This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
.~.
An Obsession with Meaning: A Critical Examination of the Pictograph Sites of the Lake of the Woods
Alicia]. M. Colson Department of Anthropology
Mc Gill University
June 2006
A Thesis submÏtted to McGill University in partial fulfi1ment of the requirements of the degree of Doctor of Philosophy
NOTICE: The author has granted a nonexclusive license allowing Library and Archives Canada to reproduce, publish, archive, preserve, conserve, communicate to the public by telecommunication or on the Internet, loan, distribute and sell theses worldwide, for commercial or noncommercial purposes, in microform, paper, electronic and/or any other formats.
The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission.
ln compliance with the Canadian Privacy Act some supporting forms may have been removed from this thesis.
While these forms may be included in the document page cou nt, their removal does not represent any loss of content from the thesis.
L'auteur a accordé une licence non exclusive permettant à la Bibliothèque et Archives Canada de reproduire, publier, archiver, sauvegarder, conserver, transmettre au public par télécommunication ou par l'Internet, prêter, distribuer et vendre des thèses partout dans le monde, à des fins commerciales ou autres, sur support microforme, papier, électronique et/ou autres formats.
L'auteur conserve la propriété du droit d'auteur et des droits moraux qui protège cette thèse. Ni la thèse ni des extraits substantiels de celle-ci ne doivent être imprimés ou autrement reproduits sans son autorisation.
Conformément à la loi canadienne sur la protection de la vie privée, quelques formulaires secondaires ont été enlevés de cette thèse.
Bien que ces formulaires aient inclus dans la pagination, il n'y aura aucun contenu manquant.
Errata:
Appendix 6 commences on page 551 instead of page 550 as stated in the Table of Contents. Page 550 do es not exist.
Table of Contents Abstracts .............................................................................................................................................. vi Acknowledgements ........................................................................................................................ vüi Note on spelling .............................................................................................................................. xüi Note on Kl.EIO lAS ...................................................................................................................... xiv List of Figures ................................................................................................................................... xv List of Tables ................................................................................................................................. xxiv Chapter 1 - Introduction ....... ....................................... , ................................................................... 1 Where are these sites? .......................................................................................................................... 1 Overview of rock image studies ......................................................................................................... 2 The use of informants ......................................................................................................................... 4 \Vhat is an interpretive appraach? ..................................................................................................... 5
The culture-historical approach (Level 1): .......................................................................... 6 The contextual approach (LeveI2): ...................................................................................... 6 The intuitive approach (Level 3): .......................................................................................... 7 The analogical approach (Level 3): ....................................................................................... 9 The homogical approach (Level3): .................................................................................... 10
The debate between the "lumpers" and the "splitters" ................................................................ 11 The structure of this thesis: ............................................................................................................... 16 Chapter II - Lake of the Woods: Regional Setting of the Sites ......................................... 17 Section 1: Geohistory and geography of the region ...................................................................... 17
(a) Geology ............................................................................................................................. 17 (b) Changing lake levels ........................................................................................................ 23
Section 2: History and ethnography ................................................................................................ 27 (a) Who lived in this regian during the early historical times? ........................................ 28 (b) The unresolved debate between historians and archaeologists over ethnic identification at the time of contact with Europeans ....................................................... 35 (c) Understanding the archaeological record of the Boreal Forest ............................... .48 (d) Can biological evidence resolve the problem of ethnic identity? ............................. 62
Section 3: The images created and used by the Algonquian-speaking peoples ......................... 74 (a) Is establishing the ethnic origins and its objects possible? ........................................ 76
.~. (b) Images as a means of communication .......................................................................... 97 Conclusion: ........................................................................................................................................ 123 Chapter III: A Review of the Literature concerning the Study of Rock Images Sites in the Lake of the Woods in the Canadian Shie1d prior to 2001. ................................... '" ..... 127 Section 1: The earliest investigation of these images .................................................................. 127 Section 2: More recent literature ................................................................................................... 128 Section 3: Techniques used to record the pictographs ............................................................... 136
(a) Subdividing the sites into sections ............................................................................... 136 (b).Terminology ................................................................................................................... 161 Cc) The development of the techniques used ta record the images ............................ 163
Section 4: Technical Analysis .......................................................................................................... 168 (a) Technical issues involved in recording and interpreting rock image sites ............. 168 (b) Conservation and Analytical Techniques ................................................................... 167
Section 5: Culture-Historical Approach .............................................. ......................................... 173 Section 6: Contextual Approach .................................................................................................... 191 Section 7: Literature concerned with meaning ............................................................................. 194
Conclusion ......................................................................................................................................... 209 Chapter IV - My Work on the Rock Image Sites ................................................................. 213 Section 1: Numbers of images ...................................................................................................... 217 Section 2: Defining sites .................................................................................................................. 218 Section 3: What methods were used? ............................................................................................ 218
(a) Examination of maps from different periods ............................................................ 218 (b) Methods for coilecting data ......................................................................................... 223 (c) Image analysis of potential sites in the field ................................ ; .............................. 232
Section 4: The method used for describing shapes and how does one consider time: .......... 235 (a) Exfoliation ...................................................................................................................... 235 (b) White mineraI deposits ................................................................................................. 236 (c) Lichen and rock tripe growth and encroachment .................................................... 237
Section 5: Assessing age .................................................................................................................. 238 Section 6: Description of a shape .................................................................................................. 239 Section 7: The tools used to aid the analysis of the data ........................................................... .242
Ca) IZLEIO lAS ................................................................................................................... 242 (b) Image analysis using Adobe Photoshop 6.0 and VIPS ............................................ 244 Conclusion ............................................................................................................................ 245
Section 8: Vocabulary used to de scribe each type of shape found ........................................... 245 Section 9: Possible relationships between rock image sites and other archaeological sites .. 252 Conclusion: ........................................................................................................................................ 253 Chapter V: Culture-History Approach .................................................................................... 255 Section 1: Blurring/overlay discussion and clarity of images ................................................... .254 Section 2: Distribution of images ................................................................................................... 267 Section 3: Analysis of the rock image style in relation to Northern Ontario and to adjacent
areas ............... : ...................................................................................................................... 268 Section 4: The techniques used to make paintings ..................................................................... .277 Section 5: Discussion of the offerings left at sites ....................................................................... 279 Section 6: Discussion of dates ........................................................................................................ 297 Conclusion ........................................................................................................................................ 387 Chapter VI - Contextual Approach .......................................................................................... 391 Section 1: ............................................................................................................................................ 391
Caves: ............................................................................................................................. 412 Summary for Section 1: ...................................................................................................... 414
Section 2: ........................................................................................................................................... 415 Summary for Section 2: ...................................................................................................... 417
Section 3: ........................................................................................................................................... 417 Summary for Section 3 ....................................................................................................... 435
Conclusion ......................................................................................................................................... 439 Chapter VII - Application of Approaches Concerned with Meaning ........................... .441 Section 1: The two useful approaches concerned with meaning ............................................. .441
Section 2: An exarnination of four birch bark scroils ................................................................ .452
Pageiii
(a) How were the seroIls examined? ............................................................................... .455 (b) Culture-historical Approaeh: ....................................................................................... 456
Comparison of the images found on bireh bark seroIls and on the roek image sites ................................................................................................................................. 458
(e) Contextual Approaeh: .................................................................................................. 460 The eontext of the information on the seroIls: ..................................................... .460
Coneluding Remarks: ........................................................................................................ 472 Seetion 4: The eontinuing importance of the images and the sites .......................................... .4 72 Conclusion: ........................................................................................................................................ 473 Chapter VIII: There is no Holy Grail: An Obsession with Meaning . .......................... ..477 Introduction: ..................................................................................................................................... 477
What has been going on here? ................................................................................................ 481 Seetion 1: Four Goals ...................................................................................................................... 484 Seetion 2: The theoretieal and methodologieal implieations of this thesis: ............................ .497 Seetion 3: Further researeh - What should be done next and whieh way should the field
go? ............................................................................................................................................... 500 Appendices ...................................................................................................................................... 502 Appendix 1: Summary of the information available on the pietograph sites prior to 2001 .. 502 Appendix 2: What some of the Tools of Adobe PhotoShop 6.0 ean reveal. ......................... .525 Appendix 3: A Blank Copy of Form A and Form B for Sites on Cliffs and Caves ............... 531 Appendix 4: The Mod file used by KLEIO lAS ......................................................................... 534 Appendix 5: Uehen, Roek Tripe, Exfoliation, and White Mineral Deposits in eonjunetion
with Site Orientation .......................................................................................................... 542 Appendix 6: The Different Types of the Shape Loosely Categorised as a 'Creature' ............ 550 Appendix 7: Examples of the Different Types of Shapes in Each Group .............................. 557 Appendix 8: The Geology, Geomorphology, and Type of Land Mass of Eaeh Site ............. 565 Appendix 9: Possible Correlations between the Type of Track, Reserves, Offerings in 2001
and before, and Type of Site ................................................................................... 570 Appendix 10: Indian Reserves, Offerings in 2001 and before, Type of Site, Type of Land
Mass, and Orientation of Sites ................................................................................ 577 Appendix 11: Data regarding the Type of Aecess to the Pictograph Site, Vegetation at base
of site, water levels, and the number of eliffs used .............................................. 582 Appendix 12: Do ail the sites have the necessary physieal features for pietographs? ............ 585
Pageiv
Appendix 13: Sites grouped according to the number of physical attributes present ............ 588 Appendix 14: Sites grouped according to number of physical attributes present plus offerings .
