from a r e l a t i v e l y l a r g e sample of a r c h a e o l o g i c a l m a t e r i a l s . I t would be h i g h l y i m p r a c t i c a l to d i v i d e stages very narrowly. In the extreme, Muto (1971b: 111) has s t a t e d that i t i s p o s s i b l e to regard each r e - d u c t i o n event as a "stage". I t was t h e r e f o r e decided to t e s t f o r m e t r i c v a r i a b i l i t y i n q u i t e general terms, simply: e a r l y , middle and l a t e r e d u c t i o n stages. E a r l y r e d u c t i o n stages are defined as a l l events of core reduc- t i o n , i n c l u d i n g both s i n g l e p l a t f o r m and b i p o l a r core forms, r e - gardless of the number of events i n v o l v e d . Middle stages are the primary trimming stages of t o o l s , measured as a l l the r e d u c t i o n events of marginal retouch t o o l s , and the f i r s t h a l f of the r e d u c t i o n events of 107 a l l other tools, whether unifacia l or b i f a c i a l . Late stage re- duction then, is defined as the latter half of the reduction events of unifacia l and b i fac ia l implements. I believe that this is a just- i f iable way to divide the reduction process, since core reduction is undertaken to derive flake blanks, regardless of method, mar- ginal flaking and i n i t i a l uni fac ia l and b i fac ia l flaking a l l involve straightening edges and removing the most of excessive mass, and the later events of unifacia l and b i fac ia l flaking are undertaken to refine the intended shape of the tool . This method of defining stages requires no subjectivity as to what exactly constitutes "primary" or "secondary" trimming,( e. g. Coll ins 1975). The number of events in core reduction ranged from 4 to 42, and from 9 to 186 in tool reduction (Table 2). Middle stage events range from 7 to 93, and late stage events vary in frequency from 8 to 93. These events, i t should be noted, are not the dividing points in stages that were noted by individual knappers on the re- duction recording forms. In addition to the three stages, I thought i t useful to d i s - tinguish b i f a c i a l and bipolar reduction flakes objectively. Such a dist inct ion would add to the dimensions of the c lass i f icat ion , per- mitting more refined interpretations of archaeological assemblage v a r i a b i l i t y . Thus while biface reduction flakes (BRF's) do exhibit platforms and are generically "PRB's", several of the analyses to follow attempt to demonstrate the distinctiveness of BRF's. 108 4.6. Debitage V a r i a b l e s Several s t u d i e s r e p o r t the use of u n i v a r i a t e , b i v a r i a t e and m u l t i v a r i a t e s t a t i s t i c a l techniques to reduce debitage v a r i a b l e l i s t s (e.g. F i s h 1976; Katz 1976; Pokotylo 1978). Fo l l o w i n g Pokotylo (1978) and the p i l o t study (Magne and Pokotylo 1981), I decided to s e l e c t v a r i a b l e s from the r e s u l t s of these, to dev- elop a robust short l i s t of v a r i a b l e s . Two weaknesses charact- e r i z e most approaches to debitage v a r i a b i l i t y : an over-emphasis on d i s c r e t e , r a t h e r than continuous or o r d i n a l v a r i a b l e s , and stu d i e s of b r i t t l e s o l i d f r a c t u r e dynamics w i t h l i t t l e e x p l i c i t b e h a v i o r a l value (e.g. Speth 1972, 1975; Bonnichsen 1977; P a t t e r - son and S o l l b e r g e r 1978). Barton (1979), Bonnichsen (1977) and Speth (1972) attempt to def i n e v a r i a b l e s t h a t can d i f f e r e n t i a t e hard-hammer and s o f t - hammer percussion, by c o n t r o l l e d experiments, the l a t t e r two going as f a r as using cut gl a s s cores. While a d i f f e r e n c e i n t h i s type of knapping may be i n t e r e s t i n g , there i s l i t t l e attempt to s t a t e what differences could mean i n b e h a v i o r a l terms. C o n t r o l over l i t h i c f r a c t u r e i s a l s o the prime concern i n Phagah's (1976) notable d i s c u s s i o n of the value of 28 d i f f e r e n t f l a k e v a r i a b l e s . However, as i n most s i t u a t i o n s where v a r i a b l e s are a p p l i e d to a r c h a e o l o g i c a l debitage without recourse to c o n t r o l l e d experiments, the meaning of v a r i a b i l i t y i s untested, and merely suggested. Thus, Phagan's (1976) i m p r e s s i o n i s t i c , and not s t a t i s t i c a l l y i n f e r r e d r e s u l t s are weakened. Some of the Ayacucho assemblages seemed to have been produced by 109 s p e c i a l i z e d groups at long-term occupation s i t e s , but when assemblages are mixed, no i n t e r p r e t a t i o n s are p o s s i b l e (Phagan 1976: 104 - 110). Phagan notes (1976: 110) h i s e x p l i c i t app- roach to waste f l a k e s and t e c h n o l o g i c a l systems i s a c o n t r i b ^ u t i o n to b e h a v i o r a l approaches to assemblage v a r i a b i l i t y that g r e a t l y improves upon t r a d i t i o n a l t o o l t y p o l o g i c a l approaches, because i t seeks to consider a l l the t e c h n i c a l aspects of f l a k e production. Two s t u d i e s that have d i r e c t relevance to t h i s study i n t h e i r o r i e n t a t i o n and.purpose are Pokotylo (1978) and Katz (1976). Pokotylo (1978) a p p l i e d a 19 v a r i a b l e l i s t to 198 f l a k e s w i t h remaining s t r i k i n g platforms that were obtained from f i v e s i t e s i n Upper Hat Creek V a l l e y . An R-mode f a c t o r a n a l y s i s reduced t h i s l i s t to f i v e v a r i a b l e s . Nine were f i n a l l y used to d e r i v e settlement p a t t e r n and b e h a v i o r a l i n f o r m a t i o n f o r 44 s i t e s . The f a c t o r a n a l y s i s i n d i c a t e d that f l a k e s i z e i s the most important metric f a c t o r i n debitage (PRB) v a r i a b i l i t y , f o l lowed by f l a k e angle ( p l a t f o r m to d o r s a l or v e n t r a l f a c e s ) , d o r s a l f l a k e scar count, d o r s a l scar p a t t e r n i n g , presence of v e n t r a l l i p p i n g , and bulb of f o r c e " s a l i e n c y " (Pokotylo 1978: 204 - 208). Katz' (.1976) s i m i l a r study evaluated nine a t t r i b u t e s u s i n g a sample of 293 f l a k e s from a refuse p i t at the D e i s t e r s i t e , a Kan- sas C i t y Hopewell occupation. A P r i n c i p a l Components a n a l y s i s pro- duced three meaningful v e c t o r s , c o n s i s t i n g of weight, number of dor- s a l s c a r s , and p l a t f o r m angle (Katz 1976). S i x t e e n d i s c r e t e v a r i a b l e s 110 were evaluated using a non-metric multidimensional scaling procedure, unlike Pokotylo's (1978) data reduction analysis, where continuous and ordinal variables were analysed simultan- eously. Three clusters of attributes were derived, including raw material type, applied force, and control over flake re- moval. (Katz 1976). Katz retained eight variables, and using the presence of cortex to "pin down" the early stages of manu- facture, posited a six-stage sequence of l i t h i c manufacturing for the assemblage. Overall, Katz' and Pokotylo 1 s-findings are very similar: the size of flakes, their evidence of prior flake removals, and their platform angles are highly useful in describing tool manu- facturing sequences. Both studies have the weakness of deriving short variable l i s t s from archaeological debitage, inferring the meaning of those variables, and then applying them to a- larger sample of archaeological debitage from the same region, or even the same site. This is a rather circular process poss- ibly inducing a sample bias, and may have served to affirm the consequent, especially in Katz' (1976) analysis, where only one site is being examined. Both studies, however, present the most robust examinations of debitage v a r i a b i l i t y presently available, yet i t i s clear that both pin-pointed specific variables that, when used, w i l l reduce overall metric redundancy in archaeological application, but s t i l l do not answer the basic question of sequen- t i a l v a r i a b i l i t y , except by inference gained by co-association of I l l variables. This .problem i s especially crucial\fof weight and size variables. As in the pilot study for the present experiment, both studies indicate that weight or size is the variable that accounts best for overall metric v a r i a b i l i t y in debitage. This means that a l l other variables co-vary with size better than with other variables, and not necessarily that weight varies in any other, independent direction such as reduction sequence. Such an interpretation then, s t i l l remains to be tested by experimental means. Pokotylo and Katz each e f f i c - iently reduced the number of variables to be coded on each flake, and found a short hand way of measuring overall debitage va r i a b i l i t y , with some theoretical grounds for proposing that the selected variables were correlated with reduction stages. Six variables were retained for use in the present anal- yses (see Figure 11). These are defined below, with expectations of how each might pattern through sequential reduction. Four- letter abbreviations are also given, to be used as conventions in following discussions. The l i s t is deliberately short. I think that quite enough redundancy has been demonstrated by prior workers, and I needed to keep data gathering time relatively brief, since archaeological analysis was yet to come. The application of this variable l i s t to the experimental debitage required six weeks of almost daily work. I estimate that recording time would have been doubled simply by the addition of two variables such as length and width, that would have required the use of vernier c a l l - 112 FORCE APPLICATION FLAKE REMOVAL Dorsal Side PLATFORM REMNANT BEARING SHATTER F i g u r e .11. D e b i t a g e a t t r i b u t e s employed i n the e x p e r i m e n t a l program. 113 ipers. Attributes of the variables were recorded on 80 column computer coding forms, along with flake termination type, knapper, raw material, tool number, event sequence, flake within event (arbitrary), and f i n a l flake number (acc- urate to event only). The data were keypunched and stored as disk f i l e s . 1. Weight (WEIT): The weight of each flake was taken to .10 gram with an electronic balance. As reduction proceeds, i t can be expected that the weight of individual items w i l l strong- ly tend to decline. This variable is used as a general measure of size. 2. Dorsal Scar Count (DOCO): This is the number of flake scars vi s i b l e on the dorsal face of the flakes, counting only those greater than 5 mm in size. One can expect that the number of flake scars on dorsal faces w i l l tend to increase through the reduction process. 3. Dorsal Scar Complexity (DOSC): This is a new variable, mod- ifi e d from that of Munday (.1976: 123) and Pokotylo (.1978) . Here flakes are centered on polar coordinate paper divided into 10 vectors, and the number of directions that flake scars originate from are counted. This measure should increase with reduction sequences. Note that a flake may have several scars, but low complexity. 4. Platform Scar Count (PLCO): This i s the number of scars re- gardless of size actually occurring on the flake platform, and is 114 a p p l i c a b l e to PRB's only*-- This does not i n c l u d e f l a k e scars formed on the d o r s a l surface of f l a k e s adjacent to p l a t f o r m s , sometimes r e f e r r e d to as "preparation s c a r s " (Phagan 1976: 49). Recording t h i s v a r i a b l e was f a c i l i t a t e d by using a 2X i l l u m i - nated m a g n i f i e r , although platforms l e s s than 2 mm deep were oft e n d i f f i c u l t to code, and c l a s s e d as s h a t t e r . This measure i s expected to increase as r e d u c t i o n proceeds. 5. P l a t f o r m Angle (PLAN): The d o r s a l angle of PRB's i s measured to the nearest 5 degrees w i t h a goniometer that makes contact at 1 cm, 5 mm or 2 mm distances along the p l a t f o r m and•: d o r s a l faces simultaneously, depending on p l a t f o r m depth. Again, f l a k e s w i t h platforms l e s s than 2 mm deep were d i f f i c u l t to measure and were of t e n coded as s h a t t e r . This v a r i a b l e should decrease w i t h sequen- t i a l f l a k e removal (see Raab, Cande and Stahle 1979) , although Katz (1976) i n f e r r e d that p l a t f o r m angle increases through sub- sequent stages. 6. Cortex Cover (COCO): This i s the amount of weathered surface evident on the f l a k e s ' d o r s a l s u r f a c e s , measured i n s i x increments of 25% ( i n c l u d i n g 0% and 100%), and assessed v i s u a l l y . This mea- sure i s expected to decrease very sharply f o l l o w i n g core r e d u c t i o n . 4.7. Hypothesis Testing F o l l o w i n g the completion of the p i l o t study, and p r i o r to conducting the experiment i n i t s e n t i r e t y , the f o l l o w i n g hypoth- eses were formulated w i t h the goal of demonstrating that stone t o o l 115 manufacturing stages can be reconstructed from quantitative analysis of l i t h i c debitage, employing the six continuous and ordinal variables. HI: The weights of individual flakes are the best indicators of the reduction stages from which they originated. H2: Bif a c i a l reduction flakes and bipolar reduction flakes are discrete items indicative of each type of reduction and can be accurately identified by the same variables used to predict early, middle and late stages. H3: Reduction stage quantification is independent of raw material type. The kind of s t a t i s t i c a l technique that i s required to test these hypotheses, and especially the general stage question i s some kind of factor analysis, where the "factors" are known (i.e. stages, BRF's, BPO's), but the significance of variables is not. This technique w i l l also need to be able to identify variables that best sort factors. Also needed is some form of non-normal or non-parametric test of significance to identify discrete patt- erning of the variables, with respect to stages, so that a class- if i c a t i o n can be constructed. The two techniques chosen to sat- isfy these requirements are multiple discriminant analysis (MDA: 116 Klecka 1975), and the chi-square test of independence in con- tingency tables (Mendenhall 1975). Discriminant analyses have been used in l i t h i c experimentation studies by Chandler and Ware (1981), and in a combined experimental and archaeological study by Burton (1980). Stepwise MDA (Wilk's method) was used to see i f differences exist between groups, and to discover which variables are most useful. Simply stated, MDA uses the six variables to classify individual flakes into the pre-set classes as defined by within-group co-variance parameters (see Klecka 1975). 4.7.1. Stage Prediction Five groups were identified for this analysis: early, middle and late stages, as well as biface reduction (BRF) and bipolar reduction (BPO) flakes. The f i r s t stepwise discriminant analysis employed a l l flakes with platforms, including a l l BRF's and 15 BPO's. (N = 994). An overall accuracy of 58.15% in discriminating the five flake classes was achieved (Table 4). This is a significant result, 38% above the 20% "prior probability" of accurate c l a s s i f i - cation, and this i s well above the 25% mark recommended by Hair et al (1979). PLCO is the most important discriminating variable in this analysis, and accounts for 66.7% of the variance of a l l variables combined, as well as accounting for 95% of the variance of the f i r s t canonical discriminant function derived for this sample. Table 4 shows that early PRB's and BRF's are the most accurately class- i f i e d groups (75% and 84.6% respectively) followed by BPO's (66.7%) and 117 A c t u a l group // of cases P r e d i c t e d Group Membership EARLY MIDDLE LATE BRF BPO EARLY 180 135 25 3 2 15 75.0% 13.9% 1.7% 1.1% 18.3% MIDDLE 484 96 237 115 15 21 19.8% 49.0% 23.8% 13.1% 4.3% LATE 192 12 57 92 28 3 6.3% 29.7% 47.9% 14.6% 1.6% BRF 123 0 3 16 104 0 0.0% 2.4% 13.0% 84.6% 0.0% BPO 15 2 2 1 0 10 13.3% 13.3% 6.7% 0.0% 66.7% PERCENT OF GROUPED CASES CORRECTLY CLASSIFIED: 58. TABLE 4 . MDA C l a s s i f i c a t i o n Results of A l l Flakes A c t u a l group EARLY w i t h Platforms. N = 994) # of P r e d i c t e d Group Membership cases EARLY MIDDLE LATE BRF 0 0 83.3% 16.7% 0.0% 0.0% MIDDLE LATE 12 0 0.0% 4 0 0.0% 9 2 1 75.0% 16.7% 8.3% 0 3 1 0.0% 75.0% 25.0% BRF 0 2 1 3 0.0% 33.3% 15.7% 50.4„% PERCENT OF GROUPED CASES CORRECTLY CLASSIFIED: 71.43% TABLE 5. MDA C l a s s i f i c a t i o n Results of Obsidian PRB's 25% Random Sample (N = 28). 118 then middle stage PRB's (49%) and late stage PRB's (47.9%). When sorted into basalt and obsidian raw materials and sampled randomly at a 25% rate, essentially the same result i s achieved with basalt PRB's (56.72% overall), but obsidian PRB's are more accurately classified at 71.43% (Table 5). It should be noted here that BPO's were not adequately represented in the sampling to be worth testing in the obsidian sample, nor was the chert sample adequate. PLCO is the best discriminating variable in both these analyses, and as MDA is very prone to more accurate discrimination of small samples (basalt 25% PRB = 201; obsidian 25% PRB = 28; see Magne and Matson 1982), this difference in accuracy of the two analyses does not seem very important. Analysis of shatter, using only three groups (no BRF's or BPO's) and four variables (no PLCO or PLAN), and sampled at 10% rates, gave very similar results. In basalt shatter, overall correct discrimination was obtained in 54.24% of the flakes (N = 118). In the smaller obsidian sample (N = 33), 78.79% of the shatter were correctly classified. DOCO is the most important discriminating variable in both analyses. These analyses show that debitage can be assigned to the defined reduction groups with ca. 60% accuracy. However, i t is apparent that the results are not very robust, especially in prediction of middle and late stages. To finalize results, only that set of PRB's resulting from the reduction events of experienced knappers were selected, and randomly sampled at 50% (N = 222). In this analysis (Table 6), 119 Actual group # of cases Predicted Group Membership EARLY MIDDLE LATE BRF EARLY 73 68 93.2% 5 6.8% 0 0.0% 0 0.0% MIDDLE 73 6 8.3% 51 69.09% 12 16.4% 4 5.5% LATE 36 0 0.0% 11 30.6% 17 8 47.2% 22.2% BRF 40 0 1 6 13 0.0% 2.5% 15.0% 82.5% PERCENT OF GROUPED CASES CORRECTLY CLASSIFIED: 76.13% TABLE 6. MDA Classification Results, Debitage Produced by Experienced Knappers, 50% random Sample (N = 222). STAGE EARLY MIDDLE LATE 0-1 139 (87) 65 (90) 14 (41) 2 4 (38) 60 (39) 30 (18) 3 Or more 1 (19) 23 (19) 23 (9) 144 148 67 Chi-Square = 146.13, d.f. = 4, p = .001 TABLE 7. Chi-Square contingency table, PLCO by STAGE, PRB's Produced by Experienced Knappers (bracketed values are expected, rounded to nearest whole number). 120 PRB's were a c c u r a t e l y c l a s s i f i e d at an o v e r a l l r a t e of 76.13%, and again PLCO i s s i n g l e d out as the most important d i s c r i m i n - a t i n g v a r i a b l e . I t can be seen i n Table '6 that middle and l a t e stage PRB's are the l e a s t w e l l c l a s s i f i e d (69.09% and 47.2%), that these two c l a s s e s mix moderately among themselves, and that l a t e stage PRB's s l i g h t l y tend to be cla s s e d as BRF's (22.2%). To a s c e r t a i n the s i g n i f i c a n c e of the v a r i a b l e s PLCO and DOCO, chi-square t e s t s were undertaken. Using the chi-square s t a t i s t i c on a l l PRB's produced by experienced knappers (N = 359, no BRF's or BPO's) and c o l l a p s i n g c e l l s at both extremes of the PLCO range to meet the requirements of the t e s t , the d i s - t r i b u t i o n of PLCO by r e d u c t i o n stage i s s i g n i f i c a n t at p = .001 (Table 7). The same procedure on.shatter from experienced knap- pers, using the d i s t r i b u t i o n of DOCO by r e d u c t i o n stage, showed s i g n i f i c a n t d i f f e r e n c e s , a l s o at p = .001 (Table 8). By i n s p e c t i n g the chi-square t a b l e s , and the means and medians of PLCO and DOCO w i t h i n stages (Table 9, Figures"12 and 13) , i t can be seen that e a r l y PRB'§ can be cla s s e d as those having 0 or 1 d o r s a l s c a r s , middle PRB's have 2, and l a t e PRB's have 3 or more. E a r l y s h a t t e r have 0 or 1 d o r s a l s c a r s , middle g u t t e r have 2, and l a t e s h a t t e r have 3 or more. The l e s s e r d i s c r i m i n a t i n g power of the weight v a r i a b l e i s discussed below. In sum, the general problem of stage i d e n t i f i c a t i o n i s r e - so l v e d , and o r d i n a l c l a s s i f i c a t i o n of a l l types of debitage can i d e n t i f y general r e d u c t i o n stages, using p l a t f o r m and d o r s a l scar 121 STAGE EARLY MIDDLE LATE D 0-1 23 (14) 21 (.23) 2 (10 46 0' C 2 13 (.15) 26 (26) 12 (11) 51 0 3 or more 6 (14) 24 (23) 16 (10) 46 42 71 30 143 Chi-Square = 21.73, d.f. = 4, p = .001 TABLE 8. Chi-Square contingency table, DOCO by STAGE Shatter produced by Experienced Knappers, (Bracketted values are expected, rounded to nearest whole number) 50% random sample. 122 STAGE SHATTER PRB'S Weight (grams) Dorsal Scar Count Weight (grams) Platform Scar Count Mean 1.104 1.514 1.34 ,0.201 EA RL Y Median Standard Deviation 0.165 3.086 1.492 0.906 0.227 3.764 0.095 0.510 Mean 0.123 2.409 0.162 1.770 MI DD LE  Median Standard Deviation 0.039 0.532 2.265 1.446 0.069 0.482 1.650 0.874 Mean 0.77 3.320 0.090 2.194 Median 0.031 3.222 0.032 2.150 LA TE  Standard Deviation 0.327 1.498 0.417 0.839 CO Mean 0.693 3.427 P H Median 0.158. 3.379 m Standard Deviation . : „ 2.541 1.002 TABLE 9. Mean, Median and Standard Deviations of Weight, Plat- form and Dorsal Scar Counts, Debitage Produced by Ex- perienced Knappers, Broken Down by Stage of Reduction. 1 2 3 5 - i 4 i < * 3H C3 x 2H o UJ 3 I H 0-1 M f O . 23 STAGE E A R L Y 5 "I 4H 2 O 3 CC < </) 2 cr o < _ J o. 2H i H o-1 0.Q7T MIDDLE 0"."03 —t~ L A T E + 0.16 BRF / - r - F i g u r e 12. G r a p h o f m e a n , m e d i a n a n d s t a n d a r d d e v i a t i o n v a l u e s o f w e i g h t a n d p l a t f o r m s c a r c o u n t v a r i a b l e s , P R B ' s p r o d u c e d b y e x p e r i e n c e d k n a p p e r s . 124 5-1 4 4 co 2 < o 3 H i- x o 2H UJ 5 H O - 1 STAGE M 4 0.17 E A R L Y MIDDLE -0-03 L A T E 5-1 3 o tr < u co < CO tr o Q 3H Figure 13. Graph of mean, median and standard deviation values of weight and dorsal scar count variables, Shatter produced by experienced knappers. 125 counts, and recognition of bi f a c i a l and bipolar reduction techniques. The power of the experimental discriminations is apparently higher in the debitage produced by experienced knappers than in that produced by novices. HI: Weight as a Stage-discriminating variable: Negated Nowhere in the tests of HI does the weight of individual flakes appear to be a significant factor in identifying re- duction groups. In a l l the MDA analyses, weight was the third or fourth important variable, contributing less than 5% to overall variance, and achieving correlations on the order of 0.1 with dis- criminant function coefficients, while PLCO and DOCO contributed on the order of 90% to overall variance, and correlated with discriminant functions with about 0.9 correlations. This contradicts the results of the pilot study, which was in several ways less rigorous than the present one, especially in relying on professional judgement to evaluate variable significance., Thus; while i t is only logical that as tools are reduced, they w i l l become smaller, the same is not necessarily true of the debitage. Similar results are reported by Baker (1981) in an exper- imental analysis of cement block reduction. Furthermore, while other re- searchers have supported weight as a valid indicator of reduction stages in experimental situations (Burton 1980, Stahle and Dunn 1981), the control factor in this and the pilot- study of removing from the analyses a l l large flakes of debitage that would be suitable for shaping in an "ideal" technological-system, as far as I am aware, has 126 not been previously applied. Table 9. and Figures 12 and 13 depict the reason why platform- scar counts and dorsal scar counts are more accurate reflections of general manufacturing strategies than is weight of debitage. The figures also lend some insight to how the weight factor may be approached in future studies. First of a l l , i t is apparent that weight declines sharply from early to middle stages, and then f l a t - tens out to a nearly equal value for late stages, in both PRB and Shatter samples from the experienced knappers. Platform scar counts on PRB's, when plotted by stage (Figure 12), show an increase.from early to late that rises slightly to late, and sharply to biface re- duction. The shatter' flakes show a clearer progress in dorsal scar counts (Figure 13), sharper than do the PRB's curve. A secondary f a i l i n g of the weight factor is that the values of the central tendency measures are in 100 ths of grams. This i s unwieldy for macro-debitage analysis, and i s a clear indication of some data limitations, since weight in this study was recorded to 10 ths of grams only. This finding is also supportive of micro- debitage analysis, and more research along the lines of that under- taken by Fladmark (1982c) is required to determine the u t i l i t y and stage prediction capabilities of small debitage. H2: BRF's and BPO's as indicators of b i f a c i a l and bipolar reduction: Supported There i s ample evidence that BRp's and BPO's are excellent i n - dicators of b i f a c i a l and bipolar reduction, respectively (Table 2), 127 with the slight possibility that some bipolar events w i l l produce BRF's. These classes also seem to be quite objective classes in themselves, with BRF's accurately classified 84.6% of the time and BPO's 66.7% of the time among a l l PRB's (Table 4). It must be noted that these analyses provide only some indications of how BRF's and BPO's pattern across re- duction stages. However, when misclassified in the discriminant analyses, BRF's tend to be "late" and BPO's tend to be "early" (Tables 4 and 6). H3: Stage definition as independent of raw materials: Supported The differences observed between raw materials in stage varia- b i l i t y are not very great. There are d i f f i c u l t i e s in the experiment with the sample sizes of raw materials, especially with chert, for which the sample was considered too small for testing in any case but inclusion in the f i r s t MDA evaluation of HI. The slightly better dis- criminating power of obsidian in comparison to basalt is considered to be a sort of systematic error factor, due to greater f a c i l i t y in act- ually observing flake scars on the black glossy Anahim Lake region obsidian than on basalt. This factor can be evaluated archaeologically by testing for differences in assemblage complexity by raw materials. 4.8. Summary of Experimental Findings This experiment was designed to test ideas about using debitage to identify general stone tool manufacturing stages, and to develop units of measure that are technologically meaningful and reliable. Previous work relied on professional judgement and impressions gained from experience in l i t h i c replication. As in the identification of BRF's and BPO's, experience certainly plays a role in any complex 128 and s p e c i a l i z e d a n a l y s i s , but by d e v i s i n g an o b j e c t i v e way of c l a s s i f y i n g debitage types and v a r i a b l e s , t h i s r o l e can be g r e a t l y diminished. The most meaningful r e s u l t s of the ex- periment are that stone t o o l manufacturing stages can be acc- u r a t e l y reconstructed w i t h a minimum number of debitage types and v a r i a b l e s , and that the weight of i n d i v i d u a l f l a k e s i s un- s u i t a b l e f o r t h i s task, w h i l e f l a k e p l a t f o r m and d o r s a l scar counts appear to be much more appropriate. To apply the r e s u l t s of t h i s experiment to a r c h a e o l o g i c a l debitage, the debitage c l a s s i f i c a t i o n i n Figure 14 i s used. The c l a s s i f i c a t i o n groups a l l types of debitage i n t o r e d u c t i o n stages, f i r s t by s o r t i n g f l a k e s i n t o PRB's, Sha t t e r , BRF's and BPO's, hav- ing the l a s t two as i d e n t i f i e r s of d i s t i n c t kinds of r e d u c t i o n , and s o r t i n g the PRB's and Shatter i n t o e a r l y , middle or l a t e stages by t h e i r p l a t f o r m and d o r s a l scar counts. In l a t e r chapters, these groups are o f t e n lumped or pooled to provide g e n e r a l i z e d stages. In such cases,BPO's are pooled w i t h e a r l y Shatter arid PRB's, middle Shat- t e r and PRB's are grouped, and BRF's are added to l a t e Shatter and PRB's. The r e s u l t s of t h i s experiment were g e n e r a l l y p r e d i c t e d by John Speth, a pioneer i n c o n t r o l l e d l i t h i c experimental research, who com- mented t h a t : Further research i n t o the t e c h n o l o g i c a l aspects of f l a k e production should lead to a s i g n i f i c a n t r e - duction i n the t o t a l number of a t t r i b u t e s needed to q u a n t i f y t e c h n o l o g i c a l v a r i a b i l i t y , and to the r e - placement of dozens of a r b i t r a r i l y chosen and r e - dundant measurements p r e s e n t l y i n vogue w i t h con- s i d e r a b l y smaller numbers of a t t r i b u t e s c a r e f u l l y s e l e c t e d on the b a s i s of sound t h e o r e t i c a l p r i n - c i p a l s (Speth 1972: 57). P R B S H A T T E R Platform Scar Count Dorsal Scar Count S T A G E E A R L Y MIDDLE L A T E O- l 0 - 1 >3 2 3 DISTINCT REDUCTION T Y P E BIFACIAL BIPOLAR B R F BPO ro Figure 14. The experimental debitage c l a s s i f i c a t i o n , demonstrating f l a k e c h a r a c t e r i s t i c s required to s o r t debitage i n t o e a r l y , middle and l a t e r e d u c t i o n stages, and a l s o i n t o b i f a c i a l and b i p o l a r r e d u c t i o n c l a s s e s . 