..................................................................................................................................... 589 Appendix 15: Sites grouped according to number of physical attributes present and with
rocks at base of cliff face ......................................................................................... 590 Appendix 16: The "presence and absence" tables and the dendrograms from the Pearson
correlation and the Euclidean distance method ta consider the relationship· between the physical attribûtes of the sites, the different sites and the images, and the relationship between the sites and sorne of the largest shapè types .. .591
Appendix 17: The "presence and absence" tables for the images of the birch bark scrolls .. 635 Cited Bibliographic References ...................................................................................................... 637 Ethics Certificate .............................................................................................................................. 681
(Ail the figures in this the sis including the appendices are available for viewing on the CD in the envelope on the inside of the back cover.)
Page v
Abstract
Most researchers who study rock image sites tend to be interested in the meaning of
images, even though they could obtain more empirical information about these images and
their physical location. Furthermore, very little of the work done in the past on rock image
sites has been systematic. In this thesis I address the dearth of detailed information on the
images and their context. This thesis presents a thorough examination of the images of the
twenty-seven pictograph sites in the Lake of the Woods, in the Canadian Shield. These
pictograph sites were selected because they exhibit traits evident in rock image studies in
other parts of the world.
This study is based on data col1ected during three months of fieldwork conducted in
2001. Images were found on cliff faces and inside caves. New images and new sites were
found and identified.
Here, as elsewhere, the choice of theoretical approach influences the fieldwork,
analysis, and search for meaning. Each prescribes the types of questions asked and
determines the levels of understanding obtained about whichever form of archaeological
evidence is being considered. The different but complementary theoretical approaches
should be employed in a definite order. The same data must be examined in sequential order
using these different approaches to increase the potential quantity and quality of information
gained. Archaeologists should use the foilowing sequence of approaches: culture-historical,
contextual, foilowed by either the homological, or analogie al approaches, or a combination
of the latter two.
Classifying and describing any image is very difficult, since the level of description
given to an image affects the way in which it can be analysed, and heavily influences the
possible outcome of any discussion of perceived meaning. A rigorous examination of the
images of these sites was conducted to (a) identify the possible vocabulary of images, (b)
de termine whether combinatory rules exist, (c) reconstitute the life history of each site, and
(d) as certain whether the images can be related to other indigenous images to determine if
this can provide information about the meaning(s) of the rock images. In assessing the
meaning of the rock images, the images of a few birch bark scroils were considered, since it
was posited that a detailed investigation of the scroils, the ethnographic record, and their
pictographs might provide some answers regarding the meanings of the images found on the
rock faces.
Page vi
Résumé
La plus grande partie des chercheurs qui étudient les sites rupestres tendent à
s'interesser à la signification propre des images bien qu'ils puissent en tirer plus
d'informations empiriques sur les images et leurs cadres physiques. En outre, très peu de ces
études ont été systématiques. Cette thèse se met à considerer le grand manque de détails sur
les images dans leur contexte par le moyen d'un examen détaillé, rigoureux et objectif de
vingt-huit sites rupestres dans la région du Lac des Bois sur le Bouclier canadien. Ces sites
ont été choisis.car ils démontrent des traits communs qui se trouvent parmi bien d'autres
sites rupestres à travers le monde. Les données ont été recueillies au cours de trois mois de
travaux sur le terrain entrepris en 2001 et comprennent de nouveaux sites, falaises et abris, et
de nouvelles images identifiés pour la première fois pendant cette période.
Ici, de même qu'ailleur, la méthodologie théorétique choisie pour l'étude détermine la
direction des travaux sur le terrain, leur analyse et les questions qui mènent à la signification
des résultats. Chaque méthodologie ordonne les questions posées et le niveau de
compréhension des réponses selon la catégorie de l'evidence examinée. Des approches
différentes mais complémentaires devraient être employées dans une succession bien
définie.en examinant les mêmes données à fin d'assurer et augmenter la quantité et la qualité
des résultats. Les archéologues devraient déployer ces approches dans l'ordre suivant: sur le
plan culturel-historique; dans le contexte des données; puis par des enquêtes homologiques
ou analogiques, seule ou en combinaison.
La classification et description d'une image est difficile car le niveau de description
influe sur la façon de l'analyser de même que sur l'issue de sa signification. Un examen
détaillé et minutieux des images a été entrepris dans le but de: a) identifier un vocabulaire
possible des images; b) determiner s'il existe des règles combinatoires; c) reconstituer
l'histoire de chaque site; et d) apprendre si ces images pourraient avoir un rapport à d'autres
images hors des sites rupestres, ce qui pourraient aider à éclairer la signification des images
rupestres. Puisqu'on a proposé qu'une enquête détaillée sur les images posées sur écorce de
bouleau pourrait peut-être mener à des réponses aux questions sur la signification de l'art
rupestre, quelques rouleaux d'écorce ont été examinés.
Page vii
. r--'-.
Acknowledgements
My PhD committee is headed by Professor Bruce Trigger and consists of Professors
Toby Morantz, Michael Bisson, and Andre Costopolous from the Department of
Anthropology, McGill and Professor José Igartua from Département d'histoire, Université
du Québec à Montréal. l must thank Bruce Trigger, my supervisor, for the numerous
opportunities to play with ideas and explore their possibilities especially when he may have
thought that they were not sensible but he let me discover for myself that they were really
not very useful. l must thank him for his guidance, his patience, and for always being there
whenever l needed his help and advice and for giving me the opportunity to think carefully,
clearly, critically, and to evaluate everything with an open mind and the widest possible
perspective. l have tried at aIl times throughout this degree to follow his advice regarding
data to make sure that l squeezed the orange until nothing more could be extracted.
Imust thank Bruce Trigger, Toby Morantz, and José Igartua for aIl their thoughtful
editing and writing advice. l would like to thank José Igartua for his advice and assistance
with regard to the computing tools utilised and for his advice when writing about something
technical in computing so that l used as few technical words as possible but still got the ideas
and concepts across to the reader. l must thank Andre Costopolous for his advice and
technical assistance and Michael Bisson for his thoughtful comments. l must thank Toby
Morantz for her support and help during the final year of this PhD.
l must thank Professors Bruce Trigger and Toby Morantz for financial support
whilst writing up my thesis. l am immensely grateful for their help .
l would like to thank various members of the Anthropology department at McGill
with whom l took courses including Professors Colin Scott, Toby Morantz and Jerome
Rousseau for the opportunity to explore and obtain a solid grounding in the field of
anthropology as it is perceived in North America
l must thank Dr. Scott Hamilton (Department. of Anthropology, Lakehead
University, Ontario) for his advice on various drafts of the conclusion of my thesis.
l must thank Robert von Bitter (Ontario Ministry of Culture and Communication,
Toronto, Ontario); Dr. Stewart Baldwin, Dr. Scott Hamilton, and Dr. Joe Stewart (professor
Emeritus) of the Department of Anthropology, Lakehead University, Ontario), C. S. Reid
(formally Regional Archaeologist, Ontario Ministry of Culture and Communications, Kenora,
Ontario), Dr. Jean-Luc Pilon, Curator of Ontario Archaeology, Canadian Museum of
Page viii
I~
.,.--.. ..