130 CHAPTER 5 THE ARCHAEOLOGICAL DATA BASE In this chapter the archaeological sources of data and the artifact classification scheme are described to provide background information for the following analytic chapter. Each of the 38 sites from the Eagle Lake, Mouth of the Chil- cotin, Lillooet and Hat Creek regions of the Interior Plateau (Figure 15) is described; then the artifact classification system is presented. The reduction stage classification of debitage developed in the previous chapter is used to measure the dominant stages of tool manufacture represented in the 38 assemblages. Several tool classes that are based primarily on the extent of retouch exhibited are defined, and tool attributes that were individually gathered are also described. The frequencies of artifacts are tabulated for each site, and photographs of the tools and cores are presented at the end of this chapter. 5.1. Site Descriptions This section provides descriptions of the locations where the assemblages under study were collected, including the size of the sites, features associated with the sites, the area with- in the sites that was collected or excavated, the number of tools, cores and debitage analysed in this study, radiocarbon dates i f Figure 15. Location of the four regions under study This f i g u r e i s keyed to Figure 3. 132 such are available, and some general locational information. The information was compiled from various sources including project reports (Matson et al 1979; Matson et a l 1980; Poko- tylo and Beirne 1978; Beirne and Pokotylo 1979; Stryd 1972) , graduate theses (Pokotylo 1978a; Stryd 1973; Ham 1975), B.C. Provincial archaeological site forms (Keddie 1972; others from Eagle Lake, Shuswap Settlement Patterns and Hat Creek Projects), and personal communications with the original collectors of the artifacts. An effort was made to use sites that were late pre- historic in age, or from the Kamloops Phase, although i t cannot be certain that a l l sites analysed here date to within the last 2000 years. Given the current poor state of culture history in the Interior Plateau, this i s a weakness of the present data, but does not significantly interfere with the purpose of this study: to examine assemblage var i a b i l i t y within and across several regions of the Plateau. The frequencies of tools, debitage and cores that are given in the following discussion may not match those reported in orig- inal reports or detailed analyses for three major reasons. The f i r s t of these is that only chipped stone tools, cores, debitage and hammerstones were analysed, and ground stone, bone and antler tools were not. The second reason for possible discrepancies is that the assemblages were completely re-classified for this study, and my tool classifications do not necessarily agree with those of previous researchers. In particular, this study distinguishes be- 133 tween complete and fragmentary tools, and also classifies many items as debitage that were previously classed as util i z e d flakes, when edge damage was not continuous. Furthermore, i t was apparent that bipolar cores were not well recognized in previous analyses. The third major reason is that this study only analyses debitage greater than 5 mm in size along their largest dimension. This was thought necessary to limit the amount of material that would be studied, to provide continuity with the experimental program, and to provide some control over screen size differences between projects and the size of material that is gathered by different persons in surface collection situations. The assemblages are referred to by the designations assigned by f i e l d investigators, and Borden site numbers are provided as well. In the case of most Eagle Lake, Mouth of the Chilcotin and Hat Creek sites, the identifiers used here refer to quadrats and sites within quadrats. For certain sites within these three pro- ject areas and for a l l Lillooet sites, Borden site numbers are used when the sites were known prior to project surveys. Maps showing individual site locations are found in figures 16, 17, 18, 19 and 20. The assemblages are discussed here simply in the order that they were f i r s t examined, and that order is maintained in most fur- ther tables. This practice helped to minimize the amount of editing that was required of the data, and is no great impedence to under- standing the analyses since site designations are quite arbitrary in any case. Site types, radiocarbon dates (uncorrected, uncalibrated), 134 site areas and general debitage tool frequencies are shown in Table 10. In a l l of the following pages, the kinds of sites from which l i t h i c assemblages were obtained are defined as follows: 1. Housepits: Lithic assemblages have been obtained from excavated housepit depressions. 2. Lithic scatters: These are surface scatters of stone artifacts, with no associated cultural depressions. Occ- asionally, very small areas of these have been test exca- vated to depths never exceeding 20 cm. 3. Lithic scatters with housepits: These are surface scatters only at sites that also have associated house depressions. 4. Lithic scatters with cachepits: These sites' assem- blages also occur in surface contexts, but with associated cachepits only. 5. Lithic scatters with fire-cracked rock: These are sur- face scatters of l i t h i c artifacts, with associated f i r e - related features, usually including fire-cracked rock and burnt mammal bone. One of these from Hat Creek (F8:l) is an actual roasting p it, with associated surface l i t h i c re- mains. 135 COLLECTION/ SITE SITE C14 AGE EXCAVATION CORES & TOTAL TYPE (B. P.) AREA (m ) TOOLS DEBITAGE ARTIFACTS 14:2 LS 168 2 11 3 16:1 LS 2750 5 32 37 19:1 LSCP 5000 56 1063 1119 22:1 LS 150 2 83 85 26:3 LSCP 9375 4 108 112 EA GL E I 32:1 LSHP 400 13 167 180 EA GL E I CR28 LSFCR 3575 5 34 39 EA GL E I CR64 LSFCR "2 0 42 42 EA GL E I CR40 LSFCR 100 6 117 123 CR73 HP 360 + 80 4 4 53 57 ElRw 4 LSHP 280 + 80 2 19 646 665 CR92 LSHP 860 + 80 240 46 1260 1306 EkRo 18 HP 1290 + 80 3 17 65 82 S3 M EkRo 31 HP 2 22 130 152 [L CO l EkRo 48 HP 870 + 60 [L CO l 1459 + 75 5 18 330 348 CH E CH ] 2:3 LSCP 770 + 65 1250 37 122 159 CH E CH ] 4:2 LS 300 35 954 989 CH E CH ] 4:5 LS 6000 16 340 356 4:1 LSHP 8750 24 122 146 o 5:1 LSHP 12500 25 85 110 9:1 LSHP 9750 12 141 153 9:2 LSHP 3750 13 151 164 8 12:6 LSHP 2500 12 24 36 EeRk 16 HP 1290 + 85 5 20 24 44 W o EeRl 41 HP c a . 150 16 29 23 52 EeRk 7 HP 920 + 80 36 116 2802 2918 EeRk 4:38 HP 2 20 218 238 • J EeRl 40 HP 395 + 80 18 76 1300 1376 G21:9 LS 2252 26 356 382 G23:l LS 284 4 323 327 G2:12 LS 244 8 259 267 G31:l LS 376 22 281 303 r CR EE l F 8:l LSFCR 2120 + 65 r CR EE l 2245 + 50 1676 53 641 694 r CR EE l F12:5 LS 84 6 346 352 <: EC J22:2 LS 20 2 12 14 J38:2 LSFCR 12 7 23 30 K2:l LS 33200 10 1142 1152 EeRj 1 HP 140 + 50 10 69 875 944 LS - L i t h i c scatter LSCP - L i t h i c scatter with HP - Excavated housepit cachepits LSHP - L i t h i c scatter with LSFCR - L i t h i c scatter with housepits firecracked rock TABLE 10. Summary data for the 38 assemblages under study. 136 F i g u r e 16. Eagle Lake r e g i o n s i t e s . Figure 17 j o i n s upper r i g h t . 137 Figure 17. Eagle Lake region site EIRw 4. Joins Figure 16 at bottom l e f t . 138 5.1.1. Eagle Lake S i t e s The 12 assemblages from the Eagle Lake region (Figures 16 and 17) that are analysed here were c o l l e c t e d during the 1979 season of the Eagle Lake p r o j e c t (Matson et^ a l . 1980). As i s discussed i n Chapter 3, t h i s p r o j e c t was designed to describe the settlement patterns and m a t e r i a l c u l t u r e of l a t e p r e h i s t o r i c C h i l c o t i n i n the area, to date t h e i r a r r i v a l and to compare the patterns to Mouth of the C h i l c o t i n and Hat Creek regions. For f u r t h e r d e s c r i p t i o n see Matson et a l . (1980). 1. 14:2 (EkSb 4) This s i t e i s a small (10.5 m X 16 m) l i t h i c s c a t t e r l o c - ated at the western end of Eagle Lake, 25 m n o r t h of the l a k e shore. Located at an e l e v a t i o n of 1190 m a . s . l . , the s i t e occurs i n grassland environment near discontinuous lodgepole pine and aspen f o r e s t near the l a k e shore. The s i t e was completely sur- face c o l l e c t e d , and the assemblage c o n s i s t s of two t o o l s and 11 pieces of debitage. 2. 16:1 (EkSb 5) This s i t e i s l o c a t e d at the west end of Eagle Lake, at an e l e v a t i o n of 1200 m a . s . l . , and at a d i s t a n c e of about 850 m north of the l a k e shore. An area measuring 75 m X 50 m i n a l a r g e open meadow was completely surface c o l l e c t e d , y i e l d i n g an assemblage of f i v e t o o l s , f i v e cores, and 27 debitage items. 139 3. 19:1 (EkSa 27) A l a r g e (200 m X 200 m) l i t h i c s c a t t e r w i t h a s s o c i a t e d c a c h e p i t s , rock c l u s t e r s and p o s s i b l e r o a s t i n g p i t s , t h i s s i t e i s s i t u a t e d on a low t e r r a c e of the C h i l k o River at 1160 m a . s . l . , about 2.5 km east of Eagle Lake. The s i t e was completely s u r - face c o l l e c t e d , and three a d j o i n i n g 1 m X 1 m u n i t s were exca- vated to 25 cm depth below surface. One of the excavation u n i t s contained an ash f e a t u r e 15 cm i n diameter w i t h extremely f r a g i l e c a l c i n e d bone fragments. This i s the second l a r g e s t assemblage from Eagle Lake, w i t h 56 t o o l s , f i v e l a r g e cores, 15 b i p o l a r cores and 1043 pieces of debitage. 4. 22:1 (EkSb 6) This s i t e i s located about 200 m from the n o r t h shore of Eagle Lake, at 1190 m a . s . l . i n an open grassland area near the northeast shore of the l a k e . The s i t e i s small i n s i z e (.15 m X 10 m) and was completely surface c o l l e c t e d , y i e l d i n g two t o o l s , three l a r g e cores, and 80 pieces of debitage. 5. 26:3 (EkSa 31) Located on a small e s k e r - l i k e f e a t u r e at the east end of Eagle Lake, t h i s s i t e i s a l i t h i c s c a t t e r measuring 125 m X 75 m i n area, at 1190 m a . s . l . , and occurs about 25 m. from the l a k e shore. Complete surface c o l l e c t i o n of the s i t e produced 3 t o o l s , 108 pieces of debitage, and one hammerstone. 140 6. 32:1 (EkSa 36) This is a unique s ite situated 50 m west of a small lake in lodgepole pine forest environment, and about 1 km east of Eagle Lake. The s i te contains a rectangular shallow house de- pression, two cachepits, a firecracked rock feature and a l i t h i c scatter. Altogether, the s i te area is about 40 m X 40 m. The s ite was tentatively identif ied as representative of an Athapaskan occupation location, on the basis of the large rectangular fea- ture, the presence of a contracting stem Kavik project i le point and a blue glass trade bead, among other features. The l i t h i c scatter was completely surface collected, yielding 162 flakes, five bipolar cores and 13 tools. 7. CR28 (EkSa 98) This s i te is a l i t h i c scatter at 1190 m a . s . l . located on a high bluff on the east side of the Chilko River, about 5 km south- southeast of Eagle Lake. Five tools and 34 flakes were collected within an area measuring 55 m X 65 m that also exhibited f i r e - cracked rock, burnt bone and a game t r a i l . 8. CR64 (EkSa 34) This s i te is located on the west side of the Chilko River, at 1070 m a . s . l . , about 200 m north of the r iver ' s edge, and approximately 10 km northeast of Eagle Lake. The s i te was re- vealed in a roadcut exhibiting bone and firecracked rock, and 141 may have o r i g i n a l l y extended over an estimated area of 50 m X 40 m. Two excavation u n i t s each measuring 1 m X 1 m were dug to depths of about 30 cm below s u r f a c e , and produced two l a r g e cores, one b i p o l a r core, and 39 pieces of debitage, but no t o o l s . S everal burnt fragments of l a r g e mammal long bone were al s o r e - covered from the u n i t s . 9. CR40 (EkSa 89) This s i t e i s on the east s i d e of the C h i l k o R i v e r , about 2.5 km east-northeast of Eagle Lake, and i s l o c a t e d about 100 m south of s i t e 19:1, at an e l e v a t i o n of 1130 m a . s . l . Complete surface c o l l e c t i o n of the l i t h i c s c a t t e r part of the s i t e pro- duced s i x t o o l s , and 117 f l a k e s . A l t o g e t h e r , the s i t e encom- passes an area measuring 120 m X 90 m. 10. CR73 (EkSa 35) This s i t e represents the only excavated housepit assemblage from the Eagle Lake r e g i o n at present. Located at 1080 m a . s . l . , on the east s i d e of the C h i l k o R i v e r , approximately 1 km south of the mouth of B r i t t a n y Creek, the housepit was p a r t i a l l y eroded by the C h i l k o R i v e r , r e v e a l i n g the s t r a t i g r a p h y of the depression. The s i t e i n c l u d e s a small c a c h e p i t , and i n a l l covers an area of 40 m X 20 m. Four excavation u n i t s 1 m X 1 m i n s i z e were dug i n the house depression, exposing roof f i l l m a t e r i a l , w e l l preserved burnt roof beams and a s i n g l e occupation l a y e r . Charcoal from one 142 of the beams was radiocarbon dated to 360 + 80 BP (SFU 15), and a dendrochronological date from the beam of AD 1561 w (outside very variable) was obtained as well. The assemblage from the site consists of four tools and 53 flakes, as well as two net sinkers, and fragments of incised slate and bone, and two small edge fragments of ground stone tools. Faunal remains included fis h and mammal bone that have not been identified to species. 11. EIRw 4 This site i s located on the north bank of the Chilko River, well outside the immediate area of the Eagle Lake region, but was studied as part of the Eagle Lake project. The site covers an extensive area (.about 750 m X 400 m) , and contains 169 house- pits, cachepits and possibly other kinds of depression features as well as firecracked rock and a light but extensive l i t h i c scatter. Three 1 m X 1 m excavation units were dug at the site, one on a high terrace and two others on a large slump bank next to the river. Only materials from the lower two units are analysed here, and these include 19.tools, five bipolar cores and 641 flakes. A radiocarbon date of 280 + 80 BP (SFU 16) was ob- tained from one of these units. 12. CR92 (EkSa 33) This site i s a large l i t h i c scatter (about 400 m X 100 m) located on the east side of the Chilko River, about 500 m south 143 of B r i t t a n y Creek. Some 20 cachepits are located nearby (re- corded as CR98, but considered here to be a part of CR92), and the materials analysed here were recovered from 250 1 m X 1 m surface u n i t s , and from two 1 m X 1 m excavation u n i t s . This i s the larges t assemblage from the Eagle Lake region studied, and i t includes 46 t o o l s , one large core, 15 b i p o l a r cores and 1244 pieces of debitage. A radiocarbon date at 860 + 80 BP (SFU 14) was obtained on charcoal removed from one of the exca- vation u n i t s . 5.1.2. Mouth of the C h i l c o t i n S i t e s The 11 s i t e s analysed from the region immediately south- west of the confluence of the C h i l c o t i n and Fraser r i v e r s are from c o l l e c t i o n s recovered by Matson, Ham and Bunyan (1979). In c e r t a i n cases d i f f e r e n t s i t e s recorded within survey quadrats bear i d e n t i c a l Borden-site numbers, because they had been pre- vi o u s l y recorded by Keddie (1972) i n a judgemental suvey of the area. The o r i g i n a l quadrat designations are retained here to f a c i l i t a t e comparisons with the findings of the Shuswap S e t t l e - ment Pattern project (Figure 18). 1. EkRo 18 This s i t e consists of 15 housepits and eight cachepits i n an area measuring approximately 300 m X 80 m. I t i s located at an elevation of 685 m a . s . l . , 1.5 km southwest of the C h i l c o t i n 144 Figure 18. Mouth of the C h i l c o t i n region s i t e s . 145 and Fraser Rivers confluence i n an open rangeland s e t t i n g . The assemblage stud i e d here c o n s t i t u t e s only those m a t e r i a l s recov- ered from three 1 m X 1 m excavation u n i t s placed i n a housepit d e p r e s s i o n , measuring 4.5 m i n diameter. The excavations r e - vealed no h o r i z o n t a l l y continuous occupation f l o o r , and a date of 1290 + 80 BP (Gak 5325) was obtained from a charcoal sample. The assemblage inc l u d e s 17 t o o l s and 65 f l a k e s . 2. EkRo 31 This s i t e i s located i n open gra s s l a n d , at 595 m a . s . l . , approximately 4 km south-southeast of the Mouth of the C h i l c o t i n R i v e r . The s i t e f eatures 11 housepits and four cachepits i n an area 225 m X 50 m. One of the housepits was t e s t excavated by means of two 1 m X 1 m u n i t s , y i e l d i n g the assemblage s t u d i e d N here, and no radiocarbon samples were processed. The assemblage c o n s i s t s of 22 t o o l s , one b i p o l a r core, and 129 pieces of debitage. 3. EkRo 48 This s i t e c o n s i s t s of seven housepits and eig h t cachepits at 655 m a . s . l . on an open t e r r a c e near to s i t e EkRo 18, approx- imately 1.5 km southwest of the Mouth of the C h i l c o t i n R i v e r . EkRo 48 was the most e x t e n s i v e l y excavated s i t e of the Shuswap Settlement P a t t e r n p r o j e c t i n 1974. F i v e 1 m X 1 m u n i t s were excavated i n one of the housepits, exposing a continuous f l o o r 3.5 m i n diameter. Charcoal from the f l o o r was radiocarbon dated 146 at 1459 + 75 BP (Gak 5327) and miscellaneous charcoal from the pit was dated at 870 + 60 BP (Gak 5326). The materials studied here include 17 tools, one hammerstone, eight bipolar cores and 322 pieces of debitage. 4. 2:3 (EkRo 87) Site 2:3 is located 1.75 km downstream from the Chilcotin- Fraser rivers' confluence, overlooking the Fraser River at 365 m a.s.l. The site consists of two cachepits and the surface l i t h i c assemblage studied here, within an area measuring 50 m X 25 m. The assemblage was obtained from an eroding bank area, and includes 36 tools, one hammerstone, four cores, four bipolar cores, and 114 flakes. A charcoal sample was removed from the eroding bank, and was dated at 770 + 65 (Gak 5324). This site was classed as a riverside site in the analyses by Matson et a l . (1979). 5. 4:2 (EkRo 31) Site 4:2 is one of three l i t h i c scatters from Quadrat 4 of the Shuswap Settlement Pattern project that are studied here. The scatter occurs here on a low rise near a creek bed, at 550 m a.s.l. The assemblage was collected by means of 12 grid units 25 m X 25 m in size, and was also studied by Bunyan (1974) in a moderately successful attempt to delimit technolog- i c a l l y distinct areas within the scatter area. 4:2 i s one of the chert debitage sites considered to be "pre-Kamloops" by 147 Matson et_ al. (1979) and the m a t e r i a l s analysed i n t h i s study i n c l u d e 35 t o o l s , one core, two b i p o l a r cores and 951 debitage items. 6. 4:5 (EkRo 31) This s i t e i s a surface l i t h i c s c a t t e r l o c a t e d approximately 75 m south of s i t e 4:2, and i s a l s o a chert debitage s i t e (Matson et a l . 1979). The assemblage was c o l l e c t e d from a low h i l l approx- imately 100 m X 60 m i n area, and c o n s i s t s of 16 t o o l s , two b i p o l a r cores and 338 f l a k e s . 7. 4:1 (EkRo 31) S i t e 4:1 i s a housepit s i t e (Matson et a l . 1979) that occurs near the two h i l l s where the 4:2 and 4:5 assemblages were c o l l e c t e d . Eight housepits and four cachepits are located i n the low area. The assemblage s t u d i e d here i n c l u d e s 24 t o o l s , two c o r e s , three b i p o l a r cores and 117 f l a k e s . 8. 5:1 (EkRo 5 and EkRo 10) This s i t e occurs i n open grassland approximately 3 km south of the confluence of the C h i l c o t i n and Eraser R i v e r s . The s i t e f eatures three housepits and s i x cachepits i n an area measuring approximately 125 m X 100 m, and i s p a r t i a l l y d i s - ected by a small g u l l y . The a r t i f a c t s analysed here i n c l u d e 24 t o o l s , one hammerstone, two b i p o l a r cores and 83 pieces of 148 debitage. Site 5:1 was also classed as a housepit site in Matson et_ al.'s (1979) f i n a l analysis 9. 9:1 (EkRo 31) Site 9:1 is another housepit site that occurs alongside site 4:1, next to a small and densely forested creek valley. Within an area of about 150 m X 60 m, the site contains three housepits and four cachepits, as well as the surface l i t h i c assemblage studied here, which includes 12 tools, seven bipolar cores and 134 flakes. 10. 9:2 (EkRo 30) This site is located 250 m north of site 9:1, in an open area approximately 75 m X 50 m in size, between a small road and a forested creek gully. The site may be continuous with site 4:6, a small site that is not addressed here. 9:2 features a single housepit arid a surface l i t h i c assemblage, however Keddie (1972) recorded EkRo 30 as exhibiting three housepits and 13 cachepits. The assemblage studied here consists of 13 tools, one core, three bipolar cores and 146 pieces of debitage. 9:2 was considered to be another chert debitage site by Matson jit al.. (1979) . 5.1.3. Lillooet Sites The five sites from the Lillooet region that are studied here occur on the east bank of the Fraser River near Gibbs Creek 149 and Kettlebrook Creek (Figure 19). These are a l l housepits excavated by Stryd at various times throughout his extended research in the area. These sites were chosen from the many that have been excavated, with Stryd's assistance, on the basis of relatively wide excavation areas and single component occu- pation horizons. Overall, these collections are those with which I am least familair, yet I chose to study them since this study required several housepit assemblages to contrast with the several l i t h i c scatter sites available from the other three regions. 1. EeRk 16 This i s a single housepit site that was excavated i n 1973. The excavations are not well described and the exact area exca- vated is not known. Apparently, a single occupation floor was present, and this was dated to 1290 + 85 BP (1-8060) (A. Stryd, personal communication). The assemblage from this site that is studied here includes 20 tools and 24 pieces of debitage. 2. EeRl 41 EeRl 41 is a single housepit site, with two exterior cache- pits, that is situated on the south bank of Gibbs Creek at an e l - evation of 360 m a.s.l. The housepit is 8.8 m X 7.2 m in area, and 90 cm deep, and the cachepits average 2.5 m in diameter. Ex- cavation of 16 m̂  revealed a single-house floor at 30 cm below 150 Figure 19. Lillooet region sites. 151 surface w i t h a cachepit i n s i d e the house depression. The occ- upation i s thought to be p r o t o h i s t o r i c and the assemblage here i n c l u d e s 29 t o o l s , one b i p o l a r core and 23 f l a k e s . S t r y d a l s o recovered a beaver i n c i s o r t o o l , an a n t l e r wedge, a b i r d bone bead and an a n t l e r h a f t h o l d i n g an i r o n t i p . Stryd (.1972) con- s i d e r s that EeRl 41 may have been a " s p e c i a l i z e d task s t r u c t u r e " because of high f a u n a l m a t e r i a l frequencies and low l i t h i c mat- e r i a l f r e q u e n i c e s . 3. EeRk 7 This s i t e f eatures three housepit depressions on the north 2 bank of Gibbs Creek. An area of 36 m was excavated i n Housepit #1, from which the a r t i f a c t s s t u d i e d here were obtained. A date of 920 + 90 BP (Gak 3284) was obtained from the base of the s i n g l e occupation f l o o r , that occurred 30 cm to 40 cm below surface. This i s the l a r g e s t assemblage from a l l the s i t e s i n t h i s study, w i t h 116 t o o l s , two cores,- e i g h t b i p o l a r cores, and 2792 pieces of debitage. 4. EeRk 4: 38 EeRk 4 i s a l a r g e s i t e w i t h 28 housepit depressions and num- erous c a c h e p i t s . The assemblage studied here i s from Feature #38, a depression approximately 2 m i n diameter i n t o which was placed a 1 m X 2 m excavation u n i t that was dug to a depth of 1.5 m. The presence of f i r e c r a c k e d rock i n upper l e v e l s and a greater 152 number of a r t i f a c t s i n lower l e v e l s lead Stryd (personal communication) to suspect that the depression may have served as a "refuse p i t " . Unfortunately, no d e t a i l e d d e s c r i p t i o n of the f e a t u r e i s a v a i l a b l e , and the context of the assemblage was forwarded to me only a f t e r the analyses to f o l l o w were completed. Thus, throughout the remainder of t h i s study, EeRk 4:38 i s t r e a t e d as a housepit assemblage. The a r t i f a c t s examined here c o n s i s t of 20 t o o l s , two b i p o l a r cores, and 216 f l a k e s . 5. EeRl 40 This i s a s i n g l e housepit s i t e w i t h a p i t f e a t u r e measuring 9.6 m X 8.9 m. S e v e r a l cachepits are nearby on the same f l a t 2 above Gibbs Creek. An area of 18 m was excavated i n the house- p i t , w i t h the occupation f l o o r o c c u r r i n g at 35 cm below s u r f a c e . A date of 395 + 80 BP (.1-9025) was obtained from the f l o o r dep- o s i t . A r t i f a c t s examined in'.this study i n c l u d e 75 t o o l s , one hammerstone, four b i p o l a r cores and 1296 f l a k e s . 5.1.4. Hat Creek S i t e s The 10 s i t e s s t u d i e d from the Hat Creek V a l l e y were a l l c o l - l e c t e d as part of the Hat Creek A r c h a e o l o g i c a l p r o j e c t (Figure 20). Most s i t e s recorded during the three year operation were l i t h i c s c a t t e r s , some of enormous s i z e , and the assemblages analysed here were chosen w i t h the a s s i s t a n c e of Dr. David Pokotylo on the b a s i s of probably l a t e p r e h i s t o r i c age, and manageable s i z e . Figure 20. Hat Creek r e g i o n s i t e s . 154 Six of these ten were also studied previously by Pokotylo (1978a). 1. G21:9 (EeRj 42) This site i s located on the north site of Anderson Creek in an open grassland area at 1035 m a.s.l. elevation. Lithic 2 materials were surface collected from an area of 2252 m . G21:9 was one of the assemblages studied by Pokotylo (1978a). In that study, the site was characterized as exhibiting a wide range of stone tool manufacturing processes as well as indications of intensive tool use. The assemblage studied here consists of 26 tools, and 359 flakes. 2. G23:l (EeRj 52) Surface artifacts from this site were collected ̂ rom a 284 2 m area on a high ridge at 1130 m a.s.l., 250 m south of Ambusten Creek. The site also features a rock cairn, the only such feature observed to date in the Hat Creek Valley. Pokotylo's (1978a) ana- lyses characterized this site as featuring debitage indicative of late stages of tool manufacture, and the tool assemblage was i n - ferred to have resulted from short term hunting and butchering ac t i v i t i e s . The materials included in this study are four tools, five cores, three bipolar cores and 315 flakes. 3. G2:12 (EeRj 20) 2 G2:12 i s a small (244 m ) l i t h i c scatter located about 350 m north of Finney Creek at 975 m a.s.l. The site contains no fea- 155 tures other than the surface l i t h i c assemblage, that consists of eight tools, one bipolar core and 258 pieces of debitage. In Pokotylo's (1978a) analyses, G2:12 was one of the wide- ranging manufacturing assemblages as revealed in the debitage, and the tools were inferred to have resulted from expedient, short-term usage. 4. G31:l (EeRj 64) This site i s situated on an open grassland bench west of Hat Creek at an elevation of 1005 m a.s.l. Materials were col- 2 lected from an area measuring 376 m . This site was also i n - cluded in Pokotylo's (1978a) dissertation, where i t was said to exhibit debitage resulting from late stage manufacturing pro- cesses, and the tools were part of the cluster of sites inferred to represent a wide range of intensive ac t i v i t i e s . The assem- blage consists of 22 tools, four cores, three bipolar cores and 274 flakes. 5. F8:l (EeRj 71) This site was one of the few recorded in the Hat Creek pro- ject forest stratum quadrats, although the majority of the site occurs on open ground. F8:l contains a l i t h i c scatter measuring 2 1676 m in areav and a circular cultural depression that is 5.6 m in diameter. The depression feature was test excavated by means of four 1 m X 1 m units, that revealed a main firecracked rock 156 i basin and a d d i t i o n a l smaller r o c k - l i n e d basins. Charcoal from the primary and secondary basins were dated at 2120 + 65 BP (S-1453) and 2245 + 50 BP (S-1642) r e s p e c t i v e l y . The d e p o s i t s contained f a u n a l bone m a t e r i a l of which one speciman was iden- t i f i e d as mule deer (Odocoileus hemionus), p l u s carbonic and p l a n t remains. Thus, the f e a t u r e appears to have served as a subsistence resource processing l o c a t i o n , and the presence of small basins w i t h i n i t i n d i c a t e that i t was used p o s s i b l y sev- e r a l times. The F 8 : l a r t i f a c t s from the surface c o l l e c t i o n were a l s o included i n Pokotylo's (1978a) study, where the assemblage was c h a r a c t e r i z e d as being s i m i l a r to that from G21:9. The F 8 : l deb- i t a g e appeared to r e s u l t from a wide range of r e d u c t i o n a c t i v i t i e s and the t o o l s were i n d i c a t i v e of a wide range of i n t e n s i v e uses. The m a t e r i a l s examined i n t h i s study i n c l u d e 53 t o o l s , s i x l a r g e cores, s i x b i p o l a r cores and 629 pieces of debitage. 6. F12:5 (EeRj 8) This s i t e a l s o occurred i n a f o r e s t quadrat of the 1976 Hat Creek P r o j e c t survey. Located at 850 m a . s . l . , 250 m west of Hat Creek, the surface a r t i f a c t s were c o l l e c t e d from an area of 2 84 m . Pokotylo's (1978a) a n a l y s i s found t h i s s i t e to e x h i b i t a wide range of debitage c h a r a c t e r i s t i c s , and a set of t o o l s f e a - t u r i n g low d i v e r s i t y , probably i n d i c a t i n g a b r i e f period of use. The assemblage analysed here in c l u d e s s i x t o o l s , one core, f i v e b i - p o l a r cores and 340 f l a k e s . 157 7. J22:2 (EeRj 176) This small (6 m X 11.5 m) l i t h i c scatter was found in the northwest end of the Hat Creek Valley in an area called the Houth Meadows. The site was not completely surface collected, and artifacts were only removed from a 2 m X 10 m wide transect placed through the center of the surface scatter, across a b u l l - dozed logging road. The site was collected in 1977, and thus i s not a part of Pokotylo's (1978a) study. The assemblage consists of two tools, and 12 pieces of debitage. 