Civilization, Hull, Québec), Dr. J. Peter Denny (professor Emeritus, Department of
Psychology, The University of Western Ontario), Dr. Richard Preston (professor Emeritus,
Dept. of Anthropology, McMaster University, Ontario), Dr. Rhonda Telford (Historical
Research and Consulting, Hamilton, Ontario), W. A. Ross (formally Regional Archaeologist,
Ontario Ministry of Culture and Communications and currently Ross Archaeological
Research Associates, Thunder Bay, Ontario), Dr. Martin Magne (Manager, Cultural Resource
Services, Western and Northern Service Centre, Parks Canada), Dr. Cath Oberholtzer
(Conjunct Professor, Department of Anthropology, Trent University, Ontario), Dena
Doroszenko (Ontario Heritage Trust, Toronto, Ontario), Dr. 1. N. M. Wainwright
(Analytical Research Laboratory, Canadian Conservation lnstitute, Ottawa, Ontario),
Margaret McMahon (Water Statistics Analyst, Water Resources Division, Ontario Power
Generation), Mike Larson (Minnesota Department of Natural Resources, Minnesota), Rick
Cousins (Senior Water Resources Engineer, Lake of the Woods Secretariat, Lake of the
Woods Water Control Board, Canada), Dr. Daniel Arsenault (Départment d'histoire de l'art,
Université du Québec à Montréal, Québec), Dr. Beth A. S. Shook (Department of
Anthropology, CSU Chico, California, USA), Dr. Tim Jones (Executive Director,
Saskatchewan Archaeological Society, Saskatoon, Saskatchewan), Dr. lan A. Cowx
(International Fisheries lnstitute, University of Hull, UK), Dr. Charles A. Bishop (Research
Professor of Anthropology, Union College, Schenectady, New York, USA), Dr. John Long,
(Associate Professor, Faculty of Education, Nipissing University, Ontario), Dr. Jane Down
(Senior Conservation Scientist, Conservation Processes and Materials Research, Canadian
Conservation Institute, Ottawa, Ontario), Nancy Binnie (Conservation Scientist, Canadian
Conservation Institute, Ottawa, Ontario), Mark J. Dudzik (State Archaeologist, Fort Snelling
History Center, St. Paul, Minnesota, USA), Marie-Louise Perron (Library and Archives
Canada, Canadian Genealogy Centre, Library and Archives Canada, Ottawa, Ontario),
Jonathan Moore (Underwater Archaeology Service, Parks Canada Agency, Ottawa, Ontario),
Kari Layman (Hydraulics, USACE - St. Paul District, Minnesota, USA), Dr. Lisa Dillon
(Départment de demongraphiè, Université de Montréal, Québec), Dr. Sara Louri, and Hayley
Fitzsimmons for giving me references, information, archaeological reports, data of different
types, slides, files, and opportunities to discuss ideas.
l must also thank those archaeologists who had worked on sites in the Lake of the
Woods prior to myself and who kindly provided the references of their work: C. S. Reid
Page ix
I~
(formally Regional Archaeologist, Ontario Ministry of Culture and Communications, Kenora,
Ontario), Dr. Brian L. Molyneaux (Department of Anthropology, University of South
Dakota, Vermillion, South Dakota, USA), Dr. Jack Steinbring (Department of Anthropology,
Ripon College, Ripon, Wisconsin, USA), and William A. Fox (Field Unit Superintendent,
Western Arctic District, Parks Canada).
l must thank my field assistant Melinda Bell and my volunteer Elizabeth Campbell
for their help and assistance in the field. l must also thank W. A. Ross (Regional
Archaeologist, Thunder Bay, Ontario) for technical help and advice while in the field. l must
also thank David Fraser and the diving team from Sunset Diving, Kenora, Ontario for their
assistance while surveying and recording the pictograph sites. 1 must thank the Kast family
and everyone at Tomahawk Lodge and Marina, Wendal Dafcik and his guides of Crow Rock
Lodge, the owners and the guides of Monument Bay Lodge, Paul and Beverly Eliuk, Bill and
Billie-Jean Murphy from Sioux Narrows provided advice and expertise which made the
practical aspects of conducting field work possible.
l must also thank Professor Wendy Hall, Head of School, Professor Paul Lewis, and
Dr Kirk Martinez (Intelligence, Agents, Multimedia Group, School of Electronics and
Computing Science, University of Southampton, UK) for their help, guidance, and advice
when 1 used VIPS and for the experimental work 1 did in their laboratory. l must thank Dr
Mike Westmacott, Dr M. Faizal A. Fauzi, Dr David P Dupplaw, and Dr. Fazly Salleh Abas
who were doing research for their PhDs while 1 was there as well as Dr. David Millard and
Dr. Christopher Bailey for their technical assistance and for their questions because 1 realised
that the manner in which images are described fundamentally and directly affects its
subsequent interpretation.
1 must thank Dr. Andrew Sawyer (Hon. Research Fellow, Department of Museum
Studies, University of Leicester & mwr ltd) for teaching me how to use KLEIO IAS. l must
thank both Professor Manfred Thaller (professor in Computer Science for the Humanities,
Universitat zu KaIn, Germany) and Dr. Gerhardt Jaritz (Department of Medieval Studies,
Central European University, Central European University, Prague, Hungary) for help and
advice regarding the technical and theoretical aspects of KLEIO and KLEIO lAS.
l must thank Dr. Mima Kapches for permission to consult the Selwyn Dewdney
Papers in the New \Vorld Department Collection, in the Royal Ontario Museum, Toronto,
Ontario and Lori Nelson, the Director of the Lake of the Woods Museum in Kenora,
Page x
Ontario for letting me examine the papers of several early settlers in the Lake of the Woods
and using the photograph of the Massacre Scroll.
Numerous people gave me advice on how to track down a photograph of a birch
bark scroll and their combined efforts made the chapter on birch bark scrolls possible. l
must thank Moira McCaffrey (Director, Research and Exhibitions, McCord Museum of
Canadian History, Montréal, Québec); Dr. Jban Vastokas (professor Emerita, Department of
Anthropology, Trent University, Ontario), Dr. Jennifer S.H. Brown, (Director of The Centre
for Rupert's Land Studies/Department of History, The University of Winnipeg, Manitoba),
Dr. George Fulford (Department of Anthropology, The University of Winnipeg, Manitoba),
Dr. Sally Cole (Department of Anthropology and Sociology, Concordia University, Montréal,
Québec), Dr. J. c. H. King (Keeper, Department of Africa, Oceania and the Americas,
British Museum, London, England), Dr. Teresa Schenk (Center for the Study of Upper
Midwestern Cultures, University of Wisconsin, Madison, Wisconsin, USA), Lori Nelson
(Director, Lake of the Woods Museum, Kenora, Ontario), and finally Dr Heidi Bohdaker
(SSHRC Postdoctoral Fellow, Institute for Comparative Studies in Literature, Art and
Culture, Carleton University, Ontario).
l must thank several friends and colleagues who took cime from their own work to
read, comment, and discuss various versions of my thesis: Harry Lerner, Dr. Csilla Dallos
(Department of Anthropology, St. Thomas University, New Brunswick), Apryl
Wassaykeesic, Cecil Chabot, Loretta Flegel, Sheila Oakley, Yasir Khan, Dr. André LeBlanc
(Visiting Assistant Professor, History of Science and Technology Programme, University of
King's College, Nova Scotia) and Dr. Carolyn Podruchny (Department of History, York
University, Ontario). l must thank Elizabeth Campbell for editing this thesis, Francis Carson
who translated my abstract into French, and Neha Gupta and Rebekah Jobling who helped
convert sorne electronic fIles from one format to another.
l must thank the interlibrary loans department and the Blackadder-Lauterman,
Redpath, McClennan, Schulich libraries and particularly Rosa Orlandini and Joanna Hobbins
from the Walter Hitschfeld Geographie Information Centre for their assistance. l am grateful
for the continual assistance and guidance of Rose Marie Stano, Cynthia Romanyk, and Diane
Mann from the Department of Anthropology at McGill.
l must thank several people for technical support and advice regarding any computer
hardware problems that l had including Sean Colson (pC Consultants, Ryde, Isle of Wight,
Page xi
UK), University Bytes (Montréal, Quebec), Computech (Kenora, Ontario), Ian Shaw
(Managing Director of mwr InfoSecurity, Basingstoke, Hampshire, UK) and Joanna
Hobbins.
l must thank the following organizations for providing funding for m)' research: Max
Bell Fellowship for Canadian and Northern Studies, Sigma Xi Grant-in-Aid, Sigma Xi,
MacKenzie Ward Trust, Sioux Narrows, Ontario, Ontario Rock Art Conservation
Association, Canadian Historical Association, two Alma Mater Student Travel Grants, as weIl
as in-kind funding from the Ontario Ministry of Culture, Tourism and Recreation.
l must thank those friends who have not been involved in any aspect of my research
for having provided moral support: Alice Anderson, Denise Arthur, Andrea Cardarello, Sara
and Jim Busby, Nancy Chapman, Dr. Cristina Redko Meir, Dr. Caroline Gooch, Dr.