8. J38:2 (EeRj 180) J38:2 is also a small (7 m X 5 m) l i t h i c scatter in the northwest end of the Hat Creek Valley. The assemblage here was also collected from a transect (2 m X 6 m), rather than completely collected. Also found at this site was a surface feature contain- ing several small fragments of burnt and calcined bone, and f i r e - cracked rock. The artifacts studied here include seven tools, one core and 22 flakes. 9. K2:l (EeRj 90) The l i t h i c scatter covers a large area (400 m X 100 m) immediately east of Hat Creek, opposite i t s confluence with Anderson Creek. An area thought to represent 83% of the site surface was collected by means of 2 m X 2 m grid units. The site occurred along the edge of a survey quadrat and artifacts 158 are known to extend beyond i t s a r t i f i c i a l boundaries but no attempt was made to record them. The assemblage here consists of 10 tools, six bipolar cores arid 1136 pieces of debitage. 10. EeRj 1 EeRj 1 is a complex site approximately 200 m X 200 m in size that is located at the north end of the Hat Creek Valley, just in the bend where the creek turns to flow northeast to- wards the Bonaparte River. The site exhibits a large l i t h i c scatter approximately 100 m X 100 m in area, and 15 cultural depressions. Four of these depressions are thought to be house- pits, the other 11 are inferred to be roasting pits since they contain abundant charcoal and firecracked rock. The assemblage studied here was obtained from test excavations in one of the housepits (Culture Feature #10). Culture Feature #10 was tested by means of 10, 1 m X 1 m excavation units that were placed in a discontinuous line across the depression. Both traditional stone and bone artifacts and historic age goods were found, as well as many fragments of both f l o r a l and faunal materials. The depression also contained at least one hearth feature, from which a radiocarbon date of 140 + 50 BP (S-1582) was derived. Thus, by a l l evidence the depression feature appears to have been occupied during the early historic period. The l i t h i c artifacts analysed in this study include 69 tools, one core, four bipolar cores and 870 flakes. This is? the 159 only excavated housepit assemblage from the Hat Creek Valley. 5.2 Artifact Classification The data base of the archaeological component of this study consists of 15,566 chipped stone artifacts, of which 861 are tools, and 14,705 are flake and core debitage. This section provides descriptions of the assemblage classification system, and provides the basic frequency data for the analyses to follow. For a l l 38 assemblages, as required, the artifact categories were maintained across the five raw material classes: vitreous basalt, granular basalt, obsidian, chert/chalcedony and quartzite/other. In addition to artifact type and raw material, a l l tools were described by eight continuous and ordinal variables. Gathering the data took the greater part of about four months of straight laboratory time, and would have been lessened by per- haps no more than 25% if. a 10 mm, rather than 5 mm debitage size cut-off had been applied. Debitage collections that were not i n - dividually catalogued and wrapped were much faster to tabulate; however, recent damage was noted in assemblages that had been ex- cavated on the order of 10 years ago. Assemblages that have a cumulative history of archaeological collecting also posed some frustrating but resolveable problems by exhibiting changing Borden numbers, altering cataloguing, removed and altered cataloguing, and several means of storage. Data were written onto 80-column by 28 row blank forms onto which appropriate categories had been added. 160 These were keypunched, and were stored as cards, and as disk f i l e s , at the UBC Computing Centre. 5.2.1. Debitage (N = 14705) The system of general stage classification for debitage developed in Chapter 4 is applied here, yielding eight classes for flakes and two for cores. To be sorted in early, middle or late reduction stages, flakes are sorted into PRB and Shatter categories. PRB's with cortical or plain platforms are early, those with two platform scare are middle stage, those with three or more platform scars are late stage, and those with three or more platform scars on acute angled platforms are BRF's, or b i - face reduction flakes. Shatter with cortical dorsal faces or with only plain dorsal faces are early stage, Shatter with two dorsal scars are middle, and those with three or more dorsal scars are counted as late. Bipolar reduction flakes (BPO's) are con- sidered Shatter (as opposed to PRB's), because platforms are crushed, and include those flakes with evidence of simultaneous, opposing percussion and at least one dorsal face platform area that exhibits irregular hinge and step scarring. As is discussed in the experimental study, PRB's with platforms less than 2 mm wide were often d i f f i c u l t to code reliably and were often coded as Shatter and stage-evaluated by their dorsal scar counts. Cores and bipolar cores are also debitage. By definition these items bear no evidence of.use, or hafting retouch, 161 and have flake scars adequate to have yielded useful blanks. Hand-held cores (CORES) usually bear cortex, one or two per- cussion planes, minimal platform preparation and no b i f a c i a l flaking. Bipolar cores (BPCO) are pieces with evidence of sim- ultaneous percussion, with full-length scars, and extensively battered platforms. Pieces esquillees (PEEQ) are addressed as a tool class, and the contentions bearing on their identity are discussed below. Of the 14,705 pieces of debitage, the PRB and Shatter flake frequencies combined yield 5217 early stage items, 4991 middle stage, and 3325 late stage. BRF's total'595 for the 38 assemb- lages, and there are a total of 413 BPO's. A l l together, 164 cores were examined, of which 120 are BPCO's and 44 are CORES. Tables 11 and 12 show the frequency distribution for the debitage of each assemblage, and the percentage of the frequency categories per assemblage. 5.2.2. Tool Classification (N = 861) Tools are analysed in two manners, each designed to reveal different kinds of trends in implement occurrence, and assemblage complexity. The typological classification of tools combines attributes of retouch and u t i l i z a t i o n extent ( f a c i a l , marginal, utilized) with attributes of shape and occasionally, plausible function (e.g. projectile point, endscraper), and size. In and of themselves most classes do not yield much "functional" infor- 162 PRB Shatter Cores Sit e E M L BR E M L BP BC CO Total 14:2 16:1 19:1 110- . . 107.. 107 273. • • 275. • 115 22:1 20. • . . . o . . . . . 0. . . . . 0 . 36. . . . 21. . . 3. . . . . , 0 . . . . . o. • . . . 3 . . • • • 83- • • 26:3 6.. . . 10.. .23.. ... 8 . 15. .. 18. ..28.. . . . o. • . . . 0- • • • 0- • •  -108- • • 32:1 9.. .. 13.. . . 10. . ..4 . 39. ... 37. ..34.. .. 16.. . . . 5. . .. 0- • • • 167- • • CR28 CR64 4.. .. . 2.. .. .3.. ... 0 . 12. 7. .. 11.. ... 0. .. . . 1.. . . 2- • . . . 42- • • CR40 6. . ..17.. .. 13.. . . 15. 13. .. 24. .. 29.. . .. 0... . . o . • . . . 0 . • . . 1 1 7 . . . CR73 ... 8.. EIRw 4 55.. .. 53.. . . 52.. . 32 117. .. 161. . 164 .. . 7.. . CR92 135.. .. 83. . .. 57.. ..39 418. .. 256. . 210. .  .46... . 15. • ... 1.. • 1260- • • EkRo J8 EkRo 31 ..19.. . .. 7.. 8 EkRo 48 33.. ..26.. ... 5. . .. 13. . 94. ...81. ..47.. ..23... .. 8.. . . . 0 . . • • 330- • • 2:3 4:2 45.. ..44.. .. 60.. .. 22. 259. . . 311. 192 .. 18... 4:5 20.. ..18.. .. 20.. ... 3 106. ..110. . 56 ...5... 4:1 7, , . . . 6 . . 5: 1 4. . . . .3. . 9:1 8, , ...6.. . .. 0. 31 9:2 10. . ..20.. ..13.. ... 4. .39. ...35. ..25.. . ..1... .. 3-- .. 1.. . . 1 5 1 . . . 12:6 . . . 2. . 4 EeRk 16 1 . . . 2. . . . .4. . EeRl 41 4 ...5.. ...3.. . . . 1. 0 4. EeRk 7 467. . .241.. ..83.. . . 22. 918. . .683. .279 • 2802. • • EeRk 4:38 26 . .25.. . . 17. . . .8. 48 ...57. EeRl 40 138.. .132.. ..68.. ..51. 273. ..408. .194.. ..32... • 1300- • • 21:9 18.. ..54.. ..41.. . . 12. 49. .. 109 ..73.. . . . 0. . . 0. . 23:1 32.. .. 16.. ..10.. ..16. .85. . . i o o ..56.. . . .0... . 3. . . . . 5 . . 2: 12 11 ..37.. ..52.. . . .9. 71 31:1 27. . .. 11.. ..11.. . . 15. 72. . . .64. .67.. . . . 7 . . . .. 3.. . . 4. . . 281. .. F8:1 62. . . . 7 7 . . ..64.. . .55. .63. . .140. .163 . . . 5 . . . . . 6.. . . 6.. 641. .. F12 : 5 52. . ..21.. . . . 2. . . . .5. 114 ...95. ..46.. . . . 5 . . . . J22:2 2 . . . 1. . . . .0. . . . 1. 5 J38:2 0 . . . 2. . . . .0. . . . 1. 3 9. ...7.. .. .0. .. K2: 1 68. . ..78.. ..45.. .133. 229. . .325. .231... . .27. . . ..£> . . . 0 . . . 1142. .. EeRj 1 119. . . . 77. . .55.. . .89. 201. . .147. .151.. -.31. . . . 4 . . . 1. . . . 875. . . TOTALS 1557 1250 871 595 3660 3741 2454 413 120 44 E = Early BP = Bipolar Flakes M = Middle BC = Bipolar Cores L = Late CO = Cores BR = B i f a c i a l Reduction TABLE 11. Assemblage debitage classes, raw counts, a l l raw materials. 163 PRB SHATTER CORES S i t e E M L BR E M L BP BC CO 14:2 18.18 9.09 36.36 0 9.09 9.09 18.18. 0 0 . 0 16:1 35.48 0 0 0 35.48 12.90 3.23 & 0 12.90 19:1 10.35 10 07 10 .07 28 25.68 25.87 10.82 4 .99 1.41 .47 22:1 24.10 0 0 0 43.37 25.30 3.61 0 0 3.61 26:3 5.56 .9.26 21.30 7.41 13.89 16.67 25.93 0 0 0 32:1 5.39 7.78 5.99 2.40 23.35 22.16 20.36 9 .58 2.00 0 CR28 8.82 5.88 11.76 5.88 20.59 11.76 35.29 0 0 0 CR64 9.52 4.76 7.14 0 28.57 16.67 26.19 0 2.38 4.76 CR40 5.13 14.53 11.11 12.82 11.11 20.51 24.79 0 0 0 CR73 7.55 15.09 16.98 9.43 16.98 16.98 16.98 0 0 0 EIRw, 4 8.51 •8.20 8.05 4.95 18.11 24.92 25.39 1 .08 .77 0 CR92 10.71 6.59 4.52 3.09 33.17 20.32 16.67 3 .65 1.19 .08 EkRo 18 24.62 7.69 9.23 13.85 9.23 9.23 24.62 1 .54 0 0 EkRo 31 7.69 14.62 5.38 6.15 11.54 18.46 30.00 5 .38 .77 0 EkRo 48 10.00 7.88 1.52 3.94 28.48 24.55 14.24 6 .97 2.42 0 2:3 4.92 8.20 6.56 6.56 13.93 19.67 25.41 8 .20 3.28 3.28 4:2 4.72 4.61 6.29 2.31 27.15 32.60 20.13 1 89 .21 .10 4:5 5.88 5.29 5.88 .88 31.18 32.35 16.47 1 47 .59 0 4:1 5.73 4.10 4.92 1.64 23.77 28.52 16.39 82 2.46 1.64 5:1 4.71 11.76 3.53 0 22.35 34.12 20.00 1 18 2.35 0 9:1 5.67 11.35. 4.26 0 21.99 32.62 13.48 5 67 4.96 0 9:2 6.67 13.33 8.67 2.67 26.00 23.33 16.67 67 2.00 .67 12:6 25.00 8.33 0 0 16.67 25.00 0 8 33 16.67 0 EeRk 16 4.17 8.33 16.67 0 12.50 12.50 45.83 0 0 0 EeRl 41 17.39 21.74 13.04 4.35 0 17.39 21.74 0 4.35 0 EeRk 7 16.67 8.60 2.96 .79 32.76 24.38 9.96 3 53 .29 .07 EeRk 4:38 11.93 11.47 7.80 3.67 22.02 26.15 12.39 3. 67 .92 0 EeRl 40 10.62 10.15 5.23 3.92 21.00 31.38 14.92 2. 46 .31 0 21:9 5.06 15.17 11.52 3.37 13.76 30.62 20.51 0 0 0 23:1 9.91 4.95 3.10 4.05 26.32 30.96 17.33 0 .93 1.55 2:12 4.25 14.29 20.08 3.47 10.42 22.78 24.32 0 .39 0 31:1 9.61 3.91 3.91 5.34 25.62 22.78 23.84 2. 49 1.07 1.42 F8; 1 9.67 12.01 9.98 8.58 9.83 21.84 24.43 . 78 .94 .94 F12:5 15.03 6.07 .58 1.45 32.95 27.46 13.29 1. 45 1.45 .29 J22:2 16.67 8.33 0 8.33 41.67 16.67 8.33 0 0 0 J38:2 0 8.70 0 4.35 13.04 39.13 30.43 0 0 4.35 K2:l 5.95 6.83 3.94 11.64 20.25 28.46 20.32 2. 36 .53 0 EeRj - I 13.60 8.80 6.29 10.17 22.97 16.80 17.26 3. 54 .46 .11 E = Early M = Middle L = Late BR = B i f a c i a l Reduction BP = Bipolar Flakes BC = Bipolar Cores CO = Cores TABLE 12. Assemblage Debitage Classes, Percent by Count, A l l Raw Materials -164 mation, (Table 13), but such w i l l be attempted by searching for co-occurrence of types. Fragment type was also recorded for the tools, but in the analyses to follow only a distinction between complete and fragmentary tool -classes is maintained. The 23 tool class frequencies across the 38 sites are shown in Table 14. Photographs of the tools and cores, minus some uti l i s e d flakes, appear in Figures 21 to 69. 1. Lanceolate bifaces (LANC) and fragments (LABF) Complex bifaces with straight, or slightly curved edges, and extensive f a c i a l flaking (>5 mm from edges). - 2. Large bifaces (LABC) and fragments (LABF) These are b i f a c i a l tools in assemblages that are markedly larger than other tools, or i f they were complete, would be. There is not a s t r i c t limit imposed here, but objects on the order of 10 cm in any dimension, or fragments suggesting such a size are classed as large. 3. Bifaces (BIFC) and fragments (BIFF) These are items with flaking on two adjoining faces that extend over 5 mm from the edge. 4. Bimarginal tools (BIMC) and fragments (BIMF) These have adjoining—face retouch that extends between 5 mm and 2 mm from the edge, regardless of the actual number of edges bearing marginal retouch on both faces. 165 1. Lanceolate b i f a c e 2. Large b i f a c e 3. B i f a c i a l retouch t o o l 4. Bimarginal retouch t o o l 5. Large u n i f a c e 6. U n i f a c i a l retouch t o o l 7. Unimarginal retouch t o o l 8. U t i l i z e d f l a k e 9. P r o j e c t i l e p o i n t 10. G r a v e r / d r i l l 11. Endscraper 12. Piece e s q u i l l e e 13. S p a l l t o o l 14. Core t o o l 15. Hammerstone 16. U t i l i z e d b i f a c i a l r e d u c t i o n f l a k e NOTE: F a c i a l retouch i s greater than 5 mm; marginal i s between 2 mm and 5 mm, and u t i l i z e d i s l e s s than 2 mm lengths of f l a k e scars perpendicular to the edge. TABLE 13. Tool morphology c l a s s e s 166 TOOL TYPE u fa C J fa CJ fa o fa 55 CJ fa C J fa rJ C J fa OS C/3 Cf o o s EH rH CO oa fa fa s a fa fa H O o Q Q w EH H S u < < < M r-1 M M 55 s EH ai oi erf 55 fa PH O <3 Oi o Site , J i J • J ca 03 ca CQ D 3 CU W OH C/} c j rc CO EH 14:2 1 1 2 16: 1 1 1 1 2 5 19:1 2 1 2 11 2 1 4 5 4 20 1 1 2 1 56 22:1 1 1 2 26:3 1 2 1 4 32:1 1 3 1 1 1 4 2 13 CR28 2 1 1 1 5 CR64 0 CR40 1 1 2 2 6 CR73 1 1 1 1 4 EIRw 4 1 5 2 4 4 2 1 19 CR92 1 1 15 2 1 2 1 4 2 2 8 1 1 3 2 46 EkRo 18 1 1 12 3 17 EkRo 31 1 1 3 1 4 2 4 1 1 4 22 EkRo 48. 1 1 1 2 1 1 6 2 1 1 1 18 2:3 1 2 2 6 1 .1 1 1 7 6 3 5 1 37 4:2 • 1 2 7 2 1 4 9 5 2 1 1 35 4:5 2 8 6 16 4:1 1 2 3 2 2 10 1 1 2 24 5:1 1 2 3 2 2 3 1 2 3 4 1 1 25 9:1 1 1 2 1 1 1 1 1 3 12 9:2 1 3 2 4 1 1 1 13 12:6 1 2 5 2 1 1 12 EeRk 16 2 1 1 1 8 2 1 1 3 20 EeRl 41 1 1 5 2 1 14 2 1 1 29 EeRk 7 2 2 7 9 8 1 1 2 63 4 4 2 4 3 1 1 2 116 EeRk 4:38 1 3 5 2 8 1 1 20 EeRl 40 1 2 2 6 2 2 4 42 3 6 3 2 1 76 G21:9 1 2 2 1 8 3 1 2 2 7 1 26 G23:1 2 1 1 4 G2:12 1 2 1 1 2 1 8 G 3 i : l 3 4 5 7 3 22 F8:l 6 37 2 3 2 2 1 53 F12:5 1 1 2 1 1 6 J22:2 1 1 2 J38:2 1 2 2 1 1 7 K2: 1 1 5 1 2 1 10 EeRj 1 1 1 4 13 1 1 4 4 4 2 24 2 2 2 4 69 TOTALS rH CN rH 1 CM .- rH m 00 o r o o r o - i m oo m T -a- 00 OS rH •H r-l rH ' IT! VD r-l m r o rH CM m r o 00 i rH i - i CM rH 1 TABLE 14 . Tool type frequencies by s i t e 167 5. Large unifaces (LAUN) This i s an uncommon tool class, consisting of only two items that are large and exhibit unifacial flaking. 6. Unifaces (UNFC) and fragments (UNFF) These are items with unifacial retouch that extends greater than 5 mm from the edge. 7. Unimarginal tools (UNMC) and fragments (UNMF) These are items with retouch between 5 mm and 2 mm fromithe edge, on one face only, regardless of the number of edges bearing marginal retouch. 8. Utilized flakes without retouch damage that extends over 2 mm from the edge, regardless- of number of damaged faces. Continuity along edges must be maintained for the extent of the damage. 9. Projectile points (PROC) and fragments (PROF) These are projectile points, regardless of type, and side- notched, corner-notched and stemmed points are included in the c o l - lections, although types of points do not form analytic units. 10. Gravers/drills (GRDR) These are items with deliberately retouched projections, and not happenstance durable points. 168 11. Endscrapers (ENDS) These are items that have been retouched i n t o rounded ends, u s u a l l y u n i f a c i a l l y , and at times only e x h i b i t marginal retouch. 12. Pieces e s q u i l l e e s (PEEQ) These are items that have b i p o l a r b a t t e r i n g , where s i n g l e f l a k e scars do not extend across the e n t i r e faces of the a r t i f a c t . Pieces e s q u i l l e e s o f t e n e x h i b i t b i p o l a r f l a k i n g from perpendicular axes, w i t h four edges being about equal i n l e n g t h , and others have s p l i t s on l a t e r a l margins the l e n g t h of the item. These r i g h t - angled s p l i t s are not considered to be scars r e s u l t i n g from the detachment of u s e f u l blanks. Pieces e s q u i l l e e s and b i p o l a r cores are a t o p i c of l a s t i n g debate i n l i t h i c technology (Hayden 1980; S o l l b e r g e r and P a t t e r s o n 1976; B i n f o r d and Quimby 1963). I w i l l not add to the considerable d i s c u s s i o n , but i n d i c a t e that b i p o l a r r e d u c t i o n i s and was a con- t r o l l a b l e technique f o r f r a c t u r i n g stone, and that pieces esqu i l l e e s owe t h e i r form to some u t i l i z a t i o n technique, but that has eluded a r c h a e o l o g i s t s to date. 13. S p a l l Tools (SPTO) These are l a r g e f l a k e s , u s u a l l y obtained from granular b a s a l t s and other dense igneous r o c k s , that bear retouch i n v a r i o u s ways to provide a h a f t end and a scraping end. The ends i n f a c t are r a r e l y both worked, and o f t e n retouch i n d i c a t e s d e l i b e r a t e b l u n t i n g of the 169 scraping end. Sev e r a l of the s p a l l t o o l s studied here have con- s i d e r a b l e use p o l i s h , both a c t u a l rounding of the stone and apparent d e p o s i t i o n of organic m a t e r i a l s (see Ham 1975: 153 - 156). 14. Core Tools (COTO) These are l a r g e items, u s u a l l y w i t h much o r i g i n a l surface of a cobble present, fashioned i n a rough manner, and e x h i b i t i n g l e s s e r retouch that straightened edges, or that r e s u l t e d from heavy use. 15. Hammerstones (HAMM) These are not f l a k e d stone p i e c e s , but pebbles and small cobbles of dense rock, that bear b a t t e r i n g on one or both ends. Hammerstones were considered frequent enough to add as a p o t e n t i a l c l u e to l i t h i c t e c h n o l o g i c a l processes by patterns of a s s o c i a t i o n w i t h other types, but l a c k i n g chipped stone a t t r i b u t e s themselves, are not included i n a l l analyses. 16. U t i l i z e d B i f a c e Reduction f l a k e s (BRUT) These are BRF f l a k e s as recognized i n the debitage c l a s s i f i - c a t i o n , that e x h i b i t u t i l i z a t i o n or marginal retouch, on f l a k e edges apart from the pl a t f o r m . 170 Figure 21 a: b: 14:2 tools piece esquillee uniface 171 4 & M c d e « g h i j k l m n o p q r s t u v w x y z a o c ^ ^ ^ ^ m n o p q u ' v w x 1 a Figure 2 3 a,b c d e,f g-c' d'-q' r'-v' w'-a'1 19:1 t o o l s l a n c e o l a t e b i f a c e s g r a v e r / d r i l l piece e s q u i l l e e bimarginals p r o j e c t i l e p o i n t s b i f a c e s u n i f aces u t i l i z e d f l a k e s * o b c d e f 9 h Figure 2 4 a-o p.q 19:1 t o o l s and cores b i p o l a r cores s p a l l t o o l s core t o o l 172 173 * Figure 27 26:3 tools l B a: lanceolate biface b,c: unimarginals 174 * i t 1 Figure 29 a,b c d e CR28 t o o l s l a n c e o l a t e b i f a c e s b i f a c e p r o j e c t i l e p o i n t u n i f a c e Figure 30 CR64 cores a: b i p o l a r core b,c: cores CR40 t o o l s l a n c e o l a t e b i f a c e b i f a c e s u n i faces 175 Figure 32 CR73 t o o l s a,b: p r o j e c t i l e p o i n t s c: b i f a c e d: u t i l i z e d f l a k e M e f 9 h a b c d y * I I M * | i j k I m n • - t f P q * • « • Figure 33 EIRw t o o l s and cores a-f g,h i - n o p-s t-x b i f a c e s u n i f aces p r o j e c t i l e p o i n t s g r a v e r / d r i l l u t i l i z e d f l a k e s b i p o l a r cores w 176 Wm c d e f g h J •; i t % * q s I U v w x y z • b e t f ' I II ^ h' j m n 4 A M * * o p q r t » # » l t t l s' t' u" v ' w' x y z Figure 34 CR92 tools and cores k 1 m-o p-f g'-i j ' - l m' -p q'-r s'-g* projectile points graver/drill endscraper pieces esquillees bifaces bimarginals unifaces unimarginals utilized flakes bipolar cores 177 a b e d > » » * • I f g h I * f * * 4 Figure 36 EkRo 18 tools a b c ,d e-n biface uniface utilized BRF's utilized flakes a b c d e f 4 4 # 4 t • n o Figure 37 EkRo 31 tools and core a: biface b--e: unifaces f: unimarginal g--1: projectile points m--p: utilized flakes q: piece esquillee r: graver d r i l l s: bipolar core t: spall tool 178 a b b i j k I q r » t u v a b 4 * * «M f n w r 4 #1 U V t I q r s I . b c w x y z 0 Figure 38 EkRo 48 tools and cores a,b: bifaces c,d: projectile points e,f: unifaces g: unimarginal h-n: utilized flakes o: spall tool p: piece esquillee q-v: bipolar cores Figure 39 2:3 tools and cores a-m n-s t u-v w-c' bifaces projectile points uniface unimarginal utilized flakes bipolar cores Figure 40 2:3 tools a-d: spall tools e: core tool Figure 41 2:3 tools and cores a-d: cores e: core tool 180 9 h m n o ^ 4 1 | <%j S t U V x y z a' b' c ' d ' e' Figure 42 4:2 tools and cores k-o p-r s-v w-e f g'-h i j k bifaces p r o j e c t i l e points unifaces unimarginals u t i l i z e d flakes u t i l i z e d BRF bipolar cores piece e s q u i l l e e core s p a l l t o o l a b c d I f I J * m n o Figure 43 4:5 tools and cores a,b c-g h-o p,q r bifaces p r o j e c t i l e points u t i l i z e d flakes bipolar cores g r a v e r / d r i l l 181 #•4 c i g h i j k I m n o P a r s t u v K • • • t i 4 I m I I B | 9 n o p Figure 44 4:1 tools and cores a,b c d,e f-k 1 ,m n-w x-z a\b' Figure 45 a-f g,h i-k l,m n-p q,r s,t u-w X bifaces projectile point ut i l i z e d BRF's unifaces unimarginals uti l i z e d flakes bipolar cores cores 5:1 tools and cores bifaces bimarginals pieces esquillees projectile points utilized flakes bipolar cores unifaces spall tools core tool f It 182 Figure 46 9:1 tools and cores a: lanceolate biface b-d: bifaces e: bimarginal f,g: unifaces h: unimarginal i : u t i l i z e d flake j-1: projectile points m-s: bipolar cores m n o 4 a b e d Figure 47 9:2 tools and cores a-d bifaces e,f unifaces g-j utilized flakes k utilized BRF 1-n bipolar cores o core P graver/drill q core tool 183 V • « t * tr f g i Figure 48 12:6 t o o l s and cores a-h i 1 m-p b i f a c e s piece e s q u i l l e e u n i f a c e s u t i l i z e d f l a k e b i p o l a r cores 184 e 9 h * h t Figure 49 EeRk 16 t o o l s a-c d-f h i-p q r - t b i f a c e s p r o j e c t i l e p o i n t s u n i f a c e u t i l i z e d f l a k e s s p a l l t o o l u t i l i z e d BRF's b e d e f g i J I m ¥$m if o p q r Figure 50 EeRl 41 t o o l s and cores a: l a r g e b i f a c e b-g: b i f a c e s h: bimarginal i , j : p r o j e c t i l e p o ints k: unimarginal l,m: unifaces n-r: u t i l i z e d f l a k e s s: endscraper t : g r a v e r / d r i l l u: b i p o l a r core 185 Mi*/ e f g h i ~ 1 m n u v w x y i o b § « 4 # b e d 4 4 • ** f g h i Mr » 1 j k I 14 Figure 51 EeRk 7 t o o l s a-d e-t u-b' l a r g e b i f a c e s b i f a c e s p r o j e c t i l e p o i n t s Figure 52 EeRk 7 t o o l s a-1 j - l m-p q-t unifaces unimarginals u t i l i z e d f l a k e s pieces e s q u i l l e e s 186 b e d Figure 53 EeRk 7 tools and cores a spall tool b-e endscrapers f-m bipolar cores n-r utilized BRF's s,t graver/drill u,v utilized BRF's w,x cores r J k t m »• m n o p VI Figure 54 EeRk 4:38 tools and cores a-h: bifaces i : projectile point j-k: unifaces 1-p: utilized flakes q: utilized BRF's r,s: bipolar cores 187 h i j k I m n o p q r s t u v w x a b b c h i m n Figure 55 EeRl 40 t o o l s a-c d-k l - o p-x y-b' l a r g e b i f a c e s b i f a c e s bimarginals p r o j e c t i l e p o i n t s unimarginals Figure 56 EeRl 40 t o o l s and cores a - i j , k 1-n o-r u t i l i z e d f l a k e s s p a l l t o o l s g r a v e r / d r i l l s b i p o l a r cores 188 1# u v VI f i Figure 57 G21:9 t o o l s a - i j - l m n o-q r,s t-z b i f a c e s bimarginals p r o j e c t i l e point l a n c e o l a t e b i f a c e u nifaces unimarginals u t i l i z e d f l a k e s Figure 58 G23:l t o o l s and cores a,b c,d e - i j - l b i f a c e s u n i faces cores b i p o l a r cores 189 > M d e f 9 h Figure 59 a-c d,e f g h G2:12 t o o l s and cores b i f a c e s p r o j e c t i l e p o i n t s endscraper u n i f a c e u t i l i z e d f l a k e b i p o l a r core c b W c d e f g h 1 I — k I m n o ***** dfr Figure 60 G31:l t o o l s and cores a-g h-j k-o p-v w-z b i f a c e s p r o j e c t i l e p o ints u nifaces u t i l i z e d f l a k e s cores b i p o l a r cores 190 j k I m n s ^ ^ ^ ^ t * u ^ ^ ^ ^ r A 4 * 4 v w x y 4 Figure 61 F 8 : l t o o l s a - i ' : b i f a c e s k I m Figure 62 F 8 : l t o o l s and cores a-h i . j l,m n,o p»q r-w x-c' b i f a c e s unimarginals u t i l i z e d f l a k e s p r o j e c t i l e p o i n t s g r a v e r / d r i l l s cores b i p o l a r cores s p a l l t o o l Figure 63 a b c d g h-1 F12:5 t o o l s and cores b i f a c e p r o j e c t i l e p o i n t u t i l i z e d BRF l a r g e b i f a c e l a r g e unifaces core b i p o l a r core Figure 64 a: b: J22:2 t o o l s p r o j e c t i l e p o ints unimarginal 192 4 j j a b Figure 65 J38:2 t o o l s and cores a-c d,e f b i f a c e s p r o j e c t i l e p o ints core b e d e f t • I » g h | | f l | Figure 66 K2:l t o o l s and cores a-f i j k-p b i f a c e s p r o j e c t i l e points uniface u t i l i z e d BRF b i p o l a r cores 193 I i l i | $ k I i w X a b c d e f g h i i I I m Q I s t o J * « > l 4 • Figure 67 EeRjI tools a-s t u-x bifaces bimarginal projectile points Figure 68 EeRj 1 tools and cores a-p q r-u v-w x-a* utilized flakes core bipolar cores pieces esquillees utilized BRF's * y 194 c d e Figure 69 EeRjI tools a - i j-o p-w unifaces unimarg inals ut i l i z e d flakes h i P Q r s 195 CHAPTER 6 A MULTIREGIONAL PERSPECTIVE ON LITHIC ASSEMBLAGE VARIABILITY 6.1. Introduction The theoretical frameworks developed by Binford (1979), Ebert (1979), Goodyear (1979) and Pokotylo (1978) provide the be- havioral perspective within which the analyses in this chapter are undertaken. The analyses seek an understanding of the basic causes of l i t h i c assemblage va r i a b i l i t y in the central and southern In- terior Plateau, through the derivation of consistent multivariate and bivariate patterns from which technological strategies can be inferred. The analyses proceed by exploring inter-assemblage var- iations with respect to major factors thought to determine the character of l i t h i c technological practices, including stages of l i t h i c reduction, kinds of raw materials, tool maintenance, and tool and debitage co-occurrences within major settlement site types. Three general hypotheses are tested in the following analyses of the context and l i t h i c content of the 38 assemblages from the four regions of study. This part of the study evaluates the u t i l i t y of the debitage classification that has been formulated in the ex- perimental program of this study as a useful and reliable means of inferring l i t h i c technological strategies that were employed by the prehistoric residents of the central and southern Interior Plateau. 196 The f o l l o w i n g hypotheses have been formulated on the b a s i s of the current models of l i t h i c technology and settlement p a t t e r n s that have been developed by E i n f o r d (1979), Ebert (1979) and Good- year (1979), discussed at l e n g t h i n Chapter 2: 1. Obsidian and chert raw-materials- e x h i b i t v a r i a - b i l i t y that is- the r e s u l t of ' extensive economizing p r a c t i c e s . This i s expected becaus-e these materials- are r e l a t i v e l y r a r e or completely absent w i t h i n the regions of study, w h i l e v i t r e o u s b a s a l t i s the domi- nant raw m a t e r i a l w i t h i n a l l of the regions. 2. Regardless of the importance of raw m a t e r i a l f a c t o r s , l i t h i c maintenance p r a c t i c e s are important determinants of the v a r i a b i l i t y of assemblages. These should be emphasized d i f f e r e n t i a l l y among s i t e s w i t h i n and across re g i o n s . 3. General s i t e occupation purposes across the four r e - gions can be r e l i a b l y p r e d i c t e d on the b a s i s of debitage and t o o l c o - v a r i a t i o n s . 6.2. Reduction Factors The major i s s u e of i n f e r r i n g l i t h i c t e c h n o l o g i c a l behavior by means of re d u c t i o n stage measures of l i t h i c debitage i s best answered i n a m u l t i v a r i a t e manner, to d e r i v e major f a c t o r s of v a r i a b i l i t y , from data that can be p a r t i t i o n e d i n many ways. 197 In the present case, i t is major patterns of inter-assemblage vari a b i l i t y that are sought (see Matson 1980), and individual assemblage inferences are offered only after the entire set of hypotheses has been evaluated. There exist many possible ways of computing similarities between assemblages, including various correlation coefficients and similarity and difference measures (see Sneath and Sokal 1973) . I selected a City Block distance measure calculated on standard- ized percentages of the 10 debitage classes within each assemb- lage (Table 12). Percentage calculations are necessary because variable sampling rates and wide variation in sample size would otherwise automatically severely bias the analysis. Visual group- ings and data reduction or "factoring" are accomplished by f i r s t clustering the sites, using Ward's Error Sum of Squares method (Sneath and Sokal 1973) , an algorithm option available in a pack- age of cluster routines developed by Wood (1973). The City Block distance matrix was also factored by Metric Multidimensional scal- ing (Matson 1978; Matson and True 1974; Torgerson 1958), following standardization of the percentage data, in which the mean of each variable becomes zero with a standard deviation of one. It is im- portant that what is being reflected in the multivariate analyses i s generalized reduction stage patterning and not sheer abundance of material. Standardization was conducted on the percentage data to emphasize va r i a b i l i t y within site units, rather than within the debitage variables, because inter-site patterns are being sought 198 (Sokal and Sneath 1973: 178) . When t h i s i s done, the problem w i t h s i z e f a c t o r biases that i s prevalent i n s c a l i n g and o r d i n a t i o n techniques i s g r e a t l y reduced (Sokal and Sneath 1973: 178). Both the c l u s t e r and s c a l i n g analyses are conducted i n Q-mode f a s h i o n where the s i t e cases are grouped on the ba s i s of the debitage v a r i a b l e s . For d e t a i l e d d i s c u s s i o n of c l u s t e r i n g and s c a l i n g techniques, i n c l u d i n g those used here, see Matson and True (1974), Matson et a l . (1979), Matson et a l . (1980) and Pokotylo (1978). The c l u s t e r diagram i s not reproduced here, but the three major c l u s t e r s derived i n "that a n a l y s i s are shown i n the TSCALE p l o t of the f i r s t two dimensions of v a r i a b i l i t y ( Figure 70). To i n t e r p r e t the major f a c t o r s of v a r i a b i l i t y , rank-order c o r r e l a t i o n c o e f f i c i e n t s (Spearman's r ) , are computed on the debitage c l a s s e s ' percentages against the p o s i t i o n of the assemblages on the dimen- s i o n s . This r e v e a l s that Dimension I accounts best f o r the amount of l a t e stage debitage i n s i t e s , of PRB and Shatter percentages com- bined ( r s = 0.95), and Dimension I I i s explained by the percentage of combined PRB and Shatter middle stage debitage ( r s = 0.76). Both of these c o r r e l a t i o n s are s i g n i f i c a n t at p : .005. These two dimen- sions account f o r 47% and 22% of Trace v a r i a b i l i t y i n the data over- a l l . The remaining 31% of Trace r e q u i r e s a f u r t h e r four dimensions, none of which i s r e a d i l y i n t e r p r e t a b l e i n terms of the r e d u c t i o n c l a s s e s . I t i s notable a l s o i n t h i s metric s o l u t i o n , that no t r i a n g l e i n e q u a l i t i e s (Anderberg 1973) were v i o l a t e d , adding to the confidence i n the i n t e r p r e t a t i o n s . The s o l u t i o n shows that i f general i n t e r - 199 CM — O ro o rr CM ' / I * / ~ ' ro ro CM 3 cr I \ I ^ \ \ i \ \ - \"-'-w \" \ * CM ro O CC CO E ro rr CJ CO CM cr o rr \ \ H V- P J lO E — 0> ro UJ CM CM 6S CO ro — CC ro o» \ \ I CO <T o CC \ \ UJ \ \ i n \ \ \ CJ O c j cn 2 — — LU E _ i x • • -4 * CO \ <->• DC ' co / y sz O) Q • Q 2 5 LU I- • < CC CVJ CM \ _ \ CM \ " \ Figure 70. Torgerson's Metric Multidimensional Scaling of debitage class percentages, City Block Dis- tance. Dimension 1 accounts for 47% of Trace, Dimension 2 for 22%. Broken lines indicate Ward's clusters. 