Judy Petch, Masami Waki, Rieko Sasa, Teresa deI Blanco, Nuzhat Dunn, Debra Irons, Sarah
Phillips, Lucy Braga, Rita Weidendorfer, Jennifer Famery, Heather Podruchny, Simone
Abuhab, Belinda Beaton, Alessandra Crippa, Eleanor Roff, and Dan Friesting. l come from a
large family and all of them have been incredibly supportive while undertaking this degree
but l must specifically thank a few of them: Professor Richard and Betty Preston, Finbar and
Susie Coison, my godparents Professor Henry and Mercedes Ettinghausen, my brother-in
law Ian Shaw, my sister Beth Coison, and finally my parents Dr. R. Frank and Jean Coison
l would like to dedicate this the sis to two people: my paternal grandmother Kathleen
Cotter Colson and my maternaI grandfather MacKenzie Belfield Ward. My grandmother
taught me that it was important to stand up for the freedom of speech and thought no
matter how hard it rnight be to do this. My grandfather, MacKenzie Ward had a reputation
of being both strong rninded and highly opinionated. Both of us disagreed with each other
numerous times about many things but ultimately we both knew that it was better to agree to
disagree because our disagreements stemmed from each of us having different perspectives
and attitudes about the world and the way it functioned. He repeatedly told me that it was
better to stand up for what l believed, if l believed my perspective was weIl grounded and
researched, regardless of how hard it might be at the time than to buckle and back down
because of someone else's opinion.
Page xii
r--' ..
N ote on Spelling
The author of this thesis has used standard British English grammar, punctuation,
and spelling from the O:x.ford English Dictionary throughout her research and her thesis since
she started at McGill in September 1998.
Page xiü
Note on KLEIO lAS
The combination of the two letters 'xq' plus a number appear throughout several of the chapters in this thesis. This is not a special code. The letters 'xq' indicate that the result was provided as a consequence of a query numbered 'xq and a number'. It is a convention of KLEIO to use these letters and numbers as references to a particular database query.
Page xiv
List of Figures
Figure 1.1: Physical Location of the Lake of the Woods ............................................................... 2 Figure 1.2: Examples of shapes loosely categorised as 'crescents' and the sites that they come
from in the Lake of the Woods .................................................................................... 13-15 Figure 2.1: Cliffs in the southern end of Whitefish Bay with clearly defined high water levels ............................................................................................................................................................... 24 Figure 3.1: Map of the sites that Lawson recorded in 1885 ....................................................... 127 Figure 3.2: Lawson's published drawing of the images at DiKo-2, the petroglyph site. In:
Lawson, A. C 1885. Ancient Rock Inscriptions on the Lake of the Woods. Amm'can NaturalistXIX (7): 654-657.Pp.: 656 ................................................................................ 128
Figure 3.3: Lawson's published drawing of the images at DiKp-l, the pictograph site from 1885. Lawson, A. C 1885. Ancient Rock Inscriptions on the Lake of the Woods. American Naturalist XIX (7): 654-657. Pp: 657 ............................................................... 128
Figure 3.4: Annotated photograph of DiKp-l taken in June 2001 ........................................... 129 Figure 3.5: The four sites that Cameron recorded in 1978 ........................................................ 131 Figure 3.6: Fox's photograph of DhKm-l in 1974. Courtesy of Ontario Ministry of Culture
and Communications .......................................................................................................... 133 Figure 3.7: Dewdney's 15 July 1960 field record of Site # 92A, now called DhKm-l and
Figure 3.12: The Close Proximity of DhKm-4 to DhKm-l ...................................................... 140 Figure 3.13: Annotated photograph of DhKm-l in 2001 by Cols on according to Dewdney's
divisions ................................................................................................................................ 141 Figure 3.14: Annotated photograph by toIson of DhKm-1 taken in 2001 according to
Ontario Museum, Toronto Canada .................................................................................. 142 Figure 3.16: Part 1 of Dewdney's field drawing of Site # 91A and B, undertaken on 15 July
Royal Ontario Museum, Toronto Canada ...................................................................... 146 Figure 3.22: Photograph of DgKl-l in July 2001 with Pastershank's 1989 subdivisions ...... 146
Page xv
Figure 3.23: DgKl-2 in 1989 bl' Pastershank. Courtes)' of Ontario Ministry of Culture and Communications ................................................................................................................. 147
Figure 3.24: Dewdney's published images from this site. In: Dewdney S. and K. Kidd. 1967. Indian Paintings if the Great Lakes. 2nd Edition Toronto: University of Toronto Press. p.l10 ..................................................................................................................................... 148
Figure 3.25: Three of the new images at DgKl-2 in July 2001 .................................................. 148 Figure 3.26: The new image at DgKl-2 not painted upon the older ochre images in July 2001
............................................................................................................................................... 149 Figure 3.27: Annotation by Coison in 2004 of DgKl-2 taken by Pastershank in 1989 ......... 150 Figure 3.28: Field recording of detail of Face l, DgKI-2, Pastershank in 1989. In:
Pastershank, G. 1989. An Archaeological Sun1ry if Sabaskong B'!)', Lake if the Woods. Ontario Conservation Archaeology Report #13, North-western Region. Kenora: Ontario Ministry of Culture and Communications. p. 56 ............................................ 150
Figure 3.29: Photograph of Face 1 by Pastershank in 1989. Courtesy of Ontario Ministry of Culture and Communications ............ ; ........................................................................... 151
Figure 3.30: Annotation by Coison in 2004 of photograph, of Face II by Pastershank in 1989. Courtesy of Ontario 1linistry of Culture and Communications .............................. 152
Figure 3.31: Field drawing with detail of Morph 4 at DgKl-2. In: Pastershank, G. 1989. An Archaeological S urvry if S abaskong B'!)', Lake if the Woods. Ontario Conservation Archaeology Report #13, North-western Region. Kenora: Ontario Ministry of Culture and Communications. p. 58 ................................................................................ 152
Figure 3.32: Field recording of detail of Face III. In: Pastershank, G. 1989. An Archaeological Survry if Sabaskong B'!)'J Lake if the Woods. Ontario Conservation Archaeology Report #13, North-western Region. Kenora: Ontario Ministry of Culture and Communications. p. 60 ...................................................................................................... 153
Figure 3.33: Field recording of Face IV of DgKl-2. In: Pastershank, G. 1989. An Archaeological Survry if Sabaskong B'!)'J Lake if the Woods. Ontario Conservation Archaeology Report #13, North-western Region. Kenora: Ontario Ministry of Culture and Communications. p. 61 ................................................................................ 154
Figure 3.34: DgKI-2, Detail of Face IV by Pastershank in 1989. Courtesy of Ontario Ministry of Culture and Communications ...................................................................................... 155
Figure 3.35: Pastershank's Field recordings of Faces V, VI and VII. In: Pastershank, G. 1989. An Archaeological Survry if Sabaskong B'!)'J Lake if the Woods. Ontario Conservation Archaeology Report #13, North-western Region. Kenora: Ontario Ministry of Culture and Communications. p. 62 ................................................................................ 155
Figure 3.36: U sing Pastershank's (1989) subdivisions of rock surface to divide the paintings in 2001 into Faces V, VI and VIl ..................................................................................... 156
Figure 3.37: Face l of DgKl-17 by Pastershank in 1989. Courtesy of Ontario Ministry of Culture and Communications ........................................................................................... 158
Figure 3.38: DgKl-17 _ Pastershank's FaceII and II in 1989. In: Pastershank, G. 1989. An Archaeological Survry of Sabaskong B'!)'J Lake of the Woods. Ontario Conservation Archaeology Report #13, North-western Region. Kenora: Ontario Ministry of Culture and Communications. p. 66 ................................................................................ 159
Figure 3.39: Pastershank divided this Face into NO parts called Panel i and ü in 1989. In: Pastershank, G. 1989. An Archaeological Survry if Sabaskong B'!)'J Lake if the Woods. Ontario Conservation Archaeology Report #13, North-western Region. Kenora: Ontario Ministry of Culture and Communications. p.68 .............................................. 159
Figure 3.40: DhKm-5 in June 2001 ............................................................................................... 160
Royal Ontario Museum, Toronto Canada ...................................................................... 164 Figure 3.43: The plastic being taped to the surface of the cliff face and the mineraI
precipitation and patination being recorded by a black pen by Rusak in 1991. Courtesy of Ontario Ministry of Culture and Communications .................................. 167
sites. In: Dewdney S. and K. Kidd. 1962. lndian Paintings of the Great Lakes. Toronto: University of Toronto Press, p. 18 ................................................................................... 174
Figure 3.46: The Massacre Scroll photographed in 1993 by J. Rusak. Collections of the Lake of the Woods Museum, Kenora, Ontario ....................................................... : .............. .190
Figure 3.47: Part of DhKn-l beside Obabikon Channel ........................................................... l92 Figure 3.48: The serpents and the turtles described by Dewdney in 1960 .............................. 199 Figure 4.1: Examples of cliff faces examined for pictographs during 2001 ............................ 220 Figure 4.2: DhKm-18 (an "oId" site) and DhKm-20 Ca "new" site) ....................................... .221 Figure 4.3: A map of where the diving took place ...................................................................... 222 Figure 4.4: Overview of DgKI-2 taken in 2001 ........................................................................... 224 Figure 4.5: Physical location of DgKI-2 in Sabaskong Bay, Lake of the Woods. C4nadian
Hydrographic Services. "Sabaskong Bay, Lake of the Woods, Southeastern Channel" [hydrographic chart]. 1 :40,000. Sheet 6214. Ottawa: Department of Oceans and Fisheries, 1989 .................................................................................................................... 225
Figure 4.6: Field drawing of DgKI-2 by M. Bell in 2001 ............................................................ 229 Figure 4.7: Red/orange lichen on the western end of Crowrock Island ................................. .233 Figure 4.8: Part of DjKn-l used as the control file ..................................................................... 234 Figure 4.9: DiKp-l in June 2001 annotated by Colson in 2002 to indicate locations of previously unknown images ......................... ; .............................................. 241 Figure 4.10: Map indicating location of other known archaeologïcal sites to pictograph sites ............................................................................................................................................................. 253 Figure 5.1: DkIZn-6 in 2001 ............................................................................................................ 256 Figure 5.2: The different shapes of 'creatures' at DhKn-l in 2001 .......................................... 257 Figure 5.3: Overview of DhKm-4 in Whitefish Bay in 2001 ..................................................... 257 Figure 5.4: DhKm-l has images affected by exfoliation, white mineraI deposits and rock tripe
accumulation of white mineraI deposits at DhKm-S is not uniform .......................... 264 Figure 5.12: Screenshot showing what occurs to the same im4ges using different tools when 1
mathematically tried to rem ove the white pixels deemed 'noise" ................................ 266 Figure 5.13: A map of the Lake of the Woods area showing the numbers of images at each