200 assemblage variation is being investigated, i t is feasible to reduce the 10-state debitage classification somewhat, to the two major factors of early/late and middle, but i t also indicates that general v a r i a b i l i t y in BRF's, BPO's, BPCO's and CORES is harder to account for, and thus these classes should be retained. The sites from a l l four regions are distributed across the scaling diagram, with the Eagle Lake sites appearing to exhibit most va r i a b i l i t y , ranging from 3.23% late debitage to 54.54%, 12.9% to 35.94% middle debitage, and 16.24% to .83.68% early deb- itage. The Mouth of the Chilcotin sites are also highly variable, ranging from 0% to 35.38% late, 16.92% to 45.88% middle, and 19.23% to 41.67% early stage material. Among the Lillooet sites, varia- b i l i t y is constrained, perhaps bedause only five assemblages are represented, but these are s t i l l quite varied in content, with re- duction ranges of 12.92% to 62.50% late, 20.83% to 41.53% middle and 16.67% to 49.5% early stage debitage. Among the Hat Creek assemblages, a great deal of var i a b i l i t y is also exhibited, with the sites having reduction ranges of 8.33% to 44.04% late, 25.0% to 47.83% middle, and 14.67% to 58.43% early reduction stage items. Sites 16:1 and 22:1 from Eagle Lake are the extreme cases of early reduction sites. These are low frequency debitage assemblages with large flakes and cores. Sites 14:2, CR28, CR40, CR73 and 26:3 from Eagle Lake are a l l at the late stage end of the TSCALE diagram. These assemblages have no cores, and are small collections. Site CR92, while within the "early" cluster, i s clearly more related to 201 site 19:1 and sites CR64, 32:1 and EIRw 4. These sites, ex- cept for CR64 and 32:1, have relatively abundant artifacts, some cores and/or bipolar cores, and while not exceptionally high on the middle reduction stage scale, exhibit broader spreads of the relative percentages of the debitage classes. The Mouth of the Chilcotin patterns are different from the ELP case. Here site 12:6 is clearly by i t s e l f , with a small assemblage containing a relatively large number of bipolar cores and no late stage debitage at a l l . EkRo 48 and 9:2 are s p l i t in the cluster analysis, but in the TSCALE diagram are related to each other perhaps more than to the group of sites 4:2, 4:5, 5:1 and 9:1. A l l of these sites exhibit wide ranges of reduction stages within their assemblages, and have cores and/or bipolar cores, but the five latter sites are clearly very similar in most respects, and especially in the high amounts of middle stage deb- itage present." Sites EkRo 18, EkRo 31 and'2:3 have predominantly late stage trends, however 2:3 contains the greatest percentage of CORES of a l l of the MOC assemblages. The five Lillooet assemblages occur in three separate clusters. EeRk 7 emphasizes early stages and has a very abundant assemblage that contains cores, bipolar cores and bipolar flakes in relatively large quantities. EeRl 40 and EeRk 4:38 contain relatively low amounts of late stage debitage, high middle stage percentages, and moderately high early stage percentages of debitage. Both sites have bipolar cores and flakes, but no hand-held cores. EeRl 41 and 202 EeRk 16 are most interesting, with very l i t t l e early, and re- latively large amounts of late stage debitage. These two sites also contain relatively few core materials. Among the Hat Creek sites, J22:2 and F12:5 are similar early stage-predominant l i t h i c scatters, both with very l i t t l e late stage material, but each is very different from the other in terms of actual abundance of total debitage. Sites EeRj 1, G31:l, G23:l, K2:l, and J38:2 a l l occur within the "middle" stage cluster, but are widely spaced within i t . EeRj 1 is relatively low on the middle scale, and J38:2 is the highest middle stage content assem- blage of the entire 38 sites. J38:2 also contains a f a i r amount of late stage material, and has the highest number of CORES for a l l Hat Creek sites. Sites G21:9, F8:l and G2:12 are the late stage sites from Hat Creek, with F8:l being the odd one here with several cores and bipolar cores. On the whole, the sites clustering in the "middle" cluster have more-or-less evenly spread debitage stage distributions; those in the "early" cluster and in the "late" cluster have more restricted patterns. Most sites in association withlate stage reduction have biface reduction flakes, although these range from being relatively common to being completely absent. The following sites are out- standingwlth regard to the high percentages of BRF's contained in their assemblages: CR40 at ELP, EkRo 18 at MOC, K2:l and EeRj 1 at HAC, and none are outstanding in the LIL sample. The assemblages' major patterns of reduction stage va r i a b i l i t y 203 can be i n t e r p r e t e d as emphasizing e a r l y / c o r e r e d u c t i o n , middle/wide range and late/maintenance. When the s i t e s are grouped w i t h respect to these i n t e r p r e t a t i o n s and by t h e i r context, s e v e r a l i n t e r e s t i n g p atterns are apparent (Table 15). No housepits e x h i b i t the extreme of e a r l y stage predominance i n t h e i r assemblages (except p o s s i b l y EkRo 48), but the 10 excavated housepit assemblages are s p l i t be- tween middle/wide ranging and those w i t h late/maintenance predom- inance. L i t h i c s c a t t e r s without features are spread among the three major r e d u c t i o n f a c t o r s , but other l i t h i c s c a t t e r s are more l i m i t e d i n content. L i t h i c s c a t t e r s w i t h housepits i n c l u d e both e a r l y / c o r e r e d u c t i o n and middle/wide ranging assemblages, but only one of these i s an e a r l y / c o r e r e d u c t i o n type of s i t e . The l i t h i c s c a t t e r s w i t h cachepits and those w i t h f i r e c r a c k e d rock features are spread among the middle/wide ranging f a c t o r and the late/main- tenance f a c t o r . At t h i s stage of the analyses, the debitage c l a s s i f i c a t i o n appears to have considerable a b i l i t y to r e v e a l b a s i c patterns of l i t h i c t e c h n o l o g i c a l processes of assemblage formation. I t should be noted here that the patterns revealed among the Hat Creek s i t e s do not completely agree w i t h Pokotylo's (1978a) i n t e r p r e t a t i o n s of the s i x s i t e s here that were included i n h i s study. One reason fo r t h i s apparent discrepency i s that the debitage c l a s s i f i c a t i o n employed i n t h i s study i s much more d e f i n i t e i n i t s assignment of debitage to stages* whereas Pokotylo's inferences depended on choos- ing p atterns from s e v e r a l v a r i a b l e s (1978: 250 - 258). 204 ASSEMBLAGE CONTEXT Excavated Housepits L i th ic Scatters L i th ic Scatters with Housepits L i th ic Scatters with Cachepits Li th ic Scatters with Firecracked Rock EARLY/ CORE 16:1 22:1 F12:5 J22:2 12:6 MIDDLE/ WIDE RANGE EkRo 48 EeRk 7 EeRk 4:38 EeRl 40 EeRj 1 4:2 4:5 G23:l G31:l K2:l 32:1 EIRw 4 4:1 5:1 9:1 9:2 19:1 CR92 2:3 CR64 J38:2 LATE/ MAINTEN- ANCE CR73 EkRo 18 EkRo 31 EeRk 16 EeRl 41 14:2 G21:9 G2:12 26:3 CR28 CR40 F8:l TABLE 15. Assemblage context compared with major reduction factors. 205 In a cluster analysis of Hat Creek debitage assemblage attributes, Pokotylo found four of the six sites to occur in a cluster (Cluster 3) interpretable as exhibiting a "wide range" of reduction steps, with no one stage predominating (F8:l, F12:5, G21:9, G2:12). In this study, G21:9, G2:12, and F8:l appear to contain debitage indicative of late stages. Sites G23:l and G31:l are not as extreme in diversion from Pokotylo's findings, in that both here are understood to con- tain middle stage debitage, and in the purely Hat Creek study both are "wide ranging", with G23:l trending towards late steps (Pokotylo 1978: 250 - 258). In sum, I think the danger of using large attribute l i s t s is that factors other than reduction stage- are being measured, such as core geometry and raw material char- acteristics. In fact I think i t feasible to eventually use other flake morphological characteristics such as platform angles and size variables to reconstruct core and tool shapes, and this is certainly an area where concise experimentation and mathematical derivation i s required. 6.3. Hypothesis 1: Raw Material Factors Since vitreous basalt is widely recognized as having been the primary l i t h i c raw material that was used in stone tool manufacture in the Interior Plateau, other raw materials such as cherts and ob- sidians may have been diffe r e n t i a l l y conserved, or used in d i f f e r - ent manners, simply by virtue of their relative regional scarcity. 206 This p o s s i b i l i t y i s worth i n v e s t i g a t i n g f o r what i t can t e l l us about hunter-gatherer m o b i l i t y , given the arguments of B i n f o r d (1979) and Goodyear (1979) discussed i n Chapter 2. In t h i s study, I cannot c o n t r o l f o r p r e c i s e source l o c a t i o n s of any raw m a t e r i a l s , although i n a separate paper (Magne 1979) I have d i s - cussed raw m a t e r i a l occurrence i n Upper Hat Creek V a l l e y . G e n e r a l l y , throughout- the I n t e r i o r P l a t e a u , b a s a l t , e i t h e r v i t r e o u s or g r a n u l a r , i s found as cobbles i n g l a c i a l t i l l s or i n stream beds. Apparent concentrations of good q u a l i t y b a s a l t , such as i n the Arrowstone H i l l s east of the Hat Creek V a l l e y , or i n the Baezeko R i v e r of the n o r t h - c e n t r a l I n t e r i o r , have yet to be stu d i e d w i t h a combined g e o l o g i c a l and a r c h a e o l o g i c a l perspec- t i v e . W i t h i n B r i t i s h Columbia, o b s i d i a n i s known to have two main sources, Mount Ed z i z a i n the f a r northwest (Fladmark 1982b), and Obsidian Creek near Anahim Peak i n the c e n t r a l I n t e r i o r ( N e l - son and W i l l 1976). The o b s i d i a n m a t e r i a l s studied here from the Eagle Lake r e g i o n are b e l i e v e d to have o r i g i n a t e d from the Obsid- i a n Creek area, but t h i s i s based on macroscopic c h a r a c t e r i s t i c s , and source s t u d i e s have not been undertaken. The very few o b s i d i a n pieces from the Mouth of the C h i l c o t i n were sourced by X-Ray f l u o r - escence, and probably came from the Obsidian Creek source area. Obsidian i s present, but very r a r e , i n s i t e s from the L i l l o o e t and Hat Creek regions. However, i t does hot occur i n any of the assem- blages stu d i e d here. Chert raw m a t e r i a l s occur both as stream.and g l a c i a l t i l l cobbles, and as outcrops. In the Hat Creek V a l l e y r e - 207 gion, outcrops occur in the northern and eastern parts of the region, in association with jaspers and agates that are actively mined by rockhounds (Danner 1970; Learning 1971). It is notable that in the Cache Creek streambed east of Hat Creek, both chert and basalt cobbles can be obtained. A .comprehensive study of l i t h i c raw material sources of the Interior Plateau is urgently needed, to provide fixed geographical l o c i from which the spread of materials can be studied, such as Choquette (1981) has i n i t - iated in the Kootenay d i s t r i c t of southeastern British Columbia. The proposition that differences in debitage and tool assem- blage v a r i a b i l i t y are due to -raw materials is tested here, by comparing vitreous basalt to obsidian at Eagle Lake, and to chert at the Mouth of the Chilcotin, Lillooet and Hat Creek. Granular basalts are not tested for differences, since these are already recognized. Granular basalt occurs almost solely as early stage debitage, and in restricted tool classes such as spall tools and core tools. As can be seen in Tables 16 and 17, this raw mat- eri a l comprises most of the 16:1 and 22:1 assemblages (89% and 99%), which are considered predominantly early/core reduction sites, based on the scaling analysis of debitage. Site CR64 contains a moderate amount of granular basalt, as do sites 19:1, CR92 and CR73. Among the Mouth of the Chilcotin sites, EkRo 48 debitage is 95% granular basalt and the other two housepits lack i t entirely. Sites 2:3, 5:1 and 9:1 contain moderate amounts of the material. -With-- in the Lillooet and Hat Creek sites, only one assemblage in each re- 208 gion (EeRl 41 and K2:l) contains a minimal amount of granular basalt (Tables 16,17). The f i r s t question to ask i s : Are obsidian and chert con- served? If so, according to Binford's (1979) and Goodyear's (1979) models, then these materials should be late stage deb- itage more often than vitreous basalt. This question is add- ressed by Chi-square tests of independence in contingency tables (Mendenhall 1975), where the flakes of each material are grouped into the general early, middle and late stage classes (Tables 18,19, 20, 21). The tests show that among the Eagle Lake, Lillooet and Hat Creek assemblages, there are no significant differences in the stage distribution of debitage by vitreous basalt or obsidian/chert materials. In the Mouth of the Chilcotin sample there are significant differences, and vitreous basalt is brought to late stages proportionately more often than the chert debitage, and the chert materials occur proportionately more often as early stage than is to be expected. Thus i t is apparent in the debitage, that obsidian and chert are not extensively maintained, and that in the Mouth of the Chilcotin region, there is a tendency to maintain vitreous basalt, such that i t is brought to late stages quite often, while chert materials are used more expediently. The second question i s : Are the patterns of conservation and maintenance evident in the tools l e f t at sites? Again, i f chert and obsidian are conserved, then we could expect the tools to be small and complex in relation to those made of vitreous basalt. The tools 209 Vitreous Granular Basalt Basalt Obsidian Chert Quartzite 14:2 100 0 0 16:1 11.1 88.9 0 19:1 61.7 14.9 22.1 1.2 22:1 0 98.7 1.2 26:3 94.4 1.0 4.6 32:1 92.6 1.2 6.2 CR28 83.8 0 10.8 5.4 CR64 76.9 17.9 5.1 CR40 90.6 0 9.4 CR73 49.1 9.4 33.9 7.5 ElRw 4 95.6 0 4.2 0.2 CR92 70.3 8.8 20.8 EkRo 18 63.1 0 36.9 EkRo 31 79.1 0 1.6 19.4 EkRo 48 4.3 94.7 2:3 52.6 21.9 25.4 4:2 20.5 .4 79.1 4:5 22.2 77.8 4:1 40.2 3.4 56.4 5:1 78.3 10.8 10.8 9:1 70.1 15.7 14.2 9:2 42.2 57.8 12:6 95.0 5.0 EeRk 16 95.8 4.2 EeRl 41 90.9 4.5 4.5 EeRk 7 99.5 .1 .4 EeRk 4:38 94.9 5.1 EeRl 40 99.5 .5 G21:9 5878 41.2 G23.-1 39.0 61.0 G2-.12 86.0 14.0 G31:l 33.2 66.8 f 8 : l 73.3 26.7 F12:5 100.0 J22:5 100.0 J38:2 100.0 K2:l 20.2 4.9 74.8 EeRj 1 76.5 23.5 TABLE 16. Percent raw material composition of debitage assem- blages by counts. 210 % Vitreous % Granular Site Basalt Basalt % Obsidian % Chert % Other 14:2 50.00 50.00 16:1 20.00 20.00 60.00 19:1 67.86 10.71 21.43 22:1 100- - 26:3 75.00 25.00 32:1 69.23 30.77 CR28 100 CR64 CR40 66.67 16.67 16.67 CR73 50.00 25.00 EIRw 4 94.74 5.26 CR92 67.39 8.70 23.91 EkRo 18 76.47 23.53 EkRo 31 68.18 18.18 4.55 9.09 EkRo 48 35.29 58.82 5.88 2:3 48.65 22.86 22.86 8.11 4:2 54.29 5.71 37.14 2.86 4:5 50.00 12.50 37.50 4:1 41.67 12.50 41.67 4.17 5:1 40.00 40.00 4.00 16.00 9:1 75.00 8.33 16.66 9:2 46.15 15.38 38.46 7.69 12:6 66.67 33.33 EeRk 16 85.00 5.00 10.00 EeRl 41 79.31 17.24 3.45 EeRk 7 96.55 1.72 1.72 EeRk 4:38 90.00 10.00 EeRl 40 93.42 1.32 2.63 2.63 G21:9 57.69 3.85 38.46 G23:l 75.00 25.00 G2-.12 75.00 25.00 G31:l 77.27 4.55 18.18 F8:l 84.91 13.21 "1.39 F12:5 100 J22:2 100 J38:2 85.71 14.29 K2:l 60.00 40.00 EeRj 1 84.06 1.45 14.49 TABLE 17. Raw material composition of tool assemblages by percentages. 211 Debitage General Reduction Stage EARLY MIDDLE LATE Vitreous Basalt Obsidian N= 2940 885 844 716 ; i: (871.55); (843.28) (730.17) 163 170 162 (176.45) (170.72) (147.83) 1048 1014 878 '2445 495 2940 X = 11.07 p = not significant at .05 TABLE 18. Chi-square test of Eagle Lake debitage general reduction stages by raw material. Debitage General Reduction Stage EARLY MIDDLE LATE Vitreous 187 278 228 Basalt (229.20) (277.34) (186.45) Chert 451 494 291 (408.80) (494.68) (332.55) 638 772 519. N = 1929 X 2 = 26.58 p = .05 TABLE 19. Chi-square test of Mouth of the Chilcotin debitage general reduction stages by raw material. 212 Debitage General Reduction Stages CO u OJ 4J Ct) S & co Pi EARLY MIDDLE LATE Vitreous 1869 1545 684 Basalt (1863. 45) (1547. 07) (685. 49) 4098 Chert 4 (9. 55) 10 ( 7. 93) 7 (3. 51) 21 1873 1555 691 4119 N = 4119 X = 7.26 p = not significant at .05 TABLE 20. Chi-square test of Lillooet debitage general reduction stages by raw material. Debitage General Reduction Stages u CD cd Pi EARLY MIDDLE LATE Vitreous Basalt 696 (693.37) 756 (793.97) 677 (641. 66) -Chert 524 (526.63) 641 (603.03.)'. 452 (487. 34) 1220 1397 1129 N = .3746 2129 1617 3746 X 2 = 8.74 p = not significant at .05 TABLE 21. Chi-square test of Hat Creek debitage general reduction stages by raw material. 213 are also analysed by Chi-square tests of independence. Here weights and scar counts are pooled into regular intervals by raw materials (Tables 22,23, 24, 25, 26, and 27). The Lillooet materials are not analysed because the sample sizes are too small to f i t the requirements of the test. The Chi-square tests demonstrate that there are no stat- i s t i c a l l y significant differences in the sizes or Complexity of tools due to raw material factors. This finding generally supports the debitage tests, and indicates that the differences observed between the Mouth of the Chilcotin chert and vitreous basalt are not consistent. The contingency tables do indicate that Eagle Lake obsidians and basalts relate to each other differently than do the Mouth of the Chilcotin and Hat Creek basalt and chert mat- erials. In the Mouth of the Chilcotin and Hat Creek regions (Tables 24, 25, 26 and 27) chert tools tend to be larger and less complex than basalt tools, while at Eagle Lake, both basalt and obsidian tools tend to be small. I suggest that the probability tests are generally reliable, given the variety and size of the entire tool- sample, and that separate study is required of varia- b i l i t y of such factors within specified tool types. Again, the data have been gathered for such analyses, but their manipulation is currently beyond the scope of this study. These findings do not necessarily indicate that raw material conservation and maintenance was not practiced, but only that in relation to each other, vitreous basalt and obsidian/chert are re- 214 Weight Intervals (Grams) cd • H Vitreous cu Basalt •u cd a S Obsidian co 0 - 1 2 3 4 £5 68 (67. 67) 28 (30.69) 25 (15. 74) 8 (8. 66) 25 (21.25) 18 (18. 33) 11 (8.31) 5 (4. 26) 3 (2. 34) 2 (5.75) | 86 39 30 11 27 144 39 183 N = 183 X = 4.62 p = not significant at .05 TABLE 22. Chi-square test of Eagle Lake tool sizes by raw material. Scar Count Intervals <j} Vitreous _ rt Basalt rt Obsidian P5 0 1 - 5 6 -'10 11 - 15 16 - 20 >21 18 (17.31) 12 (16.52) 38 (42.49) 40 (36.20) 22 (18. 10) 14 (13. 38) 4 (4.69) 9 (4.48) 16 (11.51) 6 (9.80) 1 (4. 90) 3 (3. 62) 22 21 54 46 23 17 N = 183 X 2 = 14.1 p = not significant at .05 144 39 183 TABLE 23. Chi-square test of Eagle Lake tool scar counts by raw material. 2 1 5 Weight Intervals (Grams) ca •H u ai cd S IS Pi Vitreous Basalt Chert 0 - 1 2 3 4 2 5 5 4 ( 5 0 . 5 1 ) 2 1 ( 2 0 . 7 6 ) 1 5 ( 1 3 . 1 5 ) 1 6 ( 1 4 . 5 3 ) 3 1 ( 3 8 . 0 6 ) 1 9 ( 2 2 . 4 9 ) 9 ( 9 . 2 4 ) 4 ( 5 . 8 5 ) 5 ( 6 . 4 5 ) 2 4 ( 1 6 . 9 4 ) 7 3 3 0 1 9 2 1 5 5 N = 1 9 8 X = 6 . 3 7 p = not significant at . 0 5 1 3 7 6V 1 9 8 TABLE 2 4 . Chi-square test of Mouth of the Chilcotin tool sizes by raw material Scar Count Intervals i-H cd 0 1 - 5 6 - 1 0 1 1 - 1 5 1 6 - 2 0 5 2 1 . . . . •H '.QJ. "Vitreous % Basalt a 4 5 ( 4 9 . 8 2 ) 1 5 ( 1 7 . 3 0 ) 3 0 ( 2 9 . 7 5 ) 2 0 ( 1 6 . 6 1 ) 2 0 ( 1 4 . 5 3 ) 7 ( 8 . 9 9 ) 1 3 7 £ Chert 2 7 ( 2 2 . 1 8 ) 1 0 ( 7 . 7 0 ) 1 3 ( 1 3 . 2 5 ) 4 ( 7 . 3 9 ) 1 ( 6 . 4 7 ) 6 ( 4 . 0 1 ) 6 1 7 2 2 5 4 3 2 4 2 1 1 3 1 9 8 N = 1 9 8 X = 1 2 . 8 8 p = not significant at . 0 5 TABLE 2 5 . Chi-square test of Mouth of the Chilcotin tool scar counts by raw materials. 216- Weight Intervals (Grams) 0-1 £5 Vitreous Basalt & Chert 40 (31 .98) 15- (16.38) 21 (17.94) 11 (12. 98) 101 (109.21 188 1 (9.02) 6 (4.02) 2 (5.06) 5 (3. 52) 39 (30.79) 53 41 21 23 16 140 ' 241 241- x2 = 15.65 P = not s i g n i f i c a n t at .05 TABLE 26. Chi-square t e s t of Hat Creek t o o l sizes'by raw material Scar Count Intervals CO u Vitreous 0J 4J Basalt ca S § Chert on 0 1 - 5 6 - 1 0 11 - 15 >15 47 12 49 42 38 (48. 37) (15.60) (49.93) (42. 90) (31.20) 188 15 8 15 13 2 (13. 63) (4.40) (14.07) (12. 10) (8.80) 53 62 20 64 55 40 241 N = 241 X = 10.86 p = not s i g n i f i c a n t at .05 TABLE 27. Chi-square ofsHat Creek t o o l scar counts by raw material. 217 duced in the same ways, and used to make tools of the same orders of size and complexity. What the previous tests do not show is that one raw mat- e r i a l may serve to replace another. Consider Figure 71, where the relative amount of debitage composedNof vitreous basalt, and the relative amount of tools composed of vitreous basalt are plotted per site. Here and in following figures, lines indicating 2:1, 1:1 and 1:2 ratios\;are provided to f a c i l i t a t e reference. Sites low on both scales have relatively large amounts of granular basalt (16:1 and EkRo 48), as discussed above, and the "chert debitage" (Matson et_ a l . 1979) sites from the Mouth of the Chilcotin (4:1, 4:2, 4:5, 9:2) occur as two groups. One of these groups has relatively high vitreous basalt tool contents and low vitreous basalt debitage contents (4:2, 4:5), and the ether two sites are composed of about 50% vitreous basalt tools and also about 40% vitreous basalt debitage. Site K2:l from Hat Creek occupied a place on this graph that is similar to 4:2 and 4:5, but even though i t contains a large amount of chert debitage, no chert tools were found there. Sites G23:l and G31:l are also similar to these three sites, but contain relatively more vitreous basalt tools and debitage. Sites 14:2 and 5:1 are at the opposite ends of the scale, with relatively low numbers of vitreous basalt tools, but relatively large amounts of vitreous basalt debitage. Other sites are not as extreme with respect to these measures, and cluster around the 1:1 line of the graph. CR64 is the exceptional site, in that i t contains no teds whatsoever. 218 % vitreous basalt all tool counts Figure 71. Graph of the percent of debitage derived from vitreous basalt vs. the percent of tools derived from vitreous basalt per assemblage. 219 From this graph i t appears that sites 4:2, 4:5, K2:l, G23:l and G31:l are places where the vitreous basalt tools that were deposited were replaced by tools made of chert, while at 16:1 and EkRo 48, granular basalt is the replacement material. Another way of checking these patterns is to plot the rela- tive amounts of tools versus the relative amounts of debitage that are composed of obsidian or chert (Figure 72). Here we see that at 16:1, obsidian and vitreous basalt tools were replaced by granular basalt, and that sites 4:2, 4:5, G23:1, G31:1.. and K2:l are clearly separated from the other assemblages, in that chert tools appear to have been removed from the sites following their manu- facture. Site 9:2 also patterns out in this way, but less strongly so. The other assemblages occur close to the 1:1 line on the graph, or exhibit such low percentages of obsidian/chert debitage and tools as to be beyond accurate interpretations here. Note that use of the term "replacement" in discussing the above patterns does not imply that such occurred in a single epi- sode of site occupation. Since the present analyses deal only with complete collections, at times from large areas and at others from small site areas, I cannot control for the influences of site re- occupation, but only the combined results of a l l site occupations. Overall, I assume that sites were reoccupied for the same reasons as their i n i t i a l establishment, i f at a l l . To a lesser degree, i t can also be assumed that succeeding occupants are aware of and use the materials l e f t by prior occupants, whether by design or circum- 220 FIGURE 72. Plot of the percent of debitage derived from Chert or Obsidian vs. percent of tools for the same raw material per assemblage. 221 stance. These ideas do not r u l e out d i s c o v e r i n g the r e l a t i v e average d u r a t i o n of s i t e occupations, nor do they e l i m i n a t e the p o s s i b i l i t y of comparing s i t e t o t a l lengths of occupations. Thus, when raw m a t e r i a l X i s being replaced by m a t e r i a l Y, then i n the long run t o o l s of X are being brought i n and deposited, w h i l e t o o l s of Y are being made and exported. I t should be apparent here why p r e c i s e raw m a t e r i a l source l o c a t i o n s would be u s e f u l i n a c t u a l l y mapping m o b i l i t y and trade p a t t e r n s . O v e r a l l , the r e s u l t s are appealing because at Eagle Lake s i t e s , where o b s i d i a n i s imported probably from the Obsidian Creek source area, i t can be expected that o b s i d i a n would be conserved. Figure 72 shows that t h i s i s g e n e r a l l y t r u e , w^th s e v e r a l of the Eagle Lake assemblages e x h i b i t i n g o b s i d i a n t o o l s c u r a t i o n whereas o b s i d i a n debitage i s being deposited (.22:1, 26:3, 32:1, CR28, CR64, ElRw 4 ) . This i s true a l s o of chert t o o l and debitage patterns at s i t e s EkRo31, 4:2, 4:5 and 5:1 from the Mouth of the C h i l c o t i n and s i t e s G31:l, G23:l, F 8 : l and K2:1 at Hat Creek. With respect to those s i t e s that do e x h i b i t meaningful p a t t e r n s , some subs t a n t i v e conclusions can be drawn. S i t e 16:1 appears to be an e x c e l l e n t example of B i n f o r d ' s " s i t u a t i o n a l " type of s i t e , where t o o l s of h i g h q u a l i t y m a t e r i a l s were replaced by the coarser granular b a s a l t . As f o r the Mouth of the C h i l c o t i n "chert debitage" s i t e s , i t i s p o s s i b l e , i n support of Matson est al.'s (1979) p o s i t i o n , and contrary to my p o s i t i o n i n Chap- t e r 3, that the s i t e s are evidence of high m o b i l i t y , i f chert m a t e r i a l s from a f a r were brought i n t o the r e g i o n , and then used to make t o o l s 222 that ordinarily would be made of vitreous basalt. Concerning such a possibility, Binford (1979: 260) has written of raw mat- erials that he is convinced "that va r i a b i l i t y in the proportions of raw materials found at a given site is primarily a function of the scale of the habitat which was exploited from the site location, possibly coupled with a founder effect resulting from discard on the site of items which had been manufactured pre- viously at some other location". On the other hand, given that the Canyon Shuswap were great traders, i f these sites are late prehistoric, then i t is equally possible that the chert was acquired by trade. Knowing chert source locations would great- ly aid in such a debate. 6.4. Hypothesis 2: Implement Maintenance and Curation Factors Ebert's (1979) model of stone tool va r i a b i l i t y holds that implements meant to be transported and re-used w i l l tend to be small and complex, while those that are used once and l e f t at':the l o c i of use w i l l tend to be large and simple. In Binford's (1979) terms, the small, complex items are curated personal gear, the larger, simpler tools are expedient types. A sort that Ebert does not consider, site furniture, probably is quite variable in size and complexity, depending on their specific intended purposes. Ebert's (1979) approach to inferring the relative degrees of mo- b i l i t y that produced assemblages relies on size and complexity measures for individual tools. However, i t is logical as in Bin- 223 ford's (1978b) faunal assemblage s t u d i e s , that s i m i l a r measures can be a p p l i e d to e n t i r e assemblages to c h a r a c t e r i z e s i t e form- a t i o n processes at r e g i o n a l and i n t e r - r e g i o n a l s c a l e s of compar- is o n . In Figure 73, are p l o t t e d e n t i r e assemblage analogs of Ebert's (1979) suggested s i z e and complexity measures (Table 28), w i t h the m o d i f i c a t i o n that weight of t o o l s has been s u b s t i t u t e d f o r Ebert's s i z e index (volume), since volume can be expected to be n e a r l y p e r f e c t l y c o r r e l a t e d w i t h weight, v a r y i n g only w i t h s p e c i f i c g r a v i t y of the raw m a t e r i a l s under c o n s i d e r a t i o n . At the s c a l e of a n a l y s i s undertaken here, i t i s u n l i k e l y that spe- c i f i c g r a v i t y of the raw m a t e r i a l s v a r i e s enough to be a s i g n i f - i c a n t determinant of assemblage content. The f i g u r e shows that assemblages vary a great deal w i t h respect to the r e l a t i v e complexity of t o o l s i n r e l a t i o n to t h e i r s i z e . S i t e s CR28, ElRw 4, 4:2, 4:5, 9:1, 12:6, G2:12 and J22:2 co n t a i n r e l a t i v e l y complex t o o l s i n r e l a t i o n to t h e i r s i z e , w h i l e s i t e s 16:1, 22:1, EkRo 31, EkRo 48, 4:1 5:1, EeRl 40, 2:3 and F12:5 c o n t a i n r e l a t i v e l y simple, heavy t o o l s . In the l a t t e r group of s i t e s , t h i s p a t t e r n i s due to the presence of granular b a s a l t and/or s p a l l t o o l s i n the assemblages. Other s i t e s are not r e a d i l y i n t e r p r e t a b l e , except as being " t y p i c a l " , c l u s t e r i n g along the l i n e of one scar per gram, however s i t e s CR92, EeRj 1, 19:1, F 8 : l and EeRk 7 are e x c e p t i o n a l i n terms of sheer abundance. Thus the r e l a - t i v e l y complex and small assemblages can be i n t e r p r e t e d as the r e s u l t 224 SITE TOTAL TOTAL TOTAL WEIT DOCO COUNT (grams) (#) 14:2 15.0 14 2 16:1 101.1 48 5 19:1 648 .'5 692 56 22:1 114.7 10 2 26:3 1.4 12 4 32:1 152.5 130 13 CR28 15.7 99 5 CR64 0 0 0 CR40 50.7 67 6 CR73 44.8 43 4 ElRw 4 77.5 182 19 CR92 390.1 406 46 EkRo 18 29.4 15 17 EkRo 31 413.0 151 22 EkRo 48 142.4 65 18 2:3 2728.9 399 37 4:2 245.6 250 35 4:5 32.9 99 16 4:1 322.2 78 24 5:1 778.1 180 25 9:1 51.0 88 12 9:2 63.3 72 13 12:6 43.8 114 12 EeRk 16 106.5 120 20 EeRl 41 211.0 231 29 EeRk 7 1114.9 723 116 EeRk 4:38 92.4 96 20 EeRl 40 685.2 366 76 G21:9' 167.0 136 26 G23:l 24.3 • 33 4 G2:12 38.8 78 8 G31:l 166.4 155 22 F8:l 814.3 725 53 F12:5 675.6 120 6 J22:2 3.9 22 2 J38:2 44.5 63 7 K2:l 156.9 132 10 EeRj 1 441.4 385 69 TABLE 28. Total tool weights and scar counts by site. 225 F i g u r e 7 3 . Assemblage t o t a l t o o l weight p l o t t e d a g a i n s t t o t a l t o o l s c a r counts. 226 of highly mobile tasks, or as the depositional l o c i of no longer useful personal gear, while the relatively simple, heavy assemblages can be interpreted as the results of expedient tasks or residential generalized tasks, employing furniture along with expedient tools. One of the problems with Ebert's model is that i t does not consider debitage, which reveal immediate deposition patterns at sites. If we consider that tools, regardless of types, are used and deposited, and that late stage debitage can result from the maintenance of tools, then archaeologically-expected relation- ships between actual numbers of tools at sites, and the relative abundance of late debitage in assemblages, can serve as an infer- ential model of maintenance and curation behavior with more pre- cision than Ebert's model. In this way, we can understand that assemblages resulting from relatively short-term tool maintenance acti v i t i e s should exhibit few tools, and relatively large amounts of late stage debitage, whereas assemblages that are simply short- term manufacturing l o c i should exhibit few tools, since the manu- facturing products should have been then transported to use-locations. Sites with many tools and large amounts of late stage debitage should have resulted from re-occupied tool maintenance locations, and assemblages with many tools and l i t t l e late stage debitage should be the products of long term tool use "locations where tool maintenance was not undertaken. The archaeological situation with respect to the 38 sites i s shown in Figure 74. Here sites occur in more or less discrete clus— 227 • 116 tools EeRk ? • Eagle Lake (ELP) • mouth of the Chilcotin (MOC) A Lillooet (LIL) * Hat Creek (HAC) EeRl 40 EeRj I 19:1 .CR92 ,F-8: 4:2 • G 2 I : 9 X 2:3 .EeRl 41 I 12:6 4:1 • EeRk 4 :38^ EkRo 48 4:5 , 9 : 1 - G3I:I 9:2 • • 3 2 : 1 *K2: I # I6:I # 2 2 : l * FI2:5 X G23:l J22:2 J38:2 CR64 B E k R o 31 ,EIRw 4 " EkRo 18 # C R 4 0 * C R 7 3 . G2:I2 EeRk 16 , C R 2 8 '26:3 # i 4 : 2 — i — 30 - 1 - 15 45 I— 60 % L a t e Debi tage Count F i g u r e 74. Graph o f th e t o t a l number o f t o o l s v s . the p e r c e n t of l a t e d e b i t a g e i n each assemblage. 228 t e r s that are r e a d i l y i n t e r p r e t a b l e i n the terms above. Exca- vated housepit s i t e s EeRk 7, EeRl 40 and EeRj 1 appear at upper l e f t , w i t h many t o o l s , but r e l a t i v e l y l i t t l e l a t e debitage. However, not a l l excavated housepit s i t e s f o l l o w t h i s p a t t e r n ex- pected of long-term residences. S i t e s EkRo 48 and EeRk 4:38 occur at lower l e f t , w i t h few t o o l s and r e l a t i v e l y l i t t l e l a t e debitage, w h i l e s i t e s CR73, EkRo 18, EkRo 31, EeRk 15 and EeRl 41 occur near the lower r i g h t , w i t h few t o o l s , but r e l a t i v e l y h i g h amounts of l a t e debitage. This p a t t e r n i s i n support of i n t e r - p r e t a t i o n s made i n previous s e c t i o n s , that these are d i f f e r e n t from what i s expected on housepit s i t e s , and these are perhaps r e l a t i v e l y short-term occupation h a b i t a t i o n s . S i t e s 12:6, 16:1, 22:1, J22:2, F12:5 and G23:1 appear to be l o c a t i o n s where t o o l making was a p r i o r i t y i n i t s e l f , whereas s i t e s J38:2, CR64, CR73, CR40, G2:12, CR28, 26:3 and 14:2 are l o - c a t i o n s where t o o l maintenance and low d i s c a r d r a t e s occurred i n probably r e l a t i v e l y short periods of time. S i t e s 4:1, 5:1, G31:l, 9:1, 4:5, 9:2, 32:1 and K 2 : l , grouping w i t h EkRo 48 and EeRk 4:38, appear to be occupation l o c a t i o n s , but of sh o r t e r term than 19:1, F 8 : l , CR92, 4:2, G21:9 and 2:3. These l a t t e r s i t e s , by t h e i r po- s i t i o n s on t h i s graph, below EeRk 7, EeRl 40 and EeRj 1, must be considered to have r e s u l t e d from long occupations over s e v e r a l ep- isodes, but i n open-air s i t u a t i o n s . EeRk 16 seems to be a sh o r t e r term housepit s i t e , and housepit s i t e s EeRl 41, and EkRo 31 only s l i g h t l y longer term than EeRk 16, along w i t h l i t h i c s c a t t e r and 229 housepit site ElRw 4. It should be noted that the raw tool frequency values are highly subject to sampling biases, and these are only partially cancelled by using them against debitage percentages. This i s an example of where precise functional tool data is required (but unfortunately not available in this study), for i f standardized measures of functional tool groups were compared to the debitage stage values, then the range of l i t h i c tool related tasks that occur in different kinds of sites would be much better understood. For example, greater or lesser occurrences of chopping, scraping or cutting tools in relation to resharpening stages would yield relative data on the rates at which tools are exhausted in various tasks. Regardless of the sampling bias, I think that the resolution of the patterns is high and would no doubt be increased with f u l l e r samples, especially from the excavated housepit sites. The problem could perhaps be resolved by multiplying the samples obtained by the appropriate portion of site area that they represent, but such requires assuming homogeneity across site areas and would be most reliable i f more sample units were available. In an ideal f u l l sam- ple or equally random sampled set of sites, I would expect that the 0 graph in Figure 74 would sort housepit sites more discretely, but would not substantially alter the interpretations of most of the l i t h i c scatters with or without features. Site 2:3 from the Mouth of the Chilcotin and possibly other grassland sites could be biased by thick grass growth and poor surface v i s i b i l i t y , but as at Eagle 230 Lake and most Hat Creek sites, an attempt was made to collect a l l v i s i b le l i t h i c remains. With the 5 mm cut-off applied in this study, this bias should not be serious. In sum, a model of assemblage formation that is based on the abundance of tools at sites in relation to the amount of main- tenance debitage present appears to have greater interpretive ab- i l i t y than one that is based solely on the size and complexity of tools. This is because immediate deposition processes as revealed in debitage are considered along with tool deposition processes, which theoretically are not as immediate and more influenced by curation and transport. The new model has the a b i l i t y to allow inferences that include the lengths of occupation of sites, where length of occupation includes a l l of the separate durations of site occupations. It is apparent that there is a great deal of overlap in lengths of occupation, by site type, but this is easily explained, since for example a total of 10 separate years of four month housepit occupations (40 month length of occupation) is in this sense equal to 20 occurrences of two month stays at fishing or root gathering camps, where these be annual or multiple annual in nature. The evidence presented in this section demonstrates that tool maintenance and curation factors strongly affect the character of l i t h i c tool and debitage assemblages, and i t also shows that read- i l y interpretable patterns can be obtained here also with relatively simple measures, once-reduction stages are known from debitage. .' 231 Therefore, while sampling biases should be Considered in future studies, the second hypothesis is supported and the value of the experimental study is again evident. 6.5. Hypothesis 3: Settlement Strategy Factors Before I offer site formation interpretations for each site, and prior to discussing the similarities and differences of tech- nological strategies of the four regions of study, I think i t is necessary to consider the general hypothesis that settlement strategies can be predicted on the basis of l i t h i c content of the assemblages. It is hypothesized that the 38 assemblages can be consistently interpreted as resulting from five settlement strategies on the basis of their context and presence of site fea- tures alone: 1. Excavated housepits, which represent winter hab- itations; 2. Lithic scatters without features, which represent short-term occupations, or possibly pre-housepit habitation areas; 3. Lithic scatters with associated housepit features, which rep- resent long-term open air habitation l o c i , perhaps in early hist- oric times when housepits were no longer constructed, yet stone tools were s t i l l used, or representing outdoor activities conducted during winter pithouse occupations; 4. Lithic scatters with assoc- iated cachepits, which represent salmon processing and storage l o c i ; and 5. Lithic scatters with firecracked rock, which represent poss- ibly large mammal and f l o r a l resource processing locations. 232 The analysis requires that tools be considered along with debitage so that a l l aspects of the l i t h i c technology are i n ^ eluded in "predicting" site types, and thus a meaningful class- i f i c a t i o n of tools, in functional terms, is required. This is accomplished by R-mode cluster analysis in so that meaning can be assigned to groups of tool types on the basis of their co- association. The cluster analysis is performed on those tool types and site features that occur greater than five times across the 38 sites, to reduce the probability of spurious associations, and is thus based on 21 classes (see Table 14). The analysis is based on the presence or absence of the classes, and uses Jaccard's Complement (Sneath and Sokal 1973), as a pseudo-distance measure. The Furthest Neighbour clustering routine (Wood 1973) is used to produce the groups of classes. The dendrogram is shown in Figure 75 where four clusters of tools and features are identified. Fea- tures are included with the l i t h i c tools because such was the prac- tice with Matson et ail. (1979), in a similar R-mode analysis. Thus comparisons can be made between the two studies, and since features in general represent more labour input than stone tool manipulation per se, features can be expected to be associated with "labour inten- sive" tools. Cluster I consists of unimarginal fragments, pieces esquillees, bimarginal fragments, utilized BRF's, and lanceolate biface fragments. These can be interpreted as exhausted, fragmented, f u l l y used items that would be incapable of participating further in subsistence tasks, 233 D i s s i m i l a r i t y S c a l e 1.0 0.5 I I I I I I I ! I IV II rC 0.0 III p complete unifaces complete b i f a c e s b i f a c e fragments u t i l i z e d f l a k e s p r o j e c t i l e point fragments uniface fragments f i r e c r a c k e d rock housepits cachepits complete p r o j e c t i l e points s p a l l tools g r a v e r s / d r i l l s complete unimarginals complete large b i f a c e s large b i f a c e fragment hammerstones unimarginal fragments pieces e s q u i l l e e s bimarginal fragments u t i l i z e d b i f a c e reduction flakes lanceolate b i f a c e fragments F i g u r e 75. R-Mode a n a l y s i s o f the p r e s e n c e o r absence o f 21 t o o l c l a s s e s and s i t e f e a t u r e s i n the 38 assemblages. 234 and would not be worth further curation. Cluster IT contains complete and fragmentary large bifaces, and hammerstones. This appears to be a group of large items, possibly reflecting the use of large bifaces as cores for the derivation of useful flakes. This cluster can be interpreted as the closest there is to "site furniture" items that are l e f t at sites because they are too.lbulky to transport, yet that are useful in settlement strategies for par- ticular purposes at either residential or special-purpose sites (Binford 1979). Cluster III is composed of complete projectile points, spall tools, graver/drills, and complete unimarginal tools. These items can be interpreted as large mammal hunting and pro- cessing equipment, and also as "personal gear", that is extensively curated and maintained. Cluster IV contains a sub-cluster of site features, and a larger cluster of complete unifaces, complete b i - faces, biface fragments, utilized flakes, projectile point fragments, and uniface fragments. These are interpreted as general purpose items that are useful in several kinds of tasks, although the pro- j e c t i l e point fragments are d i f f i c u l t to interpret in this sense, and this cluster may contain personal gear that is discarded once i t has been replaced or repaired. Generally, Cluster IV items are those that are present in most assemblages, but the association of the tools with the site features, along with the co-associations of complete and frag- mentary bifaces and unifaces, allows the interpretation that these items result from generalized a c t i v i t i e s . 235 The R-Mode a n a l y s i s by Matson et a l . (1979) at the Mouth of the C h i l c o t i n derived "maintenance", " s p e c i a l i z e d " , and abun- dance f a c t o r s , comparable to those derived above. Their C l u s t e r 4 i s c l o s e to the above C l u s t e r IV, except without s i t e f e a t u r e s , which are c l a s s i f i e d s e p a r a t e l y i n the Mouth of the C h i l c o t i n sample alone, and these are a good "maintenance/generalized" grouping. The above C l u s t e r I exhausted, discarded t o o l s may be p a r t i a l l y subsumed i n Matson et a l . ' s (1979) C l u s t e r IV, and the remaining c l u s t e r s of each study appear mixed, although the sep- arate t o o l t y p o l o g i e s that were a p p l i e d i s an u n c e r t a i n t y . Using the condensed t o o l c l a s s i f i c a t i o n , minus s i t e f e a t u r e s and the condensed debitage c l a s s i f i c a t i o n (Table 29),the propo- s i t i o n of assemblage v a r i a b i l i t y being r e l a t e d to settlement s t r a t - egies i s t e s t e d w i t h m u l t i p l e d i s c r i m i n a n t a n a l y s i s (Klecka 1975) that was introduced i n Chapter 4. In t h i s a p p l i c a t i o n , the known groups are the f i v e s i t e types as i d e n t i f i e d by f e a t u r e s , and a step- wise d i s c r i m i n a n t method (Wilks) i s used.that attempts to i d e n t i f y the s i t e types on the b a s i s of the frequencies of items w i t h i n con- densed t o o l and debitage c l a s s e s . Table 30 shows that o v e r a l l c o r r e c t d i s c r i m i n a t i o n i s achieved at a r a t e of 74%, which i s a good s o l u t i o n , s i n c e t h i s i s 54% above the " p r i o r p r o b a b i l i t y " of accurate c l a s s i f i c a t i o n (Hair et al.1979). The stepwise technique i n d i c a t e s that t o o l c l u s t e r s I I I and b i p o l a r cores are the most important d i s c r i m i n a t i n g v a r i a b l e s , followed by cores, and middle stage debitage. 236 TI . T i l T i l l XIV E M L BR BP BC CO Pl P2 CR73 0. . .0. . .1. .. .3. . . .13 . . .17 . .18. . .5. . . . 0 . . , . 0 . . . .0 .LSFCR LSFCH EkRo 18 3- • -0-• • -0-• • -14-• • -22 • • -11 • -22-• -9-...1.. • - 0 - • • -0 EkRo 31 1- • -0-.. -8-• • -13-• • -25 • • -43 • -14-. -8-. ..7. . • -0 EkRo 48 3- • -1-.. .1. • • -12-• -127 • -107 • -52-•13- •-23- • . .8- • • -0 •LSCP--LSHP EeRk 16 4- • -0-.. .3. • • -13- 4 5 • -15-• -0-••-0-• • -0- • • -0 TJ EeRl 41 0- • -1 • .. .4. ••-22- 4 9 • • -8-• -1-••-0-• . .1.. • -0 EeRk 7 7- • -5-• • -8-••-92- •1385 • -924 •362- •22- . .99. . . -8- • • -2 EeRk 4:38 1- • -0-• ••0-•••19- • • -74 • • -82 • -44-• -8-•.-8- • . .2- • • -0 EeRl 40 2- . .4. • ••8-• • • 60 •  -411 • -540 •262- •32- •-32-• . .4.. • -0 EeR.i 1 9- • -2-• .-6-••-52- • -320 • -224 •206- •31- • -31- • . .4. • • -1 TOTALS 30 13 35 300 2385 1962 1003 206 201 28 3 14:2 1- • -0-• • -0- 1- 3 2 • • -6-. -0-.. .0-• . .0.. • .0 .LSFCR.LSFCR 16:1 1- • -0-...0. 4. •••22 4 .. .1. . .0-...0-• . .0- • • .4 22:1 0- • -0-• • -0- 2- • • -56 • • -21 .. .3. . -0-...0-• . .0- • • -3 4:2 7- • -3-...1. • • -24-• -304 • -355 •252- •22- •-18-• . .2- • • •1 4:5 2- • -0-• • -0-•••14- ••126 • -128 ••76- . .3. ...5.. . .2- • • -0 LSFCR.LSFCR r 1 cn G21:9 6- • -0-• • -0-••-20- • • - 67 ••166 • 114-•12- ...0-• . .0. • • -0 G23:l 0- • -0-• • -0- 4. • -117 • -116 •156- •16- ..-0-• , .3. . • -5 G2:12 0- •-0- • • -0- 7. • --38 • • -96 • 115-• 19-••-0-• , . 1. . • -0 LSFCR-LSFCR- G31:l 0- • -0-.. .3. •••19- .. .99 ... 75 • -78-•15- ...7.. , .3. . , .4 LSFCR F12:5 1- • -1-... 1. 1. ••166 ••116 ••48- . .5. ...5.. , .5- • • • 1 J22:2 1- • -0-•••1- 0- 7 3 ... 1. • - l -. . .0- • • -0- •  -0 LSFCR LSFCR K2:l 1- • -0-• • -0- 9. •297 • -403 •276- • 133 f-27-• • 6- • • -0 TOTALS 20 4 6 105 1302 1485 1126 226 62 22 18 32:1 0-• -0-...3. ...10. • • - 48 • • - 50 . .44. . .4. -16-. L.5.. • -0 EIRw 4 0-• -0-.. .5. •••14- • • 172 ••214 •216- •32- ...7.• ..5.. • .0 4:1 3- • • 1-». .3. •••17- - -36 ... 52 • -26-• • 2-.. .1.. . .3- • . • 2 5:1 3- • -2-u . .5. • • • 12-...23 L..39 • • 20-. .0-... 1.. . .2- • • 0 r 1 9:1 2- - -0-» . -0- 9. * • -39 >--62 • •25-. .0-.. .8- • ..7-. • 0 LSFCR CC 13 9:2 1- • -0-... 1. •••10- .. .49 •••55 ••38- . .4. ... 1.. .. 3- • . 1 12:6 1- • • 1-• • -0-• • •10-• • • 10 L . . - 8 .. - 0-. -0- .4. • ̂ -0 . . . TOTALS 10 4 17 82 377 480 369 42 36 29 3 19:1 5- • • 1-. . .7. ..43. • -383 • -382 • 222-..3. ..53.. • 15- • ••5 r 1 26:3 3- • • 1-• . -o- 0-• • -21 •• - 28 u .41-. .8-. . . o - . •. 0-. L -0 LS LSFCR cn n CR92 8- • • 1-...5. •••29- ••553 ••339 K 267- • 39-. .46- • • 15- • • • 1 2:3 1- • • 5-...9. ••-21- •••23 .. .34 ..39. .-8- • • 10- • ..4.. . .4 TOTALS 17 8 21 93 980 783 569 58 109 34 10 CR28 2- • • 1-• • - 0- 2- • • • 10 6 * • 16-. . 2 . .. . 0- • . • 0- • • • 0 cn CR64 0- - -0-• • - 0- 0- • • • 16 9 . . 14.  • 0-. . . o - - . . 1. . • • 2 LS "1 0 CR40 1-• -0-•--0- 5. •••19 . . . 41 ••42- • 15-..-0-- . 0- • • • 0 pa F8:l 0- • -0-. . . 5. u..48- • • 125 • • 217 • 227-• 55-.. . 5.. . 6- • • • 6 J38:2 0-• -0-. . . 2-U...4. 3 • • • 11 ...7. • • 1-... 0- • . 0- • • 1 TOTALS 3 1 7 59 173 284 306 73 5 7 9 TI = Tool Cluster 1 TIV = Tool Cluster IV L - Late TII = Tool Cluster I I E = Early BR = B i f a c i a l Reduction Flakes T i l l = Tool Cluster I I I M = Middle BP = Bipolar Flakes CO = Cores BC = Bipolar Cores Pl = Predicted Class MDA A l l Sites P2 = Predicted Class MDA F8:l removed. TABLE 29. Data employed i n the settlement component discriminant a n a l y s i s . 237 Predicted Group Membership Actual # of cu cu CO 3 O 33 U O OJ •H 4-1 - C 4 J 4 J ca • H C J HH CO H u 0) cu cu cu 4-1 CO 4 J CO 4 J 4 J 4-1 4-1 M i 4-> O ca •H ca •H ca ca o O. o CM o n O O 0) CU 0 0 C J CO cu o 3 o C J C J U • H o • H ca • H - H JZ W ,c u Xi PM 4 J 4-1 4 J C J •rl —̂  •H •H o r J • J 3 PH HP 10 8 (80%) 0 0 1 (10%) 1 (10%) LS 12 0 7 (58.3%) 0 0 5 (41.7%) LSHP 7 0 0 6 (85.7%) 0 1 (14.3%) LSCP 4 0 1 (25%) 0 3 (75%) 0 LSFCR 5 0 1 (20%) 0 0 4 (80%) Overall Correct Classification: 73.68% TABLE 30. Results of multiple discriminant analysis based on functional tool classes and con- densed debitage classes. 238 Overall, l i t h i c scatters with housepits are most accurately classified (85.7%) followed by housepits and l i t h i c scatters with firecracked rock (each 80%), l i t h i c scatters with cachepits (75%) , and f i n a l l y l i t h i c scatters (.58.3%). The individual classification results are shown in Table 29 also. The stepwise discriminant analysis showed that four functions were derived to discriminate the five groups. Of these, the f i r s t two functions account for 90% of variance of the solution. The highest loading variable in the f i r s t function is tool Cluster III, and for the second function is bipolar cores. The analysis appears to have selected the opposition of highly curated personal gear with early stage bipolar core reduction, to be the most efficient way of distinguishing the site types, but note that classification of unknown sites would require entering as many as five more variables (Tool 1, Tool 4, middle debitage, bipolar flakes and single platform cores) . It is valuable In this case to consider which sites have been misclassified. Table 29 shows f i r s t of a l l , that most misclassifi- cations are l i t h i c scatters, which tend to be classed as l i t h i c scatters with firecracked rock (41.7%). One excavated housepit (CR73) and one l i t h i c scatter with housepits (9:2) are also classified as l i t h i c scatters with firecracked rock. EkRo 48 is classed as a l i t h i c scatter with cachepits, and 26:3 (LSCP) and site CR64 (LSFCR) are classed as l i t h i c scatters alone. It is also important to note that except for housepits, the site type classes are a l l improperly classified into one other class at the most. Yet the housepit class is secure, since no other site classes are improperly classed into i t . 239 The power of the two v a r i a b l e s derived by the d i s c r i m i n a n t a n a l y s i s to demonstrate s i g n i f i c a n t d i f f e r e n c e s among the s e t t l e - ment components i s t e s t e d by the chi-square t e s t of independence, i n contingency t a b l e s i n Table 31. The t a b l e shows that the n u l l hypothesis of no d i f f e r e n c e i n r e l a t i v e proportions of t o o l C l u s t e r I I I and b i p o l a r cores i s a t t r i b u t a b l e to s i t e type, i s r e j e c t e d at the l e v e l of p = .01. The t a b l e a l s o can be read as showing that almost 40% of the chi-square value achieved i s taken up by the HP/BCO c e l l . I t appears that the LSHP, LSCP and LSFCR c e l l s do not c o n t r i b u t e s i g n i f i c a n t l y to the observed t o o l and core f r e - quencies. Again, t h i s a n a l y s i s has f a c t o r e d out extremes, i n house- p i t s i t e s and l i t h i c s c a t t e r s . I t i s thus not s u r p r i s i n g that the chi-square t e s t does not 2 r e j e c t HO at the p = .001 l e v e l (X >18.46 r e q u i r e d ) . Thus, i n t e r - p r e t a t i o n s on the t a b l e cannot be pushed much beyond observing that housepits c o n t a i n more personal gear and fewer b i p o l a r cores than i s to be expected, w h i l e l i t h i c s c a t t e r s without features e x h i b i t l e s s personal gear than i s expected, and more b i p o l a r cores. These f i n d i n g s mesh very w e l l w i t h B i n f o r d ' s (1979) Nunamiut Eskimo expec- t a t i o n s , that personal gear i s e v e n t u a l l y deposited at residences, while the b i p o l a r core f a c t o r i s a c l e a r i n d i c a t i o n of ample use of l o c a l s m a l l m a t e r i a l s at l i t h i c s c a t t e r s i t e s . To f u r t h e r confirm the r e l i a b i l i t y of the f u n c t i o n a l t o o l c l a s s i f i c a t i o n and the condensed debitage c l a s s i f i c a t i o n i n p r e d i c - t i o n of settlement s i t e types, another d i s c r i m i n a n t a n a l y s i s of the 240 TOOL CLUSTER III BIPOLAR CORES Housepits 39 (28. 7) 28 (38 .3) Lit h i c Scatters 6 (12) 22 (12 .6) Lit h i c Scatters with Housepits 17 (19. 7) 29 (26 .3) Lithic Scatters with Cachepits 21 (23. 6) 34 (31 .4) Lithic Scatters with Firecracked Rock 7 (6) 7 (8) 90 120 X 2 = (Q - E ) 2 = 3.69+ 2.76+ 30 + 7.01+ .37+ .28+ .29+ .22+ .17+ .13 2 E 2 X = 17.92, Ho is rejected at p = .01 (X > 13.28) TABLE 31. Chi-square test of independence, five settlement types by personal gear and bipolar cores 241 same f i v e s i t e types i s undertaken. In t h i s a n a l y s i s , s i t e F 8 : l has been removed from the l i t h i c s c a t t e r w i t h f i r e c r a c k e d rock group, and thus only 37 assemblages are included. F 8 : l i s r e - moved here even though i t i s not m i s c l a s s i f i e d i n the f i r s t d i s - criminant a n a l y s i s , because i t c l e a r l y contains a l i t h i c assem- blage i n a s s o c i a t i o n w i t h a r o a s t i n g p i t w i t h p l a n t and mammal remains, i t i s probably on the order of 500 to 1000 years older than most other s i t e s here, and i t s assemblage s i z e renders i t much d i f f e r e n t from other small l i t h i c s c a t t e r s w i t h f i r e c r a c k e d rock. Note that F 8 : l i s removed from t h i s a n a l y s i s completely, and i s not entered as an unknown to see where i t i s c l a s s e d . The r e s u l t obtained i n t h i s second d i s c r i m i n a n t a n a l y s i s , (Table 33), a l s o Wilk's stepwise method, i s indeed cleaner than the f i r s t . Here, o v e r a l l c o r r e c t c l a s s i f i c a t i o n i s 81.08%, and again Tool c l a s s I I I and b i p o l a r cores are the most s i g n i f i c a n t v a r i a b l e s i n the f i r s t two f u n c t i o n s , which account f o r 88% of the v a r i a n c e among the four f u n c t i o n s . Again, most m i s c l a s s i f i - c a t i o n s (Table 29) are i n t o the l i t h i c s c a t t e r w i t h f i r e c r a c k e d rock c l a s s , and most of these are l i t h i c s c a t t e r s without features (14:2, 4:5, G2:12, J22:2). G31:l i s p r o p e r l y c l a s s i f i e d a t t h i s time, and 26:3 i s again m i s c l a s s i f i e d as a l i t h i c s c a t t e r w i t h f i r e c r a c k e d rock. The only other m i s c l a s s i f i c a t i o n i s again EkRo 48, which i n t h i s run i s c l a s s e d as a l i t h i c s c a t t e r w i t h housepits. However, now l i t h i c s c a t t e r s w i t h housepits are a l l c o r r e c t l y i d e n t i f i e d , as are the a c t u a l l i t h i c s c a t t e r s w i t h f i r e c r a c k e d rock. 242 Predicted Group Membership Actual Group # of Cases •H CM OJ CQ O u cu 4 J 4 J cd CJ CO CJ •H Xi 4 J •H n n M T J CU cu CU CU 4 J ca 4 J CO 4 J 4 J 4-1 4 J 4-1 4-1 o cd •H cd •H cd cd CJ CM o CM o u C O CU C O cu CO u ca CU CJ CJ CJ o •H O •H cd •H •rl rC3 .fi u x: CM 4-) 4-1 4-1 o •H •H •H o r J i - J r J 5 PM HP 10 8 (80%) 0 1 (10%) 0 1 (10%) LS 12 0 8 !66.7%) 0 0 4 (33.3%) LSHP 7 0 0 7 (.100%) 0 0 LSCP 4 0 0 0 3 (75%) 1 (.25%) LSFCR 4 0 0 0 0 4 (100%) Overall Correct Classification: 81.08% TABLE 3 2 i Results of multiple discriminant analysis based on functional tool classes and con- densed debitage classes, with F8:1 removed. 243 The chi-square test performed on the data with F8:l removed, by tool Cluster III and bipolar cores (Table 34), shows that the five site types are significantly different with respect to the proportional frequencies of these art- ifact classes. Again, the major portions of the chi-square value are obtained in the top four c e l l s of the table. Also, the chi-square value barely achieves a level significant at p = .01, and does not pass at p = .001. I b e l i e v e that this i s not as important as demonstrating that the directions of variation are consistent with those of the f i r s t discriminant analysis, which they are. The multiple discriminant analyses performed above have their greatest value in overall results, since large-scale patterns are being sought. The analyses do achieve high success rates in assigning assemblages to pre-defined classes on the basis of tool and debitage classes obtained by indepen- dent lines of evidence. The mathematical manipulations required to achieve these results are much more complex than the b i - variate analyses of raw material factors and tool maintenance and curation processes, however, the discriminant analyses op- erate on multiple covariation measures and are thus not as sub- ject to sampling biases. In sum, i t is apparent that settlement categories of sites can be discretely identified by the methods employed in this section of the study. 244 TOOL CLUSTER III BIPOLAR CORES Housepits 39. (.28.62) 28 (38. 38) Lithic Scatters 6 (.11.96) 22 (16. 04) Lithic Scatters w/ Housepits 17 (19.65) 29 (26. 35) Lithic Scatters w/ Cachepits 21 (23.49) 34 (31. 51) Lithic Scatters w/ Firecracked Rock 2 (1.28) 1 . (1. 72) 85 114 X 2 = (0-E) 2 = 3.76+ 2.81 + 2.97 + 2.21 + .36 + .27 + .26 + .20 E + .41 + .30 = 13.55 X 2 = 13.55, Ho is rejected at p = .01 (X 2> 13.28) TABLE 33. Chi-square test of independence, five settlement types without F8:1, by personal gear and bipolar cores. 