site ......................................................................................................................................... 267 Figure 5.14: D1Jn-l from Harmon Lake Courtesy of Ontario Ministry of Culture and
Figure 5.15: DjKp-l0 on Moberly Lake. Courtesy of Ontario J\1inistry of Culture and Communications ................................................................................................................. 270
Figure 5.16: DjJp-l on Moberly Lake. Courtesy of Ontario J\1inistry of Culture and Communications ................................................................................................................. 271
Figure 5.17: Close up of a rectangle with the circle and the diagonallines fromDhKm-3 in 2001 ....................................................................................................................................... 276
Figure 5.18: The rectangle with the stick figure at DhKo-l in 2001 ........................................ 277 Figure 5.19: Sites where offerings were found in 2001 .............................................................. .281 Figure 5.20: Sites where tobacco was found as an offering in 2001 ......................................... 282 Figure 5.21: Sites where money was found as an offering in 2001 ........................................... 283 Figure 5.22: Offerings discovered by Fox in 1974 at DhKm-1. Courtesy of Ontario J\1inistry
of Culture and Communications ...................................................................................... 284 Figure 5.23: Sites where clothing was found as offerings in 2001 ............................................ 285 Figure 5.24: Sites where liquor botdes were found as offerings in 2001 ................................. 286 Figure 5.25: Offerings that Dewdney found at DjKq-l0 called Site # 95, Picture Rock Point
Canada .................................................................................................................................. 300 Figure 5.48: Face 1 of DgKl-1 and Face 2 of DgKl-2 taken by Pastershank in 1989. Courtesy
of Ontario J\1inistry of Culture and Communications .................................................. 301 Figure 5.49: Face 3 of DgKl-1 by Pastershank in 1989. Courtesy of Ontario J\1inistry of
Culture and Communications ........................................................................................... 301 Figure 5.50: Photograph of DgKl-2 in 2001 ................................................................... : ............ 302
Figure 5.53: Pastershank's Face 1 by Pastershank in 1989. Courtesy of Ontario Ministry of Culture and Communications .................................... : ...................................................... 304
Figure 5.54: Faces II, III and IV by Pastershank in 1989. Courtesy of Ontario Ministry of Culture and Communications ........................................................................................... 305
Figure 5.55: Setting from distance of Site # 28, Sioux Narrows by Dewdney in 1953. Courtesy of Ontario Ministry of Culture and Communications .................................. 306
Figure 5.56: Face 2 and 3 of Site # 28 by Dewdney in 1953. Courtesy of Ontario Ministry of Culture and Communications ........................................................................................... 307
Canada .................................................................................................................................. 309 Figure 5.60: Site # 106, Sabaskong-Obabikon Channel, Watercolour reproduction of Face II
Ontario Museum, Toronto Canada .................................................................................. 312 Figure 5.65: DhKm-1 taken in September 1974 by W. Fox. Courtesy of Ontario Ministry of
Canada .................................................................................................................................. 319 Figure 5.71: Section of DhKo-1 taken by Molyneaux. In: Molyneaux, B. L. 1987. The Lake of
Toronto Canada .................................................................................................................. 322 Figure 5.76: DiKm-4 by Moll'neaux. In: Molyneaux, B. L. 1980a. Landscape Images Rotunda
13 (3): 10 .................................................................. ; ........................................................... 323 Figure 5.77: DiKm-4 bl' Pelshea in 1978. Courtesy of Ontario Ministry of Culture and
Communications ................................................................................................................. 323 Figure 5.78: DiKm-4 taken in 1991 ............................................................................................... 324
Page xix
Figure 5.79: DiKm-4 taken in 2001 ............................................................................................... 324 Figure 5.80: Dewdney's field drawing of Site # 69 called Picture Rock Island undertaken on
Royal Ontario Museum, Toronto Canada ...................................................................... 327 Figure 5.84: Black and White photograph of DhKm-3 taken by Molyneaux. In: Molyneaux,
B. L. 1980a. Landscape Images Rotunda 13 (3): 7 ........................................................... 327 Figure 5.85: The images at DhKm-3 taken by Pelshea in 1978. Courtesy of Ontario Ministry
Canada .................................................................................................................................. 337 Figure 5.97: Dewdney's field drawing of Site # 70 in Sunset Channel undertaken on August
Museum, Toronto Canada ................................................................................................. 345 Figure 5.107: DgKl-19 by Pastershank taken in 1989. Courtesy of Ontario Ministry of
Canada .................................................................................................................................. 352 Figure 5.117: Bird identified by Dewdney and Lambert beside the two vertical stick identified
in 2001 .................................................................................................................................. 353 Figure 5.118: Smear at DhKm-18 in 2001 ........................ ; .......................................................... 353
Images Rotunda 13 (3): 9 ..................................................................................................... 360 Figure 5.128: Pelshea's photograph of the le ft hand side of DjI<n-l taken in 1978. Courtesy
of Ontario Ministry of Culture and Communications .................................................. 361 Figure 5.129: Pelshea's photograph of the right hand side of DjI<n-l taken in 1978. Courtesy
Canada .................................................................................................................................. 366 Figure 5.136: DiKn-l in J une 2001 ................................................................................................ 366 Figure 5.137: Blob beside exfoliated area at DiI<n-l in 2001 .................................................... 367 Figure 5.138: Dewdney's "Face lA fig a + b" at DiI<n-l in 2001.. .......................................... 367 Figure 5.139: Red images are visible although partially hidden by the lichen and rock tripe in
the upper half of DiKn-l in 2001 .................................................................................... 368 Figure 5.140: Shapes above the main collection of shapes at DiI<n-l ..................................... 368 Figure 5.141: DhKm-20 in June 2001 ........................................................................................... 369 Figure 5.142: DhKm-19 in 2001 .................................................................................................... 370 Figure 5.143: DgKm-2 in July 2001 ............................................................................................... 371
Page xxi
Figure 5.144: DiKm-49 in June 2001 .................................................... : ...................................... ;372 Figure 5.145: DiKin-50 in June 2001 ............................................................................................ 373 Figure 5.146: DhKrn-20 created by R, Bird and signed in 1982 ............................................... 374 Figure 5.147: Two of the modern solid spray painted images at DgKl-2 ................................ 374 Figure 5.148: The third modern image which is solid at DgKl-2 .............................................. 375 Figure 5.149: The modern hollow spray painted image at DgKl-2 ........................................... 375 Figure 5.150: Modern spray paint image at DhKrn-20 ............................................................... 376 Figure 6.1: DhKm-l photographed August 27'h 2003 ................................................................ 398 Figure 6.2: Annotated photograph by Colson of the holes which are called caves at DgKI-2
in the rock above the water level.. ................................................................................... .403 Figure 6.3: Inside the cave at DgKl-2 ........................................................................................... .403 Figure 6.4: Map of the pictograph sites that have deep holes in 2001 .................................... .404 Figure 6.5: Map of pictograph sites with large cracks in 2001 .................................................. .405 Figure 6.6: Map of pictograph sites with crevices in 2001 ........................................................ .406 Figure 6.7: Map of pictograph sites with overhangs in 2001 ................................................... ..407 Figure 6.8: Map of pictograph sites with rock shelves, or benches or ledges in 2001 .......... .409 Figure 6.9: Map of pictograph sites with rocks at their base in 2001 ...................................... .411 Figure 6.10: DiKm- 50 on Horseshoe Island, Whitefish Bay ................................................... .412 Figure 6.11: The black marks on the surface of the rock's surface ........................... ..413 Figure 6.12: Sites with smears ......................................................................................................... 419 Figure 6.13: Sites with blobs ........................................................................................................... 421 Figure 6:14: Sites with unidentifiable shapes .............................................................................. .423 Figure 6.15: Sites with vertical stick figures ................................................................................. .424 Figure 6.16: Sites that the shape of the 'upper torso of a vertical stick figure' ...................... .425 Figure 6.17: Sites that had the shape of hand prints .................................................................. .425 Figure 6.18: Location of sites that had both a left and a right hand print shape ................... .426 Figure 6.19: Sites with shapes of loosely categorised creatures ............................................... ..427 Figure 6.20: An example of a shape called 'one vertical and one horizontal short line that
cross each other at ninety degrees' ................................................................................... 430 Figure 6.21: Sites with one vertical and one horizontal short line that cross each other at
ninety degrees ...................................................................................................................... 431 Figure 6.22: An example of a shape called 'one vertical and one horizontal short line that
cross each other at ninety degrees with short lines at the end of each of the vertical and horizontallines' ........................................................................................................... 431
Figure 6.23: Sites with the shape called one vertical and one horizontal short line cross each other at ninety degree angle with short lines at the end of each of the vertical and horizontallines .................................................................................................................... 432
Figure 6.24: Sites with groups of four diagonallines ................................................................. .433 Figure 6.25: Sites with shapes called 'three lines joined at one end' ........................................ .434 Figure 7.1: The left hand part of scroll number 1. Courtesy of Lake of the Woods Museum,
Kenora, Ontario, Canada ................................................................................................... 452 Figure 7.2: The right hand part of scroll number 1. Courtes)' of Lake of the Woods
Museum, Kenora, Ontario, Canada ................................................................................ .453 Figure 7.3: One of the scrolls that Densmore photographed. In: Densmore, F. 1910.