245 6.6. Assemblage Formation Summaries The preceding analyses have examined the nature of general technological factors that have contributed to interassemblage va r i a b i l i t y , and have successfully related l i t h i c technological processes and patterns of tool deposition to settlement strategies. The presence of features at many of the sites serves as a control factor, through ethnographic analogy, against which the technolog- i c a l and functional variations show consistent patterning. Exca- vated housepit assemblages provide the firmest association of con- text with assemblage deposition, and the va r i a b i l i t y that is ex- hibited within this site type, and the similarity of housepit assemblages to others provides insight to site occupation processes that would be otherwise d i f f i c u l t to infer. Schiffer (1975) has discussed in a theoretical manner, the kinds of behavior that could lead to marked differences in content of sites, when similar activities are undertaken at them, and con- cluded that "occupation span", the length of time that a site is occupied at any one time, and "curate behavior", the removal of artifacts from sites, should be the most important and visible de- terminants of assemblage differences. The methods of analysis em- ployed in this study very closely parallel those that Schiffer (1975: 268) suggested would be of use to the study of occupation span. Schiffer (.1975: 268) proposed that'debitage, ut i l i z e d flakes and waste products of various kinds" are of the greatest u t i l i t y in inferring site functions, and as we have seen, debitage reduction 246 stages, cores and personal gear discards are quite useful. Secondly, Schiffer offered that " i f curate behavior is wide- spread, then the variety of items present at sites should vary with occupation span" (1975: 268). In terms of the assemblage measures employed in this study, i t is apparent that sites vary a great deal in the predominance and intra-assemblage spread of reduction stages, and that some of this v a r i a b i l i t y is due to economizing behavior, where tools of one raw material were replaced with those from other raw materials. Schiffer's (1975) suggestion to study tool "uselives" is only barely considered in this study in the examination of Ebert's (1979) model, yet comparing assemblages in terms of tool quantities and variety of debitage, as Schiffer suggests,does relate to sites' lengths of occupation and perhaps provides the clearest ordering of sites along these lines (.Figure 74). The point that sites' assemblages may be the products of several occupations appears to me to require a refinement of Schiffer's (1975) definition of occ- upation span, where I would define i t as the sum of occupation dur- ations, and not as the duration of single episodes of use. The following summaries present each assemblage within a general group, representing major assemblage formation processes with respect to tool manufacturing stages, tool maintenance and curation, and occupation span as evidenced in the variations of the l i t h i c technological patterns within site types containing cultural depressions or firecracked rock features. 247 1. Long Term Housepits: EeRk 7, EeRl 40, EeRj 1 These three assemblages are distinct in their similiarites to each other in a l l analyses, being abundant, wide-ranging in manufacturing stages, and containing diverse tool types. These assemblages do not exhibit marked patterns of tool curation or conservation, and can be considered to be "typical" assemblages that resulted from repeated winter occupations. 2. Moderate Term Housepits: EkRo 48, EeRk 4:38 These two sites exhibit debitage that is wide-ranging in reduction stage, but that tends to early, and these also con- tain relatively sparse tool assemblages. I would suggest, from the "refuse p i t " context of EeRk 4:38, that both assemblages received their f i n a l character as the result of deliberate dis- posal processes, and not from in-house habitation a c t i v i t i e s . 3. Short Term Housepits: CR73, EkRo 18, EkRo 31, EeRk 16, EeRl 41 These assemblages have predominantly late stage debitage and sparse tool content for housepits. Excavation area sampled is not a factor here, since EeRl 41 is a relatively large area excavation, much larger than EkRo 48 and EeRk 4:38 above. EeRk 16 is somewhat of an anomaly and is perhaps the briefest occupation housepit of the lot. 4. Moderate Term Lithic Scatters: 4:2, 4:5, G21:9, G31:l, K2:l These are relatively abundant assemblages with wide-ranging reduction stages evidenced in the debitage. 4:2, 4:5, G31:l and K2:l exhibit chert tool manufacture and curation, while G21:9 ex- 248 hibits the manufacture and disposal of about equal amounts of vitreous basalt and chert tools. These appear to be sites in the Mouth of the Chilcotin and Hat Creek regions that were re- occupied several times. 5. Short Term Lithic Scatters: 14:2, 16:1, 22:1, G23:l, G2:12, F12:5, J22:2 These sites are of two basic kinds. 1. Late stage/tool maintenance sites 14:2 and G2:12, where granular basalt and chert tools were replaced by vitreous basalt tools, which were then maintained and curated. 2. Early reduction/replacement sites 16:1, 22:1, G23:l, F12:5 and J22:2, where early reduction stages predominate. At 16:1 and 22:1, obsidian and vitreous basalt tools were replaced by granular basalt, and at G23:l, chert tools were made, then exported. At F12:5 and J22:2, only vitreous basalt was employed, but in a replacement situation at J22:2, while F12:5 appears to be a good example of a simple "quarrying/manufacturing" location. 6. Moderate Term Lithic Scatters with Housepits: 32:1, ElRw 4, 4:1, 5:1, 9:1, 9:2 These are relatively abundant surface assemblages with wide- ranging reduction stages, that a l l exhibit the curation of chert or obsidian from them, but not in the extreme. I would suggest that the assemblages result principally from activities that were under- taken prior to winter pithouse occupations, including "gearing up" for long-distance hunts, and the maintenance of the pithouses them- selves . 249 7. Short Term L i t h i c S c a t t e r s w i t h Housepits: 12:6 12:6 i s an assemblage much l i k e 16:1 and 22:1, except w i t h housepits present. In t h i s case, the c u r a t i o n of v i t - reous b a s a l t t o o l s and the import of granular b a s a l t t o o l s does not seem to be associated w i t h the housepit f e a t u r e s , but the s i t e was p r o p e r l y c l a s s i f i e d i n both the MDA analyses, where- as s i t e 9:2 was not. T h i s , along w i t h w i t h presence of s e v e r a l s p a l l t o o l s , lends support to the idea that short-term occu- pations may not leave e n t i r e l y r e p r e s e n t a t i v e m a t e r i a l s behind, and that such i s achieved only w i t h repeated occupations ( S c h i f f e r 1975) . 8. Long Term L i t h i c S c a t t e r s w i t h Cachepits: 19:1, CR92, 2:3 These assemblages e x h i b i t wide ranges of t o o l manufacture, d i v e r s e and abundant t o o l assemblages, and no extreme patterns of t o o l c u r a t i o n or import. These s i t e s .were l i k e l y occupied to process salmon resources, and a l s o l i k e l y served as l a r g e mammal hunting base camps. 9. Short Term L i t h i c S c a t t e r s w i t h Cachepits: 26:3 This assemblage i s i n t e r e s t i n g i n being s i m i l a r i n s e v e r a l respects ( l a t e debitage predominant, improper MDA c l a s s i f i c a t i o n i n t o LSFCR) to s i t e s CR28 and CR40, a l s o from the Eagle Lake sam- pl e (see below). S i t e s 26:3 and CR28 e x h i b i t o b s i d i a n maintenance and export, w h i l e CR40 i s a l o c a t i o n of o b s i d i a n t o o l d e p o s i t i o n . I suggest that the important d i f f e r e n c e i s that 26:3 i s lo c a t e d adjacent to Eagle Lake w i t h c a c h e p i t s , w h i l e CR28 and CR40 are l o - 250 cated next to the Chilko River, with firecracked rock features. If the associations are correct, then perhaps 26:3 resulted from the same kind of aquisition activity at CR28 and CR40, but the resource was cached rather than immediately processed. Unfortu- nately, more information is required of the actual resource being obtained. 10. Long Term Lithic Scatter with Firecracked Rock: F8:l This is a unique assemblage, with abundant tools and late stage debitage in association with a reused roasting p i t . Mult- iple discriminant analysis of settlement components was s i g n i f i - cantly improved when this site was removed from consideration. F8:l appears to be a biface manufacturing location, where chert tools tend to replace vitreous basalt tools. That i s , basalt b i - faces are being l e f t at the site with late stage debitage, and early stage chert debitage is also being deposited. Possibly these patterns each relate to a separate episode of site occupation,,yet I believe that the parsimonious explanation is that upon exhaustion, available basalt resources were replaced by local chert materials. 11. Short Term Lithic Scatters with Firecracked Rock: CR28, CR64, CR40, J38:2 These are small assemblages, each with restricted ranges of debitage reduction stages. CR28 and CR40 emphasize late/maintenance stages, CR64 emphasizes early/core reduction stages, and J38:2 em- phasizes middle reduction stages. A l l of these appear to be the result of single occupations. At CR64, vitreous basalt tools were 251 manufactured and removed, whereas at CR28, obsidian tools were made and exported, at CR40 obsidian tools were maintained and deposited, and at J38:2 chert tools were imported but not main- tained. I suggest that a l l of these sites are related to large mammal procurement and processing. 6.1'. Summary The analyses of inter-regional v a r i a b i l i t y in stone tool and debitage assemblages have yielded results in support of previous research and current theoretical models, and also re- sults that are inconsistent with such. The debitage classification produced in the experimental program of Chapter 4 is of great ut- i l i t y in allowing inferences to be made concerning technological processes of assemblage formation, especially when extremes of the reduction processes are considered in relation to tool occur- rence patterns. As such, the general proposition stated at the outset of this chapter is supported. The f i r s t hypothesis does not fare nearly as well. In a l l four regions, obsidian and chert raw materials appear to have been reduced and used to make tools no differently than vitreous basalt. Overall, this indicates that regardless of source, raw material acquisition w a s n o t a m a j ° r subsistence activity in i t s e l f , but was undertaken during the course of other ac t i v i t i e s . The novel app- roach of comparing the relative amounts of tools and debitage that are made of particular raw materials is a very useful means of in- ferring replacement and curation behaviors, and again is most re- vealing when extremes of the patterns are considered. 252 Tool maintenance behavior is seen as being a major determinant of l i t h i c assemblage var i a b i l i t y . Ebert's (1979) model of the effects of mobility on tools i s not an entirely satisfactory way of accounting for variability, and a refined model that considers the mere amounts of tools in comparison to the amounts of maintenance debitage in assemblages is a much more revealing method of understanding assemblage formation pro- cesses. In particular, this new model appears to be able to gauge the total lengths of time that sites were occupied, but may be sensitive to sampling restrictions. Finally, general settlement strategies can be reliably pre- dicted from l i t h i c assemblages, in a complex mathematical manner. This requires tools to be assigned functional meaning, and also requires debitage reduction stages to be considered simultaneously with the tool types. The site occupation purposes predicted on." the basis of these kinds of variables are of greater precision than those achieved solely on the basis of bivariate tool and debitage variables. 253 CHAPTER 7 SUMMARY AND CONCLUSIONS 7.1. Summary The objective of this study was to examine the nature of l i t h i c assemblage v a r i a b i l i t y in relation to late pre- historic settlement patterns of the Interior Plateau of British Columbia. The research has proceeded with a behavioral perspective that assumffi the major conditioners of assemblage variations are human ac t i v i t i e s . The development of behav- ioral approaches to l i t h i c collections has been reviewed, and shown to have reached a level of sophistication where several models are available for empirical verification. Collins' (1975) general model of the operations of l i t h i c technologies is encompassed by current models of the relationships between stone tools and settlement behavior, especially those of Binford (1979) , Ebert (.1979) , Goodyear (.1979) , and Pokotylo (1978) . These models, varying in explicitness, argue that the mobility of human groups directly and indirectly causes variations in assem- blages and that the operations of settlement systems can be mon- itored by the application of non-arbitrary measures designed to reveal regional spatial variations in manufacturing stages, cur- ation patterns, and disposal processes. 254 The ethnographic l i t e r a t u r e of the I n t e r i o r P l a t e a u immediately r e l e v a n t to the area of study has been reviewed w i t h the o b j e c t i v e of showing that the e a r l y h i s t o r i c C h i l c o t i n and I n t e r i o r S a l i s h had very s i m i l a r l i f e s t y l e s . The C h i l c o t i n , Canyon Shuswap, Upper Thompson and Upper L i l l o o e t hunted and gathered e s s e n t i a l l y the same resources, obtained anadromous salmon as a p r i n c i p a l food supply, had a well-developed storage technology, and wintered i n pithouses. I t i s recognized that the ethnographies do not provide a complete p i c t u r e of pre-contact settlement systems. Nonetheless, they c o n t a i n much i n v a l u a b l e , i f o f t e n i n d i r e c t i n f o r m a t i o n . A separate review of ethnographic records of stone t o o l manufacturing has been i n - cluded here, and again, w h i l e the data are not f u l l y p r i s t i n e , and d e t a i l . • i s a problem, manufacturing techniques and owner- ship of l i t h i c resource l o c a t i o n s have been described w i t h a c l a r i t y equal to that found i n most other North American sources. The development of I n t e r i o r P l a t e a u p r e h i s t o r i c research has focused p r i m a r i l y on c u l t u r e h i s t o r y . Most previous research has sought to d e r i v e c o n s i s t e n t t y p o l o g i c a l patterns of t o o l occurrence w i t h respect to the age of the assemblages. The problems assoc- i a t e d w i t h housepit archaeology, and a l a c k of cave and r o c k s h e l t e r assemblages have s e r i o u s l y hampered c u l t u r e h i s t o r y schemes. Only the l a s t 2000 years of occupation can be r e l i a b l y i d e n t i f i e d . S e t t l e - ment p a t t e r n archaeology of the I n t e r i o r Plateau has a sh o r t e r h i s t o r y than c u l t u r e h i s t o r i c i n v e s t i g a t i o n , but appears to be on a surer 255 methodological footing. Based firmly on the direct historic approach, through the application of direct ethnographic analogy, Interior Plateau settlement pattern research appears to have strong predictive a b i l i t i e s , and the a b i l i t y to test ethnographic models. In part icular, Matson et_ a l . (.1979) argue that the late prehistoric Canyon Shuswap had a highly mobile settlement pattern, in contrast to the "sedentary" pattern that can be inferred from Teit 's (1909a) descriptions. I argue that the evidence indicates settlement behavior that was both mobile and intensive in a re- lat ive ly small area. Also, estimates of housepit and cachepit use-spans were obtained by extrapolating data obtained in the Shuswap Settlement Patterns project. Pokotylo (1978) studied pre- viously unstudied middle elevation environments, using a techno- logical approach to stone tools and debitage to demonstrate sett le- ment strategies analogous to Bonaparte Shuswap and Upper Thompson summer and f a l l subsistence practices. Pokotylo's (1978a) research was innovative in using a large number of surface assemblages, and also in the expl ic i t application of l i t h i c debitage variables to yield important clues to the past operations of mobile group sub- sistence tasks. More recently, the Eagle Lake project (Matson et̂ a l . 1980) was directed at describing the settlement patterns of late prehistoric Chilcotin in environments direct ly comparable to those studied in the Shuswap Settlement Patterns and Hat Creek projects. This research provided preliminary means of identifying the ethnic i identit ies of s i te inhabitants, and also served as a p i lot study for 256 the current study (Magne 1980). One important aspect of the Eagle Lake research was a preliminary i n v e s t i g a t i o n of v a r i a b i l i t y i n l i t h i c debitage that i s produced i n various reduction stages of chipped stone t o o l s . The p i l o t study (Magne and Pokotylo 1981) was much enlarged i n scope and sample s i z e i n the present study, with the purpose of providing a r e l i a b l e means of c l a s s i f y i n g debitage into stages of reduction. This goal was achieved, and i t was found that the weight of flakes i s not a good predictor of reduction stages, and that platform scar counts and dorsal scar counts allow about 80% r e l i a b i l i t y i n stage c l a s s i f i c a t i o n when debitage are sorted into PRB.'s and shatter f l a k e s . B i f a c i a l and bipolar types of reduction are also very d i s c r e t e , and although there are problems with sam- ple s i z e s , vitreous basalt, obsidian and chert raw materials appear to vary i n s i m i l a r fashion. The c l a s s i f i c a t i o n of debitage that i s formulated as a r e s u l t of the experimental program i s considered adequate for the large scale applications i n t h i s study, but i s c e r t a i n l y i n need of independent v e r i f i c a t i o n . While some doubt may be expressed as to the r e l i a b i l i t y of i d e n t i f y i n g middle and l a t e stages of reduction, t h i s problem i s minimized when BRF's are classed separately, and also because early stage flakes appear to be highly d i s c r e t e . C e r t a i n l y , the c l a s s i f i c a t i o n i s not completely foolproof i n that mistakes i n i d e n t i f i c a t i o n w i l l occur, but i n low r e l a t i v e frequency. So long as t h i s i s acceptable and extreme concern with p a r t i c u l a r s i s avoided, 257 then t h i s study has been s u c c e s s f u l . I t h i n k i t i s q u i t e l i k e l y that f u t u r e research employing a s i m i l a r - research design, a l s o w i t h p r e c i s e f l a k e removal c o n t r o l , perhaps new v a r i a b l e s , and greater c o n t r o l over raw m a t e r i a l samples, w i l l enable more p r e c i s e r e c o n s t r u c t i o n of stone t o o l manufacturing behavior. P a r t i c u l a r l y r e q u i r e d are more s t u d i e s of raw m a t e r i a l f a c t o r s , pressure f l a k i n g events, sub-stage v a r i a b i l i t y , use-resharpening stages, microblade manufacture and other s o r t s of s p e c i a l i z e d t o o l manufacture. Experimental work must c l e a r l y continue to enable refinement of the ideas developed here, yet the study i s a precedent i n con- t r o l l e d l i t h i c s experimentation, and the need f o r a c l a s s i f i c a t i o n of t h i s k ind i s witnessed i n i t s a p p l i c a t i o n to assemblages from Texas (Katz, personal communication 1982) , A l b e r t a ( S t r y d , personal communication 1982), northern B. C.(Magne 1982a) and Lower Mainland B. C. (Peacock 1982). The a r c h a e o l o g i c a l assemblages that are analyzed i n t h i s study were c o l l e c t e d i n the Eagle Lake, Mouth of the C h i l c o t i n , L i l l o o e t and Hat Creek regions. In t o t a l , 14,541 f l a k e s of debitage, 164 cores and 861 t o o l s have been examined. D e s c r i p t i o n s of each s i t e , and summary t a b u l a t i o n s of a r t i f a c t frequencies have been provided. The analyses were undertaken to i n v e s t i g a t e three general hy- potheses, using the debitage c l a s s i f i c a t i o n as a u s e f u l means of ob t a i n i n g patterns of assemblage v a r i a b i l i t y that are i n t e r p r e t a b l e . Assemblage v a r i a b i l i t y i n terms of re d u c t i o n stages i s examined by 258 means of mu l t i v a r i a t e c l u s t e r i n g and sc a l i n g techniques, and s i t e s are grouped on the basis of early, middle and l a t e r e - duction stages, while at t h i s point of the study, b i f a c i a l and bipo l a r reduction do not here appear to be important factors, i n v a r i a b i l i t y at the multiregional l e v e l of i n t e r p r e t a t i o n . When s i t e s are grouped on the basis of predominant reduction stages and ,. - presence of absence of housepits, cachepits and firecr a c k e d rock, several i n t e r e s t i n g patterns emerge. Housepits exhibit both wide- ranging and l a t e reduction stages, l i t h i c scatters exhibit e a r l y / core reduction, wide-ranging and late/maintenance patterns, l i t h i c s c atters with housepits exhibit early/core reduction and middle/ wide ranging patterns, and l i t h i c s catters with cachepits and l i t h i c scatters with f i r e c r a c k e d rock exhibit both middle/wide ranging and late/maintenance patterns of stone t o o l manufacture. The f i r s t hypothesis examined i s that chert and obsidian raw materials should exhibit extensive curation and maintenance patterns i n r e l a t i o n to"vitreous basalt, since i n the regions studied, natural sources of cherts and obsidians are r e l a t i v e l y rare. This proposition i s not supported, and i n the Mouth of the C h i l c o t i n region, there i s a s l i g h t tendency f or vitreous basalt materials to be c a r r i e d to l a t e r stages of reduction than chert material.. .However, > the Mouth of the C h i l c o t i n tools, .4l£'.a?-J; of e s s e n t i a l l y comparable s i z e and complexity, regardless of raw mat- e r i a l . This analysis demonstrates that the a c q u i s i t i o n of raw mat- e r i a l s i s l a r g e l y embedded i n other settlement and subsistence ac- 259 t i v i t i e s . This is not to say that tools were not economically made nor curated ; . bivariate graphs of the raw material compo- sition of debitage and tool assemblages demonstrate definite tool curation and tool replacement patterns. The second hypothesis tested is that regardless of raw material factors, curation and maintenance of tools was a major determinant of assemblage composition. This proposition i s par- t i a l l y supported in the raw material bivariate graph analysis, and is also supported in an application of Ebert's (1979) model of tool v a r i a b i l i t y . In this analysis, sites are essentially s p l i t between those with small, complex tools and those with larger^ simpler tools. A new model of assemblage va r i a b i l i t y in relation to group mobility is presented, where debitage figure prominently in relation to the simple abundance of tools in assemblages. I suggest that the total length of time that a site is occupied w i l l determine how much late stage/maintenance debitage, in relation to other debitage, w i l l be deposited, and also that the number of tools deposited at sites, regardless ofirtype, is also determined by length of occupation. This analysis provides groupings of assem- blages that are most interesting in that housepits appear to be of three different sorts: long term, moderate term and short term. Other assemblages are interpreted in similar fashion by their sim- i l i a r i ty to the various housepit assemblages. The f i n a l hypothesis tested is that a set of five site occupation purposes across the four regions can be reliably predicted on the 260 basis of tool and debitage co-associations. The f i r s t step here is to devise a shortened, functionally interpretable tool classification from the original l i s t of tool types. This is accomplished by a presence/absence cluster analysis of most tool types, and four clusters of tools are interpretable, using Bin- ford's (1979) terms of reference. Tool clusters are inferred to be personal gear, site furniture, generalized maintenance tools, and broken, exhausted tools. These tool groups are then combined with a condensed debitage classification as suggested in the reduction factors analysis, and used to predict site types of housepits, l i t h i c scatters, l i t h i c scatters with housepits, l i t h i c scatters with cachepits, and l i t h i c scatters with firecracked rock. An overall success rate of 73.68% accurate classification is achieved with stepwise multiple discrim- inant analysis, that shows peesonal gear and bipolar cores are the most useful variables in the analysis. The significance of these variables is then tested with a chi-square test, which offers sup- port for the more complex mathematical solution of the discriminant analysis. The significance of Binford's suggested variables of personal gear, and core reduction variables are supported in this analysis. Finally, the analysis attempts discriminant analysis of the five site types using the same variables and sites, but without site F8:l. Classification accuracy now rises to 81.08%, and again personal gear and bipolar cores are the significant variables ob-. tained. 261 The most significant findings of the multiregional analyses are the tool curation and replacement patterns evidenced in the raw material analysis, and the occupation duration findings of the evaluation of Ebert's (1979) model. Most findings are in agreement with Binford's (1979) expectations for assemblage varia- b i l i t y based on his Nunamiut Eskimo studies, except that his ex- pectation that residential locations should demonstrate the least amount of var i a b i l i t y (1979: 267) is contradicted here. It is clear in the analyses that excavated housepit assemblages are highly variable. If l i t h i c scatters with housepits are also con- sidered residential sites, then residential assemblages among Interior Plateau groups are much different from Nunamiut Eskimo patterns. If the Interior Plateau peoples were less mobile than the Nunamiut, then more kinds of ac t i v i t i e s , and thus greater varia- b i l i t y , i s expected at residences. To summarize the findings of the multiregional analyses, house- pit v a r i a b i l i t y patterns are used as a "baseline" to group sites by inferred occupation spans and presence or absence of housepit/cache- pit and firecracked rock features. Small, short term sites offer the best evidence of discrete activites, since tool replacement and curation processes are clear when only a few items are l e f t at sites. Large, long term assemblages such as those resulting from extended housepit occupations are essentially a blend of multiple technological and subsistence processes. Short term housepit occu- pations reveal specific instances of the kinds of behavior that re-"" occurred in housepits. This appears mainly to be "gearing up" activity, but housepits also evidence instances of outright garbage disposal. 262 L i t h i c s c a t t e r s r e s u l t i n g from the a c t i v i t i e s a s s o c i a t e d w i t h processing and s t o r i n g salmon and probably l a r g e mammal r e - sources must be considered a d i f f e r e n t s o r t of r e s i d e n t i a l r e s i d u e , but these a l s o evidence extreme p a t t e r n s . Those l i t h i c s c a t t e r s w i t h cachepits that are i n f e r r e d to r e s u l t from short term occu- pations seem to be more c l e a r l y r e l a t e d to those w i t h f i r e c r a c k e d rock features i n that they cont a i n r e l a t i v e l y l a r g e amounts of maintenance r e s i d u e s . In a l l cases, the c u r a t i o n , replacement and r e p a i r of t o o l s made of d i f f e r e n t raw m a t e r i a l s i s a comp l i c a t i n g f a c t o r i n under- standing the e f f e c t s of settlement s t r a t e g i e s on l i t h i c assemblages. I t i s important to know c h a r a c t e r i s t i c s of abandoned versus curated t o o l s . U n f o r t u n a t e l y , time l i m i t a t i o n s precluded a study of the t o o l s that would i d e n t i f y r e l a t i v e s t a t e s of exhaustion. In any event, w h i l e a l l . r a w m a t e r i a l s except granular b a s a l t appear to be eq u a l l y maintained and used to make s i m i l a r kinds of t o o l s , c u r a t i o n and maintenance appear to operate independently of settlement s t r a t - egy, except i n the short term occupation s i t u a t i o n s . 7.2. Conclusions This study has made two major c o n t r i b u t i o n s to current archaeo- l o g i c a l research i n general and to I n t e r i o r Plateau archaeology i n p a r t i c u l a r . The f i r s t i s the demonstration that general manufact- u r i n g stages of chipped stone t o o l s of s e v e r a l forms can be r e l i a - b l y i n f e r r e d from the q u a n t i t a t i v e a n a l y s i s of l i t h i c debitage. The second i s that r e d u c t i o n stage i n f o r m a t i o n i s a very i n f o r m a t i v e 263 means of inferring past processes of assemblage formation due to l i t h i c technology and settlement strategy factors. The re- search shows that there are many cross-regional regularities in such processes, and implies that large scale attempts to derive reliable culture histories can use multiregional data, yet need to consider more f u l l y the kinds of sites that the information is retrieved from. The most reliable data would appear to be in small assemblages, regardless of context. I would suggest that future studies be directed more intensively at small sites, so that the cumulative assemblages at larger sites can be better understood. The point is that since the analyses in the present study have shown l i t h i c technology to be largely embedded in settlement strategies, l i t h i c remains can be expected to change as the operations of settlement and subsistence systems change, and in predictable fashions. The reduction stage classification can be seen as being analogous to Binford's (1978a) " u t i l i t y indices" for caribou anatomy. The reduction stage model and measures enable techno-r logical strategies to be modelled in new ways, and lik e the "bulk" and "gourmet" curves of caribou usage, the curation and replacement graphs for tools and debitage of various raw materials provide fine-grained evidence of the operational characteristics of settle- ment systems in general. I would suggest that future research on the Interior Plateau be directed to providing detailed information on l i t h i c raw mat- er i a l sources, so that patterns of mobility can be tied to constant 264 locations. Secondly, more experimentation along the lines of that in this study needs to be undertaken, tool forms need to be more closely related to debitage va r i a b i l i t y , and raw mat- e r i a l factors need to be more intensively examined. This study has several 'deficiencies that can be corrected with future work at separate locations, and with the assemblages analysed here. The most severe of these concerns data, sampling and the representativeness of the individual l i t h i c assemblages as well as that of the/sites within the separate regions;- and across the central and southern Interior Plateau In general. Some of the interpretations offered here for the transect-collected Hat Creek sites would perhaps be altered with more complete samples. Equally important is the completeness of the regional samples. The Eagle Lake, Mouth of the Chilcotin, and Hat Creek sites were located with regional sampling methods, but the Lillooet assemblages were not. Furthermore, most ELP surface l i t h i c assemblages, and MOC assemblages have been studied, only about 1/20 of the known HAC sites and very few LIL region sites have been analysed here. The rationale in a l l cases was to study sites believed to be late prehistoric or "Kamloops Phase" (less than about ,2000 years) in age, but this is by no means certain for most sites. As for methodological shortcomings, the debitage reduction stage classification relies on a meagre sample of tools in relation to the quantity of archaeological material that was analysed, yet the sample of 2657 experimental flakes i s 18% of the archaeological sam- 265 p i e of 14,541 a r c h a e o l o g i c a l f l a k e debitage. However, the r e l a t i v e p r o p o r t i o n here i s spread across s e v e r a l kinds of t o o l products i n the experiments, and i s thus more r e p r e s e n t a t i v e than any other reported l i t h i c r e d u c t i o n experiment. Time l i m i t a t i o n s and the d e s i r e to i n v e s t i g a t e l a r g e - s c a l e patterns precluded more i n t e n s i v e manipulation of the data base. Two p a r t i c u l a r kinds of a n a l y s i s were not undertaken. The f i r s t i s a t t r i b u t e a n a l y s i s of the t o o l assemblages.. Using the v a r i a b l e s gathered f o r each t o o l , major patterns of v a r i a b i l i t y could have been examined to a l l o w t o o l s to be c l a s s i f i e d on the b a s i s of r e - d u c t i o n and u s e - r e l a t e d f a c t o r s . The second a n a l y s i s that was omitted i s d e t a i l e d examination of the r e d u c t i o n stages evident i n each raw m a t e r i a l at each s i t e . This approach would have t r i - p led the debitage data presented here, and g r e a t l y complicated i n t e r - p r e t a t i o n s . I b e l i e v e that the debitage vs. t o o l raw m a t e r i a l com- p o s i t i o n analyses a l l e v i a t e most of the l a c k of i n f o r m a t i o n that could have been obtained i n such analyses, but t h i s should be r e - garded as a p r o p o s i t i o n f o r f u t u r e study. The research presented i n the preceding pages has shown how a r c h a e o l o g i c a l awareness of meaning i n l i t h i c assemblages has de- veloped, from normative o r i g i n s where inferences were framed i n c u l t u r e - h i s t o r i c a l and organic e v o l u t i o n a r y terms, to current models of assemblage v a r i a b i l i t y i n r e l a t i o n to hunter-gatherer settlement m o b i l i t y . The current study has added evidence of the importance of t o o l maintenance and c u r a t i o n f a c t o r s , such as have been noted i n 266 Mousterian (Fish 1976, Munday 1976),. Acheulian, Archaic (Collins 1974), Paleo-Indian (Goodyear 1979) and the Hat Creek (Pokotylo 1978) assemblages, and has provided an advanced method of reconstructing tool and debitage manufacturing, maintenance and disposal patterns with complete samples -j, of archaeological materials. The Mousterian problem is not unique to European Middle Paleolithic cultures, but is basic to l i t h i c assemblages everywhere. Stone tool remains can be used to inform us of historical, ideological,social and techno- logical processes, but relevance and accuracy in reconstruction require adequate empirical c r i t e r i a . Furthermore, those c r i t e r i a are subject to change, and both new models and proper analytic methods cannot be constructed in isolation from archaeological history i t s e l f . 267 CHAPTER 8 REFERENCES CITED Akazawa, T. 1979 Middle Paleolithic assemblages from Douara Cave.. In Paleolithic site of the Douara Cave and paleo- geography of Palmyra Basin in Syria, Part II, Prehistoric Occurrences and Chronology in Palmyra.' Basin, edited by K. Hanihara and T. Akazawa, pp. 1 - 30. University of Tokyo Press, Tokyo Albright, S.L. 1982 An ethnoarchaeological study of Tahltan subsistence and settlement patterns. M.A. Thesis, Simon Fraser University, Vancouver, B.C. Anderberg, T 1973 Cluster Analysis for Applications. New York: Academic Press. Baker, CM. 1981 A simulative experiment in simple product manufacture. In Modern material culture, the archaeology of us, edited by R.A. Gould and M.B. Schiffer, pp. 247 - 254. Academic Press, New York. Barton, M. 1981 Technologically produced variation in l i t h i c debitage and the predictive a b i l i t y of flake attributes. Paper presented at the 46th Annual Meeting of the Society for American Archaeology, San Diego Beirne, P. and D.L. Pokotylo 1979 Hat Creek Project: inventory, assessment and evalu- ation of the cultural heritage resources. Report to B.C. Hydro and Power Authority, Department of Anthropology and Sociology, University of British Columbia, Vancouver. Binford, L.R. 1962 Archaeology as anthropology. American Antiquity 28: 217 - 225. 268 A consideration of archaeological research design. American Antiquity. 29: 425 - 441 Model-building, paradigms and the current state of Paleolithic research. In An archeological perspec- tive, edited by L.R. Binford, pp. 252 - 295. Seminar Press, New York. Interassemblage v a r i a b i l i t y : the Mousterian and the "functional" argument. In The explanation of culture change, edited by C. Renfrew, pp. 252 - 253. Duckworth, London Forty-seven trips. In Stone tools as cultural markers, edited by R.V.S. Wright, pp. 24 - 36. Australian Institute of Aboriginal Studies, Pre- history and Material Culture Series 12. Dimensional analysis of behavior and site structure: learning from an Eskimo hunting stand. American Antiquity 43: 33 - 361 Nunamiut ethnoarchaeology. Academic Press, New York. Organization and formation processes: looking at curated technologies. Journal of Anthropological Research 3: 255 - 273. Willow smoke and dog's t a i l s : hunter-gatherer settle- ment systems and archaeological site formation. American Antiquity 45: 4 - 2 0 Bones:, ancient men and modern myths. Academic Press. New'York. Behavioral archaeology and the "Pompei premise". Journal of Anthropological Research 37 (195 - 208) . The archaeology of place. Journal of Anthropological Archaeology 1 (1): 5 - 31. 269 B i n f o r d , L.R. and S.R. B i n f o r d 1966 A p r e l i m i n a r y a n a l y s i s of f u n c t i o n a l v a r i a b i l i t y i n the Mousterian of L e v a l l o i s f a c i e s . In Recent s t u d i e s i n Paleoanthropology, e d i t e d by J.D. C l a r k and F.C. Howell, pp. 238 - 295. American Anthropol- o g i s t 68 (1) P a r t 2. B i n f o r d , L.R. and G. Quimby 1963 Indian s i t e s and chipped stone m a t e r i a l s i n the n o r t h - ern Lake Michigan area. F i e l d i a n a - Anthropology 36: 277 - 307. B i n f o r d , S.R. and L.R. B i n f o r d 1969 Stone t o o l s and human behavior. S c i e n t i f i c American 220: 70 - 84. Bishop, C.A. 1974 The northern Ojibwa and the f u r trade: a h i s t o r i c a l and e c o l o g i c a l study. H o l t , Rinehart and Winston, Toro.nto Boas, F. 1980 The Shuswap. Second general report on the Indians of B.C. , S'ixth:report on the north-western t r i b e s of Canada. B r i t i s h A s s o c i a t i o n f o r the Advancement of Science, Leeds Meeting: 80 - 95. Bonnichsen, R. 1977 Models f o r d e r i v i n g c u l t u r a l i n f o r m a t i o n from stone t o o l s . N a t i o n a l Museum of Man Mercury S e r i e s , Arch- a e o l o g i c a l Survey of Canada Paper 60. Ottawa. Borden, C. 1952a R e s u l t s of a p r e l i m i n a r y survey of the Nechako r e s e r v o i r i n west c e n t r a l B.C. Unpublished m.s. on f i l e , H eritage Conservation Branch, V i c t o r i a . 1952b Results of archaeological investigations in central British Columbia. Anthropology in British Columbia 3: 31 - 43. 1952c A uniform s i t e d e s i g n a t i o n scheme f o r Canada. Anthropology i n B r i t i s h Columbia 3: 44-48. . 270 Borden, C. 1969 Early population movements from Asia into western North America. Syesis 3: 7 - 43 1979 Peopling and early cultures of the Pacific North- west Science 203: 963 - 971. Bordes, F. 1961 Typologie du Paleolithique Ancien et Moyen. Publi- cations de prtlhistoire de l'Universite de Bordeaux. Memoire 1. 1969 The old Stone Age. McGraw-Hill. Toronto. 1972 A tale of two caves. Harper and Row. New York. Bordes, F. and F. Prat 1965 Observations sur les faunes du Riss et du Wurm I en Dordogne. L'Anthropologie 69: 31 - 45 Bouchard, R. and D. Kennedy (Editors) 1979 Shuswap stories. B.C. Indian Language Project Victoria. CommCept Publishing. Vancouver. Bradley, B. 1975 Lithic reduction sequences: a glossary and discussion. In Li'thic technology, making and using stone :tools, edited by E.H. Swanson, pp.5 - 14.Aidine, Chicago Brow, J.B. 1972 A product of the cross-cultural study of ethnocentrism, Shuswap of Canada. Human Relations Area Files Book No. ECNEI 2-001. Burton, J.B. 1980 Making sense of waste flakes: new methods for invest- .... . .igating the technology and economics behind chipped stone assemblages. Journal of Archaeological Science 7: 131 - 148. Butler, R.B. 1961 The Old Cordilleran Culture in the Pacific Northwest. Occasional Papers of the Idaho State College Museum 9. Pocatello. 271 Cahen, D., L. Keeley and F. Van Noten 1979 Stone tools, toolkits, and human behavior in prehistory. Current Anthropology 20: 661 - 683 Callahan, E. 1977 Variability in the early stages of manufacture of Virginia fluted points: an experimental study. Unpublished M.A. thesis, Catholic University of America. Washington Campbell, J. 1968 T e r r i t o r i a l i t y among ancient hunters. In Anthropological Archaeology in the Americas, edited"by B.J. Meggars, pp. 1-21. Anthropological Society of Washington. Washington. Carlson, C. 1980 Excavations at the Curr Site. In the archaeology of Kamloops, pp. 87 - 123. Simon Fraser University Department of Archaeology Publication 7. Burnaby. Chandler, S. and J. Ware 1976 The identification of technological v a r i a b i l i t y through experimental replication and empirical multivariate analysis. Newsletter of Lithic Technology 25 (3): 24 - 46. Chapman, R.C. 1977 Analysis of the l i t h i c assemblages. In Settlement and subsistence along the Lower Chaco River: the CGP survey, edited by C.A. Reher, pp- 371 - 452. University of New Mexico Press. Albuquerque. Charlton, T.H. 1981 Archaeology, ethnohistory, and ethnology: interpretive interfaces. In Advances in archaeological method and theory Vol. 4, edited by M.B. Schiffer, pp. 129 - 176 Academic Press, New York. Choquette, W. 1982 The role of l i t h i c raw material studies in Kootenay archaeology. In Fragments of the past: British Columbia archaeology in the 1970's. B.C. Studies Special Issue, edited by K.R. Fladmark, pp. 21 - 36. 272 Clark, G.A. 1982 Quantifying archaeological research. In Advances in archaeological method and theory Vol. 5, edited by M.B. Schiffer, pp. 217 - 273. Academic Press, New York. Collins, M.B. 1974 A functional analysis of l i t h i c technology among prehistoric hunter-gatherers of southwestern France and Western Texas. Ph.D. dissertation, University of Arizona. University Microfilms No. 74 - 17184. 1975 Lithic technology as a means of processual inference. In Lithic technology: making and using stone tools, edited by E.H. Swanson, pp. 15 - 34. Aidine, Chicago. Cowgill, G. 1968 Archaeological applications of factor, cluster and proximity analyses. American Antiquity 33: 367 - 375. Crabtree, D. 1966 A stoneworker's approach to analysing and replicating the Lindenmeier Folson. Tebiwa 9: 3 - 39. .1968 . Mesoamerican polyhedral"cores'and prismatic blades. American Antiquity 33: 446 - 478 1972 An introduction to flintworking. Occasional Papers of the Idaho State University Museum 28. Pocatello Cranny, M. 1982 A reexamination of the Borden excavations at Chinlac. Unpublished m.s. Department of Anthropology and Socio- logy, University of British Columbia, Vancouver. Cybulski, J.S., D.E. Howes, J.C. Haggarty and M. Eldridge 1981 An early human skeleton from south-central British Columbia: dating and bioarchaeolgogical inference. Canadian Journal of Archaeology 5: 49 - 59. Danner, W.R. 1970 Cherts and jaspers of the Western Cordilleran eugeo- syncline of southwestern British Columbia and north- western Washington. Western Commemorative Volume: 533 - 553. 273 Davis, D.D. (Editor) 1978 Lithics and subsistence: the analysis of stone tool use in prehistoric economies. Vanderbilt Univer- sity Publications in Anthropology 20. Dawson, G.M. 1891 Notes on the Shuswap people of British Columbia. Royal Society of Canada, Transactions, Section II, pp. 2 - 44. Dawson Brothers. de Mortillet, G. 1869 Essai d'une classification des cavernes et des stations sous abris fondee sur le produit de 1'Industrie humaine. Comptes Rendus, Academie des Sciences: 583 - 587 Dincauze, D.F. 1978 CA Comment on: A history of flint-knapping experi- mentation, 1838 - 1976, by L.L. Johnson. Current Anthropology 19: 360 Donahue, P. 1973 Ulkatcho: an archaeological outline. Syesis 6: 153 - 178. 1975 A catalogue of surface collections from north-central British Columbia. Unpublished m.s. on f i l e , Heritage Conservation Branch, Victoria. 1977 4500 years of cultural continuity on the central Interior Plateau of British Columbia. Ph.D. dissertation, University of Wisconsin. University Microfilms No. 78-4412 Doran, J. and F. Hodson 1966 A d i g i t a l computer analysis of Paleolithic f l i n t assemblages. Nature 210: 87 - 89. Duff, W. 1951 Carrier. Unpublished f i e l d notes on f i l e , Ethnology Division, British Columbia Provincial Museum, Victoria. 1964 The Indian History of British Columbia. Vol. 1: The impact of the white man. Anthropology in British Col- umbia Memoir 5. 274 Ebert, J. 1979 An ethnoarchaeological approach to reassessing the meaning of var i a b i l i t y in stone tool assemblages. In Ethnoarchaeology: Implications of ethnography for archaeology, edited by C. Kramer, pp. 59 - 73. Columbia University Press, New York, Eldridge, M. 1974 Recent archaeological investigations near Chase, B.C. Cariboo College Papers in Archaeology 2. Kamloops. Euler, R.C. and G.J. Gumerman (Editors) 1978 Investigations of the Southwestern Anthropological Research Group. The Proceedings of the 1976 Con- ference. Museum of Northern Arizona. Flagstaff Farrand, M. 1898 The Chilcotin. British Association for the Advancement of Science 68: 645 - 648 1900 Traditions of the Chilcotin Indians. American Museum of Natural History Memoir 4, Anthropology Vol. 3, Part I, Jesup North Pacific Expedition Vol. 2, Part 7 New York Fish, P.R. 1976 The interpretive potential of Mousterian debitage. Ph.D. dissertation, Arizona State University. 1979 The interpretive potential of Mousterian debitage. Arizona State University Anthropological Research. Papers 16. 1981 Beyond tools: Middle Paleolithic debitage analysis and cultural inference. Journal of Anthropological Research 37: 374 - 386. Fladmark, K.R. 1976 Punchaw village: preliminary report, archaeology of a prehistoric settlement. Department of Archaeology Current Research Reports 3, edited by R.L. Carlson, pp. 19 - 32. Simon Fraser University. 1982a An introduction to the prehistory of British Columbia. Canadian Journal of Archaeology 6: 95 - 156 275 Fladmark, K.R. 1982b Glass and Ice: A report on the archaeology of the Mount Edziza and Spectrum ranges, northwestern British Col- umbia. Unpublished m.s. Heritage Conservation Branch, Victoria. 1982c Microdebitage analysis: i n i t i a l considerations. Journal of Archaeological Science 9: 205 - 220 Flenniken, J. 1978 Reevaluation of the Lindenmeier Folsom: A replicative experiment in l i t h i c technology. American Antiquity 43: 437 - 480. 1981 Replicative systems analysis: A model applied to the vein quartz artifacts from the Hoko River site. Washington State University Laboratory of Anthropology, Reports of Investigations 59. Freeman, L.G. 1966 The nature of Mousterian Facies in Cantabrian Spain. In Recent studies in paleoanthropology, edited by J.D. Clark and F.C. Howell, pp. 9-21. American Anthropologist 68 (2) Part 2. 1978 The analysis of some occupation floor distributions from earlier and middle Paleolithic sites in Spain. In Views of the Past, edited by L.G. Freeman, pp. 57 - 116. Aldine, Chicago. Gardener, W.M. 1976 Paleo-Indian to Early Archaic: continuity and change into eastern North America during the late Pleistocene and early Holocene. Proceedings of the IXth Inter- national Congress of Prehistoric and Protohistoric Scientists, Colloque XVIII pp. 20-46. Nice. Germann, C. 1979 Cariboo regional archaeological impact assessment, 1979. m.s., Heritage Conservation Branch, Victoria. Goldman, I. 1940 The Alkatcho Carrier of British Columbia. In Accultur- ation in Seven American Indian Tribes, edited by R. Linton, pp. 333 - 389. Appleton-Century, New York.. 276 Goodyear, A. 1979 A hypothesis for':the use. of cryptocrystalline raw . materials among Paleo-Indian groups of North America. Institute of Archaeology and Anthropology, Research Manuscript series .56. University of South Carolina. Gould, R.A. 1968 Chipping stone in the Outback. Natural History 77: 42-49, 1978 Beyond analogy in ethnoarchaeology. In Explorations in ethnoarchaeology, edited by R.A. Gould, pp. 249 - 293. University of New Mexico Press. 1979 Cariboo hunters. Review of Nunamiut ethnoarchaeology, by L.R. Binford, Science 204: 737 - 739 1980 Living Archaeology. Cambridge University Press. Gould, R.A., D. Koster and A. Sontz 1971 The l i t h i c assemblage of the Western Desert Aborigines of Australia. American Antiquity 36: 149 - 169. Gould, R.A. and M.B. Schiffer (Editors) 1982 Modern material culture: the archaeology of us. Academic Press, New York. Greaves, S, 1982 Upon the point: a preliminary investigation of ethnicity as a source of metric variation in l i t h i c projectile points. National Museum of Man Mercury Series, Archaeological Survey of Canada Paper No. 109. Gumerman, G.J. (Editor) 1971 The distribution of prehistoric population aggregates Prescott College Anthropological Papers 1. Gunn, J. 1975 Idiosyncratic behavior in chipping style: some hypotheses and preliminary analysis. In Lithic technology, making and using stone tools, edited by E.H. Swanson, pp. 35-62. Aidine, Chicago. Hair, L.F., R.F. Anderson, R.L. Tatham and B.J. Grablowsky 1979 Multivariate data analysis. PPC Books, Tulsa 277 Ham, L. n.d. Report on excavated faunal remains from the Lillooet region, B.C. Unpublished m.s., Department of Anthro- pology and Sociology, University of British Columbia, Vancouver. 1975 Shuswap settlement patterns. Unpublished M.A. thesis, Department of Archaeology, Simon Fraser University, Burnaby. Hanihara, K. and T. Akazawa (Editors) 1979 Paleolithic site of the Douara Cave and paleogeography of Palmyra Basin in Syria. Part II: Prehistoric OCCT? urrences and chronology in Palmyra Basin. University of Tokyo Press, Tokyo. Harrington, C.R., H.W. Tipper and R.J. Mott 1974 Mammoth from Babine Lake, British Columbia. Canadian Journal of Earth Sciences IT: 285 - 303 Hayden, B. 1976 Curation: old and new. In Primitive art and technology, edited by J.S. Raymond, B. Loveseth, C. Arnold, and G. Reardon, pp. 47 - 59. Department of Archaeology, University of Calgary. 1977 Stone tool functions in the Western Desert. In Stone tools as cultural markers, edited by R.V.S. Wright, pp. 178 - 188. Australian Institute of Aboriginal Studies, Prehistory and Material Culture Series 12. Canberra. 1979a Paleolithic reflections: l i t h i c technology of the Australian Western desert. Australian Institute of Aboriginal Studies. Canberra. 1979b (Editor) Lithic use-wear analysis. Academic Press, New York. 1980 Confusion in the bipolar world: bashed pebbles and splintered pieces. Lithic technology 9 (1): 2 - 6 . 1981 Research and development in the Stone Age: techno- logical transitions among hunter-gatherers. Current Anthropology 11: 519 - 548. 278 Hill-Tout, C. 1905 Report on the ethnology of the StlatlumH of British Columbia. Journal of the Anthropological Institute of Great Britain and Ireland 35: 126 - 218 Holland, S.S. 1964 Landforms of British Columbia: a physiographic out- line. B.C. Department of Mines and Petroleum Resources. Bulletin 48. Holly, G.A. and T.A. Del Bene 1981 An evaluation of Keeley's microwear approach. Journal of Archaeological Science 8: 337 - 352. Holmes, W.H. 1890 A quarry workshop of the flaked stone implement makers in the District of Columbia. American Anthropologist 3: 1-26. Honigmann, J. 1954 The Kaska Indians: en ethnographic reconstruction. Yale University Publications in Anthropology 51. Jelinek, A. 1978 Form, function and style in l i t h i c analysis. In Cultural change and continuity, edited by C.E. Cleland, pp. 19-53. Academic Press, New York. Johnson, L.L. 1978 A history of flint-knapping experimentation, 1838 - 1976. Current Anthropology 19: 337 - 372 Jorgensen, J.S. 1969 Salish language and culture. Indiana University Pub- lications, Language Science Monographs. Vol. 3 1980 Western Indians. W.H. Freeman. San Francisco Katz, P. 1976 A technological analysis of the Kansas City Hopewell chipped stone industry. Ph.D. dissertation, University of Kansas. University Microfilms No. 77 - 16283. 279 Keddie, G. 1972 The Chilcotin archaeological and ecological survey project. Unpublished m.s. Heritage Conservation Branch, Victoria. Keeley, L.H. 1980 Experimental determination of stone tool uses, a micro- wear analysis. University of Chicago Press. 1981 Response to Holly and Del Bene. Journal of Archaeological Science 8: 348 - 352. 1982 Hafting and retooling: effects on the archaeological record. American Antiquity 47: 798 - 809. Kennedy, D. and R. Bouchard 1978 Fraser River Lillooet: an ethnographic summary. In Reports of the Lillooet archaeological project, edited by A. Stryd and S. Lawhead pp. 22 - 25. National Museum of Man-, Mercury Series, Archaeological Survey of Canada Paper 73. Ottawa Kimball, L. 1980 A quantitative pattern recognition model of temporal var i a b i l i t y in debitage for early Archaic through Historic Cherokee l i t h i c assemblages in the lower L i t t l e Tennessee River valley. Paper presented at the 37th Annu'al Meeting of the Southeastern Archaeological Conference, New Orleans. Klecka, W.P. 1975 Discriminant analysis. In SPSS: Statistical package for the social sciences, edited by N.H. Nie, C.H. Hall, J.G. Jenkins, K. Steinbrenner, and D.H. Bent, pp. 434 - 467. McGraw-Hill, Toronto. Knudson, R. 1973 Organizational v a r i a b i l i t y in the late Paleo-Indian Assem- blages . Ph.D. dissertation. Washington State University. Kobayashi, H. 1975 The experimental study of bipolar flakes. In Lithic technology: making and using stone tools, edited by E.H. Swanson, pp. 115 - 127. Aldine, Chicago. Kramer, C. 1979 Ethnoarchaeology: implications of ethnography for archaeology. Columbia University Press, New York. 280 Kuhn, T. 1962 The structure of sci e n t i f i c revolutions. Inter- national Encyclopedia of Unified Science Vol. 2 (2). Lamb, K. (Editor) 1960 The letters and journals of Simon Fraser 1806 - 1808. MacMillan, Toronto Lane, R. 1953!- Cultural relations of the Chilcotin Indians of west central British Columbia. Ph.D. dissertation, University of Washington. University Microfilms No. 1970 - 5296. 1981 Chilcotin. In Handbook of North.Amerlean Indians. Vol. 6, Subarctic, edited by J. Helm, pp. 402 - 412. Smithsonian Institution, Washington. Laville, H. 1973 The relative position of Mousterian industries in the climatic chronology of the early Wurm in the Perigord. World Archaeology 4: 323 - 329. Leach, B.F. 1969 The concept of similarity in prehistoric studies, a test case ..using New Zealand stone flake assemblages. Anthropology Department, University of Otago, Studies in Prehistoric Archaeology 1. Leacock, E. 1973 The Montagnais-Naskapi band. In Cultural ecology, edited by B. Cox, pp. 81 - 100. Steward and McClelland, Toronto. Learning, S. 1971 Rock and mineral collecting in British Columbia. Geo- logical Survey of Canada Paper 72 - 53. Lee, R.B. and I. DeVore (TMitors) 1968 Man the hunter. AidIne, Chicago 1976 Kalahari hunter-gatherers. Harvard University Press. London 281 Magne, M. 1978 Variability among four Archaic l i t h i c assemblages in the Porcupine Mountain region, Manitoba. Unpublished M.A. thesis, University of Manitoba. 1979 Lithic technological resources of the Hat Creek Valley region, B.C. Unpublished m.s. Department of Anthro- pology and Sociology, University of British Columbia, Vancouver. 1980 Comparison of l i t h i c debitage from "Athapaskan" and "Salish" sites. In The Eagle Lake project: report on the 1979 season, by R.G. Matson, M. Magne, D. Ludowicz and D.L. Pokotylo, pp. '67 - 133. Unpublished m.s. Archaeology Laboratory, University of British Columbia. 1982a Heritage resource inventory, construction materials exploration areas and selected ancillary f a c i l i t i e s Stikine-Iskut hydroelectric development. ARESCO Ltd. report to B.C. Hydro and Power Authority, Vancouver. 1982b Technological analysis of Telep site debitage. Appendix III in The Telep site: a late autumn fish camp of the Locarno Beach culture type, by W. Peacock. - Report to Heritage Conservation Branch, Victoria. 1983 Review of: Archaeological investigations at Cache Creek (EeRh 3), by R. Whitlam. Canadian Journal of Archaeology 6: 229 - 233. Magne, M. and R.G. Matson 1982 Identification of "Salish" and "Athapaskan" side-notched projectile points from the Interior Plateau of Bri t i s h Columbia. In Approaches to Algonquin Archaeology, edited by M. Hanna and B. Kooyman, pp. 57 - 79. Univer- sity of Calgary Archaeological Association. Magne, M. and D.L. Pokotylo 1980 Lithic reduction sequences; a controlled experiment. In The Eagle Lake Project: Report on the 1979 season, by R.G. Matson, M. Magne, D. Ludowicz and D.L. Pokotylo, pp. 134 - 166. Unpublished m.s. Archaeology Laboratory, University of British Columbia, Vancouver. 282 Magne, M. and D.L. Pokotylo 1981 A pilot study in b i f a c i a l l i t h i c reduction sequences. Lithic Technology X (2-3): 34 - 47. Matson, R.G. 1971 Adaptation and environment in the Cerbat Mountains, Arizona. Ph.D. Dissertation, University of California, Davis. 1978 T-Scale Scaling Routine. University of British Col- ombia, Vancouver. 1980 The proper place of multivariate techniques in archaeology. American Antiquity 45: 340 - 344. Matson, R.G. and D.L. True 1974 Site relationships at Quebrada . Tarapaca, Chile: A comparison of clustering and scaling techniques. American Antiquity 39: 51 - 74. Matson, R.G., L. Ham and D. Bunyan 1979 Prehistoric settlement patterns at the Mouth of the Chilcotin River, B.C. Occasional Papers of the British Columbia Provincial Museum. in press. Matson, R.G., M. Magne, D. Ludowicz and D.L. Pokotylo 1980 The Eagle Lake project: Report on the 1979 season. Unpublished m.s. Archaeology Laboratory, University of British Columbia, Report to Social Sciences and Humanities Research Council of Canada, Ottawa. Mellars, P. 1965 Sequence and development of Mousterian traditions in southwestern France. Nature. 2 - 5 : 626 - 627. 1970 Some comments on the notion of functional va r i a b i l i t y in stone tool assemblages. World Archaeology 2: 74-89. Mendenhall, W. 1975 Introduction to probability and st a t i s t i c s . Duxbury Press, North Scltuate. Mitchell, D. 1969 Excavations on the Chilcotin Plateau: three sites, three phases. Northwest Anthropological Research Notes 4: 99 - 116. 283 M i t c h e l l , D. 1970 A r c h a e o l o g i c a l i n v e s t i g a t i o n s on :the C h i l c o t i n P l a t e a u , 1968. Syesis 3: 45 - 66 . Montgomery, P. 1978 Stone a r t i f a c t s from the Punchaw Lake s i t e (Area C): a l a t e p r e h i s t o r i c occupation i n c e n t r a l B r i t i s h Columbia. Unpublished M.A. t h e s i s . Simon Fraser U n i v e r s i t y . Morice, A. 1893 Notes a r c h a e o l o g i c a l , i n d u s t r i a l and s o c i o l o g i c a l on the Western Denes. Transactions of the Royal S o c i e t y of Canada, 1892 - 1893. 1906 The h i s t o r y of the northern i n t e r i o r of B r i t i s h Columbia 1660 - 1880. Ye Galleon Press, Washington. Munday, F. 1976 I n t e r s i t e v a r i a b i l i t y i n the Mousterian of the C e n t r a l Negev. In P r e h i s t o r y and paleoenviionments i n the C e n t r a l Negev, I s r a e l . V o l . 