Chippewa Music I. Washington Government Printing Office. Srnithsonian Institution, Bureau of American Ethnology. Bulletin 45. p. 110 ................................ .453
Page xxii
Figure 7.4: One of the serolls that Densmore photographed. In: Densmore, F. 1910. Chippewa Music I. Washington Govemment Printing Office. Smithsonian Institution, Bureau of Ameriean Ethnology. Bulletin 45. p. 110 ................................ .453
Figure 7.5: The scroll published by Dewdney in 1970. In: Dewdney, S. 1970c. Ecologicai Notes on the Objiway shaman-artist. Artscanada. August: 17-28. P.: 27-28 .............. .454
Figure 7.6: Dewdney's tracing of the scroll in figure 7.5. In: Dewdney, S. 1975b. The Sacred Scroils of the Southem Ojibwqy. Toronto and Buffalo: University of Toronto Press: Figure 146 Song record with herbaI references (ROM -3), on p. 141 ....................... .455
Figure 7.7: The nine differences between the Dewdney's drawing and the original seroll .. .455 Figure 7.8: Scroll number 2 annotated by Colson from left to right in 2005 ......................... .464 Figure 7.9: Mnemonic for the Song of Good Medicine. In: Densmore, F. 1910. Chippewa
Music 1. Washington Government Printing Office. Smithsonian Institution, Bureau of American Ethnology. Bulletin 45. p. 1 00 ....................................................................... .464
Figure 7.10: Mnemonic for the Song of the Crab Medicine Bag. In: Densmore, F. 1910. Chippewa Music 1. Washington Government Printing Office. Smithsoruan Institution, Bureau of American Ethnology. Bulletin 45. p. 102 ..................................................... .464
Figure 7.11: Mnemonie for the Song of the Fire-Charm. In: Densmore, F. 1910. ChippelJJa Music 1. Washington Government Printing Office. Smithsoruan Institution, Bureau of American Ethnology. Bulletin 45. p. 103 ....................................................................... .465
Figure 7.12: Scroll number 3 annotated by Colson in 2005. In: Densmore, F. 1910. Chippewa Music 1. Washington Government Printing Office. Smithsonian Institution, Bureau of Ameriean Ethnology. Bulletin 45. p. 106 ....................................................................... .465·
Figure 7.13: Mnemonic for the Song of the Owl Medicine. In: Densmore, F. 1910. Chippe1va Music 1. Washington Government Printing Office. Smithsoruan Institution, Bureau of American Ethnology. Bulletin 45. p. 1 04 ....................................................................... .465
Figure 7.14: Mnemonic for the Song of Starvation. In: Densmore, F. 1910. Chippewa Music 1. Washington Govemment Printing Office. Smithsoruan Institution, Bureau of American Ethnology. Bulletin 45. p.l 04 ........................................................................ .466
Figure 7.15: Mnemonic for the Song of the Man who Succeeded. In: Densmore, F. 1910. Chippewa Music 1. Washington Govemment Printing Office. Smithsonian Institution, Bureau of American Ethnology. Bulletin 45. p. 98 ....................................................... .466
Figure 7.16: Seroll number 4 annotated by Coison in 2005 ...................................................... .467 Figure 7.17: The shapes at DiKm-4 according to Ann Seymour ............................................. .468 Figure 7.18: The turde, snake, and the first lattice discussed by Seymour at DhKm-3 ........ .469 Figure 7.19: The thunderbird, line, handprint, lattice, and the human figure with hands raised
upwards at DhKm-3 identified by Seymour. ................................................................. .470 Figure 7.20: The annotated photographs of images at DhKn-l .............................................. .471 Figure 7.21: An annotated photograph ofDhKo-l ................................................................... .471
Page xxiii
List of Tables:
Table 1.1: Sequence in which approaches must be utilised ............................................................ 6 Table 6.1: Hierarchical Cluster Analysis - dendrogram using Average Linkage (Between
Groups) using Euclidean distance method to establish whether a relationship exists between the sites and the different features .................................................................. .401
Table 6.2 Hierarchical Cluster Analysis - dendrogram using Average Linkage (Between Groups) using Pearson correlation method to establish whether a relationship exists between the sites and the different features .................................................................. .401
Table 6.3 Hierarchical Cluster Analysis - dendrogram using Average Linkage (Between Groups) using Euclidean distance method to determine the nature of the relationship between smears and sites ................................................................................................. 420
Table 6.4: Hierarchical Cluster Analysis - dendrogram using Average Linkage (Between Groups) using Euclidean distance method to determine the nature of the relationship between smears and sites .................................................................................................. 420
Page xxiv
,~.
Chapter 1 - Introduction
While most people who study rock image sites tend to be interested in the meaning
of images, much more empirical information can be obtained about these images and their
physical location. Very little of the work done in the past on rock image sites has been
systematic. Hence, in this thesis l will address the dearth of detailed information on the
images and their context. l will subsequently suggest a new approach to studying rock images
that is as systematic, rigorous and objective as possible using a group of sites in the Canadian
Shield. The goal of this thesis is to rigorously conduct a detailed and thorough examination
of the images of the pictograph sites in the Lake of the Woods region to (a) identify the
possible vocabulary of images, (b) determine whether combinator)' mIes exist, (c)
reconstitute thelife history of each site, and (d) ascertain whether the images can be related
to other images outside the rock image sites to de termine if this can provide information
about the meaning(s) of the rock images.
Where are these sites?
The rock image sites of the Lake of the Woods region exhibit traits evident in rock
Image studies in other parts of the world. Here, as elsewhere, the choice of theoretical
approach influences the fieldwork, analysis, and search for meaning. This study is based on
data collected during three months of fieldwork conducted on the pictographs in the Lake of
the Woods in 2001. Petroglyph sites were not included in this study because the high water
levels in the field season of 2001 inundated these sites. l did not examine petroforms since
this study only concerns images on rocks and not images created using rocks. Only the
pictograph sites were examined.
l chose the sites of the Lake of the Woods area, in Northwestern Ontario, because
they appear representative of the data resulting from the archaeological surveys conducted
on the rock images of the Canadian Shield. The map below, in figure 1.1, indicates the
physicallocation of the Lake of the Woods with respect to the Great Lakes and Hudson Bay.