1. The Avat/Aquv area. Part 1, e d i t e d by A.E. Marks, pp. 113 - 140. I n s t i t u t e f o r the Study of Earth and Man. Reports of I n v e s t i g a t i o n s 2. Southern Methodist U n i v e r s i t y Press. A t e c h n o l o g i c a l a n a l y s i s of the e a r l y stages i n the manu- f a c t u r e of l i t h i c a r t i f a c t s . Unpublished M.A. t h e s i s . Idaho State U n i v e r s i t y . Muto, G. 1971a 1971b A stage a n a l y s i s of the manufacture of chipped stone implements. In Great Basin A n t h r o p o l o g i c a l Conference 1970: Selected Papers, e d i t e d by M.C. A i t k e n s . Univer-'. s i t y of Oregon A n t h r o p o l o g i c a l Papers 1: 109 - 117. Nechako V a l l e y H i s t o r i c a l S o c i e t y 1979 Vanderhoof: the town that couldn't w a i t . D.W. F r e i s e n , Cloverdale. Melson, E. and G. W i l l 1976 Obsidian sources i n the Anahim Peak area. In Department of Archaeology Current Research Reports, e d i t e d by R.L. Carlson, pp. 151 - 154. Simon Fraser U n i v e r s i t y Depart- ment of Archaeology P u b l i c a t i o n No. 3. 284 Newcomer, M.H. 1971 Some quantitative experiments -in hand-axe manufacture. World Archaeology 3: 85 - 93. Odell, G. 1977 The application of micro-wear analysis to the l i t h i c component of an entire prehistoric settlement: methods, problems, and functional reconstructions. Ph.D. Disser- tation, Harvard University. University Microfilms No 77-27, 354 1980 Toward a more behavioral approach to archaeological l i t h i c concentrations. American Antiquity 45: 404 - 431. Patterson, L.W. and J.B. Sollberger 1978 Replication: and classification of small size debitage Plains Anthropologist 23: 103 - 112. Peacock, W. 1982 The Telep site: a late autumn fis h camp of the Locarno Beach culture type. Unpublished m.s. Report to the Heritage Conservation Branch, Victoria. Phagan, C. 1976 A method for the analysis of flakes in archaeological assemblages: a Peruvian example. Ph.D. Dissertation, Ohio State University. University Microfilms No. 76- 24666. Pokotylo, D.L. 1978a Lithic technology and settlement patterns in Upper Hat Creek Valley, B.C. Ph.D. Dissertation, University of British Columbia, Vancouver. 1978b Preliminary results of a probability surface survey and archaeological impact assessment in the Upper Hat Creek Valley, B r i t i s h Columbia. Datum. Heritage Con- servation Branch, Victoria. 1979a An archaeological impact assessment of the proposed Hat Creek thermal electric development: phase I, In Annual Report.for the year 1976: activities of the Provincial Archaeologist's office of British Columbia and selected research reports, edited by B.O. Simenson, pp. 100-156. Queen's Printer, Victoria. 285 Pokotylo, D.L. 1981 Towards an understanding of prehistoric upland settle- ment behavior in the British Columbia southern Interior Plateau. In Networks of the past: regional interaction i... in archaeology, edited by P. Francis, F. Kense and P. Duke, pp. 379 - 396. University of Calgary Archaeolog- i c a l Association. Pokotylo, D.L. and P. Beirne 1978 Hat Creek Project: preliminary inventory, assessment and evaluation of the cultural heritage resources. Report to B.C. Hydro and Power Authority. Department Of Anthropology and Sociology, University of British Columbia, Vancouver, Pokotylo, D.L. and P. Froese ,/Beirne) .. • „ 1983 The archaeological evidence for root gathering in prehistoric plateau subsistence - settlement patterns: a case study from the Upper Hat Creek Valley, B.C. m.s. submitted to Canadian Journal of Archaeology. Raab, L.M., R.F. Cande and D.W. Stahle 1979 Debitage graphs and Archaic settlement patterns in the Arkansas Ozarks. Mid-Continental Journal of Archaeology 4: 167 - 181. Ray, V.F. 1942 Culture element distributions XXII: Plateau. University of California Anthropological Records 8 (2) . Berkeley. Richards, T. 1978 Excavations at EeRl 171, a preliminary report. Unpublished m.s. Heritage Conservation Branch, Victoria. Rogers, E. 1973 The quest for food and furs: the Mistassini Cree, 1953 - 1954. National Museum of Canada Publications in Ethnology 5. Ottawa Roper, D.C. 1979 Archaeological survey and settlement pattern models in central I l l i n o i s . I l l i n o i s State Museum, Scientific Papers Vol. XVI: Midcbntinental Journal of Archaeology, Special Paper 2. 286 Salmon, M.H. 1982 Philosophy and archaeology. Academic Press. New York Sanger, D. 1963 Excavations at Nesikep Creek (EdRk:4), a stratified site near Lillooet, British Columbia. National Museum of Canada Bulletin 193: 130 - 161. 1966 Excavations in the Lochnore-Nesikep Creek locality, B r i t i s h Columbia: interim report. National Museum of Canada, Anthropology Papers 12. 1968 Prepared core and blade traditions in the Pacific Northwest. Arctic Anthropology 5: 92 - 120. 1969 Cultural traditions in the interior of British Columbia. Syesis 2: 189 - 200. 1970 The archeaology of the Lochnore-Nesikep localit y , British Columbia. Syesis 3, Supplement 1. Schiffer, M.B. 1972 Archaeological context and systemic context. American Antiquity 37: 156 - 164. 1975 The effects of occupation span on site content. In The Cache River archaeological project, compiled by M.B. Schiffer and J.H. House. Arkansas Archaeological Survey Research Series 8: 265 - 269. 1976 Behavioral archaeology. Academic Press. New York. Sheets, P.D. 1975 Behavioral analysis and the structure of a prehistoric industry. Current Anthropology 16: 369 - 391. Singer, C.A. and J.E. Ericson 1977 Quarry analysis at Bodie H i l l s , Mono Country, California: a case study. In Exchange systems in prehistory, edited by J.K. Earle and J.E. Ericson, pp. 171 - 188. Academic Press, New York. 287 Smith, H.I. 1899 Archaeology of Lytton, British Columbia. American Museum of Natural History, Jesup North Pacific Ex- pedition, Memoirs, Vol. I, part 3: 129 - 161 1900 Archaeology of the Thompson River region, British Columbia. American Museum of Natural History, Jesup North Pacific Expedition Memoirs, Vol. 2, part 6: 401 - 442 Sneath, P. and R. Sokal 1973 Numerical Taxonomy. W.H. Freeman, San Francisco Sollberger, J. and L. Patterson 1979 The myth of bipolar flaking. Newsletter of Lithic Technology 5 (3): 21 - 22. Speth, J. 1972 Mechanical basis of percussion flaking. American Antiquity 37: 34 - 60. 1975 Miscellaneous studies in hard-hammer percussion flaking: the effects of oblique impact. American Antiquity 40: 203 - 307. Stafford, CR. and B.D. Stafford 1979 Some issues concerning Lithic Technology VIII the design of l i t h i c experiments. (.2) : 21 - 24 Stahle, D.W. and J.E. Dunn 1981 An experimental analysis of the size distribution of waste flakes from biface reduction using constrained ' least squares analyses. Paper presented at the 46th Annual Meeting of the Society for American Archaeology, San Diego Steward, J.H. 1973 The direct historical approach to archaeology. In Ev- olution and ecology by J.C.S. Steward and R.F. Murphy, pp. 201 - 207. University of I l l i n o i s Press. Urbana. Stryd, A.H. 1970 Preliminary report on the 1970 archaeological excavations near Lillooet, British Columbia, m.s. on f i l e , Heritage Conservation Branch, Victoria. 288 . Stryd, A.H. 1971a The 1971 archaeological excavations at Lillooet, British Columbia, m.s. on f i l e , Heritage Conservation Branch, Victoria 1971b An archaeological research design for the investigation of pit house archaeology. In Aboriginal man and environ- ments oh the Plateau of northwest America, edited by A.H. Stryd and R.A. Smith, pp. 36 - 43. The'Student's Press, University of Calgary. 1972 Salvage archaeology at Lillooet, British Columbia; sites. EeRl 40 and EeRl 41. m.s. on f i l e , Heritage Conservation Branch, Victoria. . 1973a The later prehistory of the Lillooet area, British Columbia. Unpublished Ph.D. Dissertation, University of Calgary. 1973b Culture history of the Canadian plateau. Paper presented to the 6th annual meeting of the Canadian Archaeological Association. Burnaby. 1973c Lillooet archaeological project (1973 season) m.s. on f i l e , Heritage Conservation Branch, Victoria. 1978 An introduction to the Lillooet archaeological project. In reports of the Lillooet archaeological project, edited by A.H. Stryd and S. Lawhead, pp. 1-21. National Museum of Man, Mercury Series, Archaeological Survey of Canada Paper 73 1980 A review of the recent activities undertaken by the Lillooet archaeological project. The Midden XII(2): 5-19. Stuiver, M. 1978 Radiocarbon timescale tested against magnetic and other dating methods. Nature 273: 271 - 274. Teit, J. 1900 The Thompson Indians of British Columbia. American Museum of Natural History, Jesup North Pacific Expe- dition, Memoirs, Vol. 2, Part 4: 163 - 392. 289 Teit, J. 1906 The Lillooet Indians. American Museum of Natural History, Jesup North Pacific Expedition, Memoirs, Vol. 2, Part 5: 193 - 300. 1909a The Shuswap. American Museum of Natural History, Jesup North Pacific Expedition, Memoirs, Vol. 2, Part 7: 447 - 789. 1909b Notes on the Chilcotin Indians. In the Shuswap. American Museum of Natural History, Jesup North Pacific Expedition, Memoirs, Vol. 2, Part 7: 758 - 789. Thomas, D.H. 1971 Prehistoric subsistence-settlement patterns of the Reese River Valley, Central Nevada. Ph.D. Disser- tation, University of California, Davis. Torgerson, W. 1958 Theory and methods of scaling. John Wiley. Toronto Tringham, R. 1978 Experimentation, ethnoarchaeology and the leapfrogs in archaeological methodology. In Explorations in ethnoarchaeology, edited by R.A. Gould, pp. 169 - 199. University of New Mexico Press. Albuquerque. Turner, N. 1978 Food plants of British Columbia Indians. Part 2. Interior Peoples. British Columbia Provincial Museum Handbook 36. Victoria. 1979 Plants in British Columbia Indian Technology. British Columbia Provincial Museum Handbook No. 38. Victoria. Turner, N., R. Bouchard and D.I.D. Kennedy 1980 Ethnobotony of the Okanagan-Colville Indians of British Columbia and Washington. Occasional Papers of the British Columbia Provincial Museum 21. Victoria. Tyhurst, R. 1982 An ethnohistory of the Chilcotin people, unpublished manuscript. University of British Columbia, Vancouver. 290 Van Noten, F., D. Cahen and L. Keeley 1980 A Paleolithic campsite in Belgium. Scientific American 242: 48-55. Von Krogh, H. 1980 Archaeological investigations at the Flood and Pipeline sites, near Hope, B.C. Heritage Conservation Branch Occasional Paper 4, Victoria. Whitlam, R. 1976 Archaeology in the Williams Lake area, British Columbia. Occasional Papers of the Archaeological Sites Advisory Board of British Columbia 1, Victoria. 1980 Archaeological investigations at Cache Creek (EeRh 3). Occasional Papers of the Heritage Conservation Branch 5, Victoria. White, J.P. and D. Thomas 1972 What mean these stones? In Models in archaeology, edited by D.L. Clarke, pp. 275 - 308. Methuen, London. White, J.P., N. Modjeska and I. Hipuya 1977 Group definitions and mental templates. In Stone tools as cultural markers, edited by R.V.S. Wright, .pp. 380 - 390. Australian Institute of Aboriginal Studies, Pre- history and Material Culture Series 12. Canberra Willard, I. 1979 Hunting methods. In Shuswap stories, edited by R. Bouchard and D. Kennedy, pp. 139 - 150. CommCept Publishing, Vancouver. Wilmeth, R. 1969 The Potlatch site, British Columbia: a preliminary report. NeVsletter of the Kansas Anthropological Association 14 (7) . 1970 Chilcotin archaeology. Canadian Archaeological Association, Bulletin 2: 42 - 43. 1971 Historic Chilcotin archaeology at Anahim Lake, British Columbia. In Aboriginal man and environments on the Plateau of Northwest America, edited by A.H. Stryd and R. Smith, pp. 56-59, University of Calgary Archaeological Association 291 Wilmeth, R. 1975 The protohistoric and historic Athapaskan occupation of B r i t i s h Columbia: the archaeological evidence. Western Canadian Journal of Anthropology 5(344):4-20. 1977 Pit-house construction and the disturbance of str a t i f i e d sites. Canadian Journal of Archaeology 1: 135 - 150. 1978 Anahim Lake archaeology and "the early historic Chil- cotin Indians. National Museum of Man Mercury Series, Archeological Survey of Canada Paper No. 82. Wilson, R. 1980 The archaeology of Kamloops, Simon Fraser University Department of Archaeology Publication 7. Wobst, H.M. 1979 Computers and coordinates: strategies for the analysis of Paleolithic stratigraphy. In Computer graphics in Archaeology. Edited by S. Upham, pp. 61 - 67. Anthropological Research Papers 15, Arizona State University. Wood, J. 1979 HCLUS-Hierarchical clustering routines programs. Museum of Northern Arizona. Yellen, J. 1977 Archaeological approaches to the present. Academic Press. New York Zipf, G.K. 1949 Human behavior and the principle of least effort. Addison-Wesley, New York. 292 APPENDIX I. ANTHROPOLOGY 406 Reduction Experiment Outline October 1980 • 293 Anthropology 406 A n a l y t i c a l Techniques i n Archaeology: R. G. Matson, P r o f e s s o r LABORATORY ASSIGNMENT -OUTLINE - October 27 LITKIC REDUCTION ANALYSIS (with Marty Magne, ANSO 0307) The purpose of t h i s lab assignment i s to teach you ways of r e c o g n i z i n g stages i n the r e d u c t i o n - manufacture of chipped stone t o o l s . We w i l l be conducting c o n t r o l l e d experiments i n t o o l making and debitage recovery, using methods I have pre- v i o u s l y used and found to be very i n f o r m a t i v e . By t h i s time, you should be s u f f i c i e n t l y f a m i l i a r w i t h chipped stone t o o l making to be able to understand the impor- tance of p l a t f o r m p r e p a r a t i o n , d e c i s i o n s to use s o f t or hard hammers, how to remove t h i c k spots from b i f a c e s , and the s l i g h t d i f f e r e n c e s i n technique required to work o b s i d i a n and b a s a l t . I f you f e e l you are s t i l l having problems that are not simply r e l a t e d to l a c k of acquired s k i l l , f o r example, i f you don't under- stand the mechanical l o g i c behind p l a t f o r m p r e p a r a t i o n , then please do not h e s i t a t e to consult myself or Dr. Matson. The goal of these experiments i s to provide i n f o r m a t i o n towards i n c r e a s i n g the r e l i a b i l i t y of r e c o n s t r u c t i n g t o o l manu- f a c t u r i n g stages using l i t h i c debitage, and at the same time de- creas i n g the amount of time required to undertake debitage a n a l y s i s . T his assignment r e q u i r e s you to undertake an i n i t i a l step of the ex- periments - I w i l l be t a k i n g the a n a l y s i s to f u r t h e r and f i n a l steps. 294 Outlined i n the following pages are the kinds of tools I want you to make, the procedures you are to follow while making the tools and recovering the debitage, and the c l a s s i f i c a t i o n of debitage that w i l l complete your p a r t i c i p a t i o n i n the experiments. A. Tools to make: Each person should t r y to make at least two of the following t o o l s : one from obsidian, the other ifrom basalt. Please, no mini-tools r e s u l t i n g from multiple errors i n manu- facture. In t h i s business, knowing where to stop i s j u s t as im- portant as knowing where to begin. Tool Code Description UF Flake, u n i f a c i a l l y retouched along one s t r a i g h t margin. BF Flake, b i f a c i a l l y retouched along one s t r a i g h t margin EF Flake, made into an endscraper, u n i f a c i a l l y retouched at least along one end which i s convex i n plan view. UC Ovoid uniface, c i r c u m f e r e n t i a l l y retouched. BC Ovoid b i f a c e , c i r c u m f e r e n t i a l l y retouched, BL Lanceolate b i f a c e , extensively flaked, thinned PP Stemmed or notched p r o j e c t i l e point, t h i s need not be too complex BP Bipolar core - using b i p o l a r reduction, remove flakes which you f e e l would be s u i t a b l e f or use as c u t t i n g / w h i t t l i n g t o o l s . PE Piece e s q u i l l e € - make tools you f e e l would be s u i t a b l e f o r use as wedges used to s p l i t open wood or bone materials.. 295 B. Procedures: 1. Core Reduction: I. Select basalt and obsidian cores, 1 of each large enough to supply you with the material needed to make the tools, plus, two small pebbles for bipolar reduction. II. Weigh, measure, and draw the cores. Include a scale on your drawings, but do not attempt too much detail. III. Lay out a clean canvas tarp or plastic over which to do the flaking. At this point teams w i l l be made up con- sisting of two people each - one to do the flaking, another to recover each flake upon i t s removal and place the flakes from each reduction event in individual cardboard trays. Each team should also have an assortment of hammerstones, antler hammers, leather pads, goggles, (to be worn by both team members), small cardboard trays, and recording forms. The forms are to be used by the knapper, to indicate by flake number at which point in the manufacturing process he/she feels they are changing technique or moving to a distinct new stage of manufacture. There is also ample room for rough notes detailing d i f f i c u l t i e s , changes in hammer type, etc. It is quite probable that at some time during the ex- periment, the person recovering debitage w i l l not be sure of the order of removal of certain flakes. The best way to solve this problem is to place the flakes back on the core or blank being reduced, but do not s t a l l for long trying to 296 f i g u r e t h i s out. The knapper should t r y not to over- load the recovery person. Again, note any t r o u b l e you have, on the forms. I t i s a l s o l i k e l y that some blows w i l l remove more than one f l a k e simultaneously, or that f l a k e s w i l l break upon removal. Here assignment may be a r b i t r a r y . At the completion of core r e d u c t i o n , each team should have s e v e r a l stacks of t r a y s (do not p i l e them so high they t i p o v e r ) , ordered f i r s t to l a s t from bottom to top, w i t h s l i p s of paper i n the t r a y s numbering the f l a k e s . Be sure to i d e n t i f y your stacks by your l a s t name. 2. Blank Reduction I . From the core debitage, s e l e c t the blanks you intend to use f o r f u r t h e r r e d u c t i o n , w r i t i n g down which f l a k e s you have removed. I I . Weigh, measure and draw these blanks. In drawing, concentrate on a c c u r a t e l y o u t l i n i n g f l a k e scars on the d o r s a l faces of the.blanks. Do not attempt any kind of shading even i f you are a g i f t e d a r t i s t . I I I . Reduce each blank to the des i r e d t o o l form, using the two-man procedure. Use the forms again to note any pl a t f o r m p r e p a r a t i o n , type of percussor, pressure f l a k e r s , e t c . I f breakage occurs, do not attempt to salvage the piece 297 unless I t i s q u i t e l a r g e , but keep the debitage produced up to that p o i n t i n t a c t - i t i s s t i l l u s e f u l i n f o r m a t i o n . IV. At the end >of t h i s part of the experiment, each team should have a set of t r a y s f o r each t o o l produced, each set of t r a y s c l e a r l y l a b l l e d as to knapper, t o o l , and o r - der of f l a k e removal, and a set of forms d e t a i l i n g the knapping methods. C. Cataloguing: Now you w i l l have to catalogue the m a t e r i a l s , so that they can be used i n the l a t t e r part of t h i s a s s i g n - ment w i t h no f e a r of l o s i n g provenience. Cataloguing should proceed us i n g a set of codes, as f o l l o w s : B a s a l t Event o b s i d i a n f l a k e number I I I f M: B: UF: 36: 1 n n M: 0: C: 110 / / / ° R : / / your ID t o o l f l a k e no. ID Core (Magne) ( u n i f a c i a l w i t h i n event r e d u c t i o n r e t . f l a k e ) Catalogue only those f l a k e s greater than 5 mm. i n any dimension, on t h e i r v e n t r a l f a c e s . Flakes smaller than 5mm. but l a r g e r than 2 mm. should be i n d i v i d u a l l y bagged w i t h catalogue numbers placed on a piece of paper. Flakes smaller than 2 mm. can be catalogued together, by core or blank from which they were produced. 298 D. Analysis - (The results of this are not used in the dissertation) Once everyone has completed a set of tools and debitage, the next step w i l l be to swap debitage (not tools) with another person from another team. I. Take the debitage given to you, and sort the debitage into flakes with remnant striking platforms (PRB's), and flakes without striking platforms (Shatter). II. Weigh each piece of debitage to the nearest gram, and sort the debitage into the following classes: less than 1 - 2 2 - 5 5 - 1 0 greater than 1 gm. grams grams grams 10 grams. SHATTER PRB's III. Count and weigh the debitage falling-into each class. IV. Using raw counts and weights, and relative measures such as percentages or indices, what inferences can you make concerning the kinds of tools made, the stages of reduction represented, and techniques used? 299 Completing the "Reduction Recording Form" Event: The use of these rows and columns should become .clear;.in the reading of the remainder of t h i s note. B r i e f l y , whenever you s t a r t a new, stage i n t o o l manufacture, OR change technique, c i r c l e the a p p r o p r i a t e f l a k e number, which should be a v a i l a b l e fromi.the debitage recovery person. Stage: In gene r a l , f o l l o w C o l l i n s ' (1975) use of the terms "primary" and "secondary" trimming. I suggest the f o l l o w i n g stage c u t - o f f s f o r thetools you are going to be making: UF: Stage 1: Retouch the f l a k e along one margin. COMPLETE BF: Stage 1: Retouch the f l a k e b i f a c i a l l y ' along one margin. COMPLETE EF: Stage 1: Retouch the f l a k e along the r i g h t and l e f t l a t - e r a l margins to produce an elongate, symmetrical form i n p l a n view. Stage 2: Choose e i t h e r the d i s t a l or p r o x i m i n a l end to retouch u n i f a c i a l l y to an edge which i s convex i n plan view. COMPLETE1, (Note: endscrapers o f t e n have i n t a c t s t r i k i n g p l a t f o r m s , w i t h the "scraper" formed a t the d i s t a l end of the f l a k e , but choose whichever end seems e a s i e s t . ) UC: Stage 1: Retouch the f l a k e on a l l margins u n i f a c i a l l y . COMPLETE .300 BC: Stage 1: Retouch the flake on a l l margins unifacially. Stage 2: Retouch the flake on the opposite face, on a l l margins. COMPLETE BL: Stage .1: Retouch the flake on whatever margins or faces required to produce a generalized lanceolate form i n plan view. Do not bother here with thinning procedures. Platform Preparation may be required (see Below). Stage 2: Using appropriate platform preparation (SEE BELOW), remove flakes required to thin the biface,while retaining a lanceolate outline. Stage 3: Straighten edges, align and sharpen point, pre- pare platforms as required. (Note: you may find that you need more "stages" to complete your b i - face. Describe these in "additional comments".) PP: Stage 1: Retouch a flake along whatever margins, or faces required to produce a triangular form in plan view. This flake should be f a i r l y thin to begin with. Use platform preparation as required. Stage 2: Remove flake required to thin the flake blank b i - fa c i a l l y . Use appropriate platform preparation. f Stage 3: Make notches, stem using pressure flaking. This requires careful isolation of flake platforms. Stage 4: Straighten edges, pressure flake the faces of the point. COMPLETE (Note: Again, you may find that 301 you go through more stages than outlined here, but please try to keep i t simple. For a good idea of how far one can go in deta i l - ing stages of manufacture of complex items, see: Flenniken, J. Jeffrey; "Reevaluation of the Lindenmeier Folsom: A Replication Experi- ment in Lithic Technology." American Antiquity 43 (3): 473 - 480. 1978. Don't even try to copy Flenniken!)v BP: Stage 1: Seat the pebble on a firm anvil, preferably with a " p i t " so the cobble w i l l not s l i p . Strike the proximal "end of the pebble with a hard hammer, remove flakes. Stage 2: The pebble can be rotated, or more blows can be directed from the same orientation as in Stage 1. Continue u n t i l you can no longer hold the core for fear of damaging your fingers. You might try to think of ways the core could be held with no danger of harming yourself. Use here the "other" column in "technical details", to mark those flakes you think would be useful as blanks for other tools. PE: Stage 1: Use the bipolar technique described above, but this time your intention is to form a tool that can be used as a wedge. Stage 2: Any retouch you need to straighten the edges of the tool. 302 3. Technical Details: Technical details are to be checked off in rows corresponding to the "flake numbers" of flakes pro- duced while the particular detail i s operative. These de- t a i l s are not mutually exclusive; usually several columns / w i l l be checked off for any single flake number. For example, i f an antler b i l l e t i s used to remove flakes from the proximal end, dorsal surface of a flake blank, then the three columns "soft-hammer", "proximal mar.", and "dorsal" would be checked off. Any techniques you used that are not covered here, can be added in either of the three columns l e f t in "other". Hard-hammer: Using a stone to remove flakes by percussion. Soft-hammer: Using an antler b i l l e t to remove flakes by pressure. Pressure: Using a pointed antler tool to remove flakes by pres- sure. Platform Preparation: The terms "platform preparation" encom- pass- .several ways of modifying flake blank (or "preform") edges to provide more secure platforms for either percussion or pres- sure flaking. Edges can be abraded or "scrubbed" unifacially or b i f a c i a l l y with a rough stone, starting at one end of the blank and working to the other, or circumferentially; individual platforms can be "strengthened" for pressure flaking or the removal of thick spots by removing material which overhangs the dorsal face of the flake you intend to remove, using either a stone or antler. Crabtree (1972: 84) defines platform preparation as follows: 303 " The grinding, polishing, facetting, bevelling of that part of the platform to receive the applied force. Usually done to strengthen the platform in order to carry off a l a r - ger flake." Isolated: This is meant to be in opposition to "circumfer- ential", or lateral or end margins when these are used to i n - dicate that a technique has been applied a l l along that par- ticular margin. For example, i f you are removing a thick spot with soft-hammer percussion, and that thick spot i s on the dis- t a l margin, then the columns "soft-hammer", "isolated", and "distal mar." w i l l be checked off. Right margin: This refers to the right margin of the flake blank when the ventral face of the flake is facing you. Can be used alone to indicate that the particular technique was applied along the margin, or in conjunction with "isolated" to indicate that the technique was applied to a specific loca- tion. Left margin: Similar to above, but referring to l e f t margin of the blank when i t s ventral face is facing you. Proximal margin: The end with the striking platform, or in the case of flake shatter, the end of the flake where the plat- forms should be, as indicated by ripples or what is l e f t of the bulb of percussion. Distal margin: Similar to above, but refers to the end opposite the platform. 3 0 4 C ircumf er entIal: This column is to be checked when you apply a technique to a l l the edges of a flake blank. Dorsal: Refers to the dorsal face of the flake blank; the face bearing evidence of previous flake removals. Ventral: Refers to the ventral face of the flake blank; the face that i s fresh from the core; bears no evidence of previous flake removals, and exhibits the bulb of percussion, ripples and perhaps eraillure flakes or hackles. Thinning: This column i s to be checked off whenever you are attempting, by percussion or pressure, to conciously thin the cross-section of any of the tools. Notching: Check this column off when you start to produce the notches or stem of your projectile point, by either per- cussion or pressure flaking. Other: There are here three potential columns that you can use to indicate techniques that are not covered here. Check with me or Dr. Matson before you try anything too original. # of Flakes: The number of debitage items produced each time the "core" i s struck, or eachi:time reduction of some sort is even, attempted. Only flakes greater than 5 mm. should be counted. 305 ANTH. 406 REDUCTION RECORDING FORM Knapper_ Recorder Item being Reduced (Material,.Core or Blank No.) .Item being Manufactured (Tool Code) Reduction Event Number Circle Row ' Number 1 1 21 2 22 l3 23 4 24 5 25 6 26 7 27 8 28 9 29 10 30 11 31 12 32 13 33 14 34 15 35 16 36 17 37 18 38 19 39 20 40 Stage (Check off stage i n i t - iation in same row as flake number 1 2 3 4 5 Technical Details 0) u PH o 4H U cd r H PH 60 U cd S 4-> Xi 60 • H PH Other tu o cd PH r H CO U •M a QJ > 60 c ti • H Xi H 60 C • H O +J O a to cd r H PHI Flake Blank Orientation ^ Platform'^/ > PROXIMAL^ } 4 DISTAL ̂Ue+Ac** Additional Comments: PUBLICATIONS 1983 (with B. Penn) FIintknapping: a hobby for s u r v i v a l . B.C. Outdoors August 1983, pp. 49-50, 55. 1982 Review of Whitlam, R.G. Archaeological investigations at Cache Creek (EeRh 3), 1980. Occasional Papers of the Heritage Conservation Branch No. 5. In Canadian Journal of Archaeology 6: 229-233. 1982 (with R.G. Matson) I d e n t i f i c a t i o n of "Salish" and "Athapaskan" p r o j e c t i l e points from the Interior Plateau of B r i t i s h Columba. In Approaches to Algonquian Archaeology, edited by M. Hanna and B. Kooyman, pp. 57-59. University of Calgary Archaeological Assocation. 1981 (with D.L. Pokotylo) A p i l o t study in b i f a c i a l l i t h i c reduction sequences. L i t h i c Technology 10 (2-3): 3 4 - 4 7 . 1980 (with C.T. Shay) FdMg 5: A Blackduck workshop in the Porcupine Mountains. Manitoba Archaeological Quarterly 4 (1): 14 - 25. 1978 Archaeological research in the Porcupine H i l l s , Manitoba. Archae-Facts 5 92-3): 14 - 15. 1978 Comments on the experimental method in cultural anthropology. MASA 4: 24 - 33. University of Manitoba. 1977 A h o l i s t i c pun - halfism. MASA 3: 1. University of Manitoba. 1976 Forecast. Scarborough Fair 3: 53. University of Toronto Press.

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
Canada 18 4
United States 14 1
Iran 8 0
Israel 1 0
Mexico 1 0
City Views Downloads
Unknown 10 4
Tiran 8 0
Vancouver 5 1
Victoria 4 0
Cincinnati 2 0
Indianapolis 2 1
Huntsville 2 0
Hollywood 2 0
Sunnyvale 2 0
Ashburn 2 0
Nelson 1 0
Clarks Summit 1 0
Edmonton 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}

Share

Share to:

Comment

Related Items