Page 1
,-MANtTOBA ~~~4
(
:
/ ......... . .'
.,,,,,CO
lKE OF THE WOODS
.' /
Figure 1.1: Physical Location of the Lake of the Woods
QUE BEC
The majority of the scholars interestèd in rock images of the Canadian Shield (for
example: Dewdney and Kidd 1962, 1967, and 1975; Fox 1974; Lanoue 1989; Lambert 1983,
1985, n.d.; Molyneaux 1975; Rajnovich 1989, 1994; Reid 1977) examined the sites in the Lake
of the Woods from 1885 (Lawson 1885) onwards. The individu al pictograph sites associated
with this lake vary in the number and range of their images. Most of the known pictographs
occur at the base of vertical granitic rock walls, either immediately beside the water or several
metres from the water's surface. None of these sites can be dated using absolute·
chronometrie techniques. l based ml' research on archaeological experience gained from
conducting fieldwork in the Canadian Shield region since 1990.
My views on rock image studies in general
In preparation for my thesis, l carried out a comparative survey of rock image studies
around the world. l examined the effects both of different theoretical approaches, and the
interpretative themes used by each scholar. The term 'rock art' is applied world-wide to
images that are placed on the surfaces of rocks. 1t occurs in many different places and
settings: Australian rock shelters, the surfaces of boulders in the Jordanian desert, vertical
Page 2
r---..
rock faces or rock outcrops on the Canadian Shield, the sides of the stone passages of New
Grange in Ireland, and the walls of deep caves in France and Spain. 'Rock art' also covers
features created using rocks of different sizes to produce 'rock,' or 'boulder alignments.'
The term 'art' is problematic because it suggests that these images have primarily a
decorative value and no intrinsic value or meaning of their own. lt also implies classification
of these images according to Western notions of high or low art, or, perhaps, a craft. These
terms have loaded meanings, since they impose the analyst's conventional values. Rock
images should not be considered within such a perspective, since, evidently, the cultural
context of the 'reader' or 'viewer' influences perception and classification. This prejudgement
affects how images are understood (Blocker 1994; Conkey 1987; Price 1989).
Not:withstanding these continuaI dangers, scholars interested in the se images
continue to use the term 'art'. Images similar to those found on rock surfaces of the
Canadian Shield are encountered in other cultural contexts (Densmore 1974 [1928]; Phillips
1999; Ritzenthaler and Ritzenthaler 1970). The designation of all these sorts of images as
'art', 'folk art' or 'hanrucrafts' has created problems for their analysts.\ There are no clearly
delineated guidelines indicating the way in which the meaning of such images should be
unpacked. Yet, whether or not these images were intended as 'art', they are a form of
communication, and, therefore, they challenge us to understand them. l rusagree with
Whitley's (2001: 22-23) argument that the term 'rock art' should not be changed since a
western intellectual tradition has used it for over one hundred years. The use of a term for a
long period of time should not justify its continued usage, particularly if the users
acknowledge that problems exist with it. Continuing such a practice or 'tradition' merely
Ieaves the arena open for continuaI disputes and discussions over whether these images are
art or not. l therefore propose that these representations should be termed rock images, or
petroglyphs and pictographs.
Rock image sites cannot be studied using the same techniques as are applied to other
archaeologicai sites. The theoretical approaches used and the questions asked may be the
same but the data sources are rarucally different and generally far more limited. These images
cannot be excavated using the techniques for recovering, cataloguing, and analysing data that
1 Frank (2000: 1-18) in her introduction clearly surrimarises the intellectual arguments, and debates concerning the development of the different theories that have led to different objects and images being termed "decorative" or "flne" arts or handicrafts. Dondis (1973), Gombrich (1984), Lay ton (1991), Morphy (1989), and Rapaport (1997) considered the question of what can be defmed as art, crafts, and visualliteracy. Haselberger (1961) proposed a method for dealing with ethnological art.
Page 3
archaeologists apply to 'conventional' archaeological sit~s. The area surrounding such images
may be excavated but the physical context of the site often provides little or no information
about the meaning(s) of the images themselves.
The subjective beliefs and ideas held by the people who created these images did
more to shape them than technological processes or the economic or political systems in
which these people lived. Therefore, the archaeologist must rely to an unusual degree on a
range of non-archaeological sources in order to establish the meaning of the images. It is
very difficult to access this information for a group whose past is available only through the
archaeological record. The difficulties in accessing the symbolic knowledge of a group of
people through the inherent attributes and physicallocation of such images may explain why
these sites have often been ignored, or merely described, in contrast to similar images found
on birch bark scrolls. Fieldwork and archivaI work must be considered as equally important
in this study, since information must be drawn from a wide range of disciplines, induding
archaeology, anthropology, history, art history, geology, and geography.
It is time to re-examine the fundamental daims regarding rock images and their
study. Rock images have fallen into 'the domain of archaeology because they are found
outside on the surface of rocks, and not on clothes, canvases, portable objects or documents
in archives. My own fieldwork has demonstrated that, contrary to common beliefs, sites in
the Lake of the Woods region did not always exist in conjunction with a body of water. It
also shows that rock images occasionally are found in caves rather than on cliff faces.
Suppositions about the known physicallocations of the pictographs have had a fundamental
influence upon their discovery, and how and by whom they are interpreted (see Brown and
(1978); Nelson (1983); Fulford (1988, 1989, and 1990), and Vastokas (1984, 1986-1987, and
1996).
Page 98
Schoolcraft was an American explorer and topographer from Watervliet, New York
State. He served as a topographer for an expedition led by the American General Cass to
explore the Lake Superior region in 1820. This experience coupled with ethnographie
fieldwork on the Upper Mississippi and in the Lake Superior region led to his appointment
as the Indian Agent for the peoples of the Lake Superior region from 1822 to 1836. From
1836 to 1841 he was Superintendent of Indian affairs in Michigan, where he settled in Sault
Ste. Marie after his marriage to an Ojibwa. He devoted much of his life to the study of
American Indians.
Schoolcraft (1851: 351) observed that the Ojibwa distinguished between
"kekeeowin" and "kekeewin". The term "kekeewin" was defined as a system of marks that
represented Ojibwa clan symbols and other images that were readily recognised and
understood by everyone in the tribe, and were used to leave messages, and on grave markers
to provide information about the person who had died. However, if these images were used
on rocks, Schoolcraft stated that they were cailed "mussinabiks". He (ibid.) defined the term
"kekeeowin" as a system of symbolic notation "for the songs and incantations of the Indian
meda [Midé] and priests, making it appear to the memory for the presentation of language
and musical notes". Schoolcraft described (ibid.) the images on the birch bark scroils as
"picture writing". He (ibid.: 361) observed that the members of the Midewiwin used these
pictographs as mnemonics to remember the words of their chants. These chants were
performed inside the sacred midewegan, 'spirit lodge', during curing rites and initiations.
Schoolcraft recorded these chants and discerned they functioned in a similar manner to
prayers or hymns. He was impressed by the Midé's ability to recoilect both the words and the
music, based upon the images on the birch bark scroils and concluded that each chant had a
key symbol wruch acted as a trigger to a body of information residing in the memory of the
shaman. Unfortunately, he did not identify the key symbols on the birch bark scroils and was
unable to elicit support for future research from the Smithsonian Institution.
Kohl (1985[1860]: xiii) provided sorne information regarding the images on the birch
bark scroils belonging to the Ojibwa Indians who lived on the south shore of Lake Superior
in northern Wisconsin. A German geographer, ethnologist, and travel writer, Kohl spent
four months with them as they adjusted to life on four of the new reservations in northern
Wisconsin. He (ibid.: 145) noticed that the "northern savages" used birch bark sheets of
varying sizes on wruch to write. He described the Ojibwa practice of creating pictographie
Page 99
dream records and suggested that many song scroils had iconographic onglns ln the
Algonquian dream or vision quest experience. He (ibid.: 143-144) observed that, at
Chequamegon, the Ojibwa also le ft records of their dreams on trees, rocks, birch bark,
blankets, and clothes. He (ibid.: 145) coilected, purchased, and copied many of these objects,
sorne which were possibly birch bark scroils. Kohl (ibid. 143-166) established through his
interpreters that the scroils were related to the Midewiwin Society and depicted different
totems as weil as the migration movement of the Ojibwa from the Atlantic to Lake
Superior.5 Unfortunately, his efforts at transcribing the images were not very comprehensive;
perhaps a consequence ofhis attitudes towards the Indians. He wrote that he (ibid.: 165-166)
became annoyed with "these savages" after his attempts to buy scroils from two of them
failed; foilowing an attempt to bribe them with pouches of sugar or tobacco.
Warren (1984[1885]) worked among the Ojibwa of Minnesota and Wisconsin and
was of mixed Ojibwa and Euro-American heritage. He wrote very little about these images
but he stated (ibid.: 67) that the songs and rites of the Midewiwin were transcribed using
"hieroglyphics". The only ex ample of these "hieroglyphics" that he (ibid.: 89-90) discussed
was of sorne images inscribed on a copper disc which recorded the number of generations of
a family of the Loon totem who had lived in the La Pointe area. The indentations indicated
that eight generations had lived ln this region. Warren (ibid.: 26-27) argued in the
introduction of his volume that he intended to conduct a detailed examination and
interpretation of the images used to record the religious idioms and the songs sung, but
unfortunately he died from a haemorrhage before beginning the projected book on the
Midewiwin (Fletcher Williams 1984: 18). Although a practising Episcopalian, he believed that
the Midewiwin was central to the Ojibwa world view.
Three decades after Schoolcraft (1856), the Smithsonian Institution provided funding
for further research into Ojibwa pictures. Hoffman, a Colonel in the US Army, carried out
fieldwork for three months amongst the Ojibwa of northern Minnesota under the
supervision of Garrick Mailory, who remained in Washington D. C. (Hoffman 1891: 156).
Mailory (1893: 202) and Hoffman (1891: 156) were concerned with complex sequences of
symbols and rejected Schoolcraft's (1851: 351) idea that any of these symbols could be
5 The debate regarding the origins of the Midewiwin, the Grand Medicine Society, will not be considered because it is not pertinent to the discussion. However, it has been discussed in sorne depth by others including Angel (2002), Bhar (1991), Balikci (1956), Dailey (1958), Grirn (1983), Hallowell (1936), Hickerson (1962, 1963, and 1970), Hoffrnan (1891), Landes (1968), Meyer (1990), Vecsey (1983, 1984); Rogers (1973).
Page 100
deciphered using a key. They rejected his distinction between "kekeeowin" and "kekeewin",
believing that he exaggerated many of his observations. Hoffman's (1883) paper focused
principaily on comparing "Eskimo" pictographs with those created by other indigenous
groups. He argued that the Ojibwa used birch bark for mnemonic records and personal
exploits (ibid.: 128) and that the style of drawing used by the Alaskans was "very similar to
that of the Ojibwa" (ibid.: 135). He (ibid.: 135) examined the images found on the stem of a
pipe presumably made by the Ojibwa, but did not distinguish between the images that could
be used and read by ail the members of the Ojibwa community and those images that only
the Midé could use. He (ibid.: 139) claimed that the method of recording transactions and
events on birch bark scroils was "almost, if not entirely discontinued". He therefore posited
that it was crucial to coilect as many examples as were stiil being used, in order to increase
the quantity of information concerning these images. Nevertheless, he concluded that the
Eskimo images were "greatly superior" to those made by any other group, especiaily with
respect to the "delineation of natural objects, especiaily animate forms" (ibid.: 146). He
placed the Ojibwa bark records second among ail the indigenous peoples' imagery. He also
stated that only through a thorough familiarity with Ojibwa "ceremonies and mythology"
could any understanding of these mnemonic images be obtained (ibid.).
Much of the information in Hoffman's (1888) article was republished almost
verbatim in his (1891) report for the Bureau of Ethnology at the Smithsonian Institution.
Yet, this 1888 article discusses two other birch bark scroils that were not described in the
1891 report. Both scroils came from Red Lake, Minnesota, and, although he did not ob tain
assistance in transcribing them, he made a copy without the knowledge of the Midé to whom
the scroils belonged. He (1888: 216) observed that the characters on both scroils were
identical and that one appeared to be copied from the other. He also (ibid.: 217) claimed to
have seen an ancient cosmological chart that apparently had never been shown to any white
man or to any non-Midé Indians. Hoffman described this map, its physical condition, and its
images but unfortunately did not provide any analysis of the images themselves.
Hoffman (1889: 215) discussed the meanings of the images, the mnemonics, on a
Midé birch bark seroil that had not been shown to a "white man" before in Red Lake in
northern Minnesota. Whether this is the same bireh bark seroil diseussed in his previous
article is diffieult to determine. Apparently, this seroil was not shown to an Indian unless he
had paid the required fees, and then only once he had undergone the necessary training and
Page 101
instruction to be initiated into the Midewiwin Society (ibid.). This seroll belonged to
Skwekom' -ik who had received it from his father-in-Iaw who in turn had received it from
Ba-da'-san the Grand Shaman and chief of the Winnibigo'shish Ojibwa. Ba-da'-san himself
had obtained this scroll, which was a copy of the original, from the Grande Midé at La Point
where the Midewiwin was held annually and ceremonies were conducted strictly in
accordance with ancient and traditional rites. Hoffman (ibid.: 216-217) subsequently
examined this scroll and a few others. He realised that, if a shaman was only a second degree
Midé, he could not actively participate in ceremonies that went beyond Midé of the second
degree. It also meant that he was not privy to information known about these same images
by a Midé who was also higher than second degree. Hoffman (ibid.: 218) discussed at length
the differences in the colours used in the sacred poles used by the Midewiwin based upon
information from Sicos 'ige, a second degree Midé. The remainder of this article considered
the facial decoration used by initiates into different degrees, the ceremonial procedures, the
types of medicine bags used by the Midé of different degrees, what occurred at a Ghost
Lodge, and finally an event where a Midé treated and brought back from the Spirit Land a
dead young man.
Hoffman's (1891) monograph discussed the collection of twenty-four song scrolls
used in religious ceremonies of the Midewiwin at the White Earth Reservation in Minnesota
during the late 1880s. He published reproductions of twenty of these scrolls along with
several other examples of images. Hoffman considered the various types of shamans existing
among the Ojibwa (ibid: 165) and stated that the Midé and Jiissakid were the most highly
regarded individuals in the community for their curing and medicinal skills. Both skills were
achieved either by visions alone or by visions and purchase. Midé priests (ibid.: 165)
customarily preserved birch bark records with incised lines that pictorially represented the
ground plan of the ceremony and the number of degrees to which the owner of the scroll
was entitled when he conducted a ceremony. Severallevels of knowledge and understanding
existed because of the hierarchy within the Midewiwin Society, and therefore some Midé
were more knowledgeable and powerful than others. This, in turn, affected the meaning of
the images depicted on birch bark scrolls. Hoffman (ibid.) established than these records
were sacred and could not be exposed to public view; they could be examined only after a
candidate had paid his initiation fee, fasted, and provided offerings of tobacco. These
pictographs were used by the Midé as mnemonic devices to remember the words of their
Page 102
chants, for the instruction of new members, to record oral traditions, and to correct
ceremonial procedures (ibid.: 191-193). Hoffman (ibid.: 267) established that the seroils
eould be read from left to right, right to left, and sometimes some seroils used both styles.
He observed that the Ojibwa used images to express ideas by shamans, hunters, and the
traveilers. However, a reeognised system of using these images that enabled one person to
completely comprehend the work of another did not exist. Essentiaily, an image acted as a
key symbol and behaved as a trigger of information, a mnemonic that resided in each
person's memory. He observed (ibid.: 289) that the songs sung by the shamans were based
upon the characters recorded on the seroils and in many cases words or phrases were not
recited. Unfortunately, he did not provide any analysis concerning the images themselves.
Densmore's (1910) publication discussed the song scroils coilected during 1907, 1908
and 1909 from the Chippewa lndians of White Earth, Leech Lake, and Red Lake
reservations in Minnesota, and from a Chippewa who lived in the Bois Fort reservation in
Minnesota, and from a Chippewa who lived in the Bois Fort reservation in Minnesota. She
discovered that the melody was more important than the words and that "in a succession of
renditions of a song" the words were often only used once (ibid.: 2). She determined that the
ide a of the song was firmly assoeiated with the melody and not with the words (ibid.).
Densmore observed that it was "permissible and customary" to compose new words for old
tunes, but she noticed that the words were "always similar in general charaeter" to the words
previously used. She concluded that although changes could occur and be introduced, the
identity of the songs remained unaffeeted. Densmore asserted that these tendencies did not
apply to songs used by the Midé.
Densmore (ibid: 8-11) considered the structure of the songs sung and established
that certain songs were harmonie and others were melodic in structure. The results she
presented in tables. She observed that the songs used by the Midé required a "special skill" to
translate (ibid.: 8) and that different songs were used when singing the Midewiwin (ibid.: 11-
12). She (ibid.: 13-14) also considered the beliefs of the Midewiwin, their goals, and aims, and
argued that their songs "represented the musical expressions of religious ideas" (ibid.: 14).
This meant that she often found translating the words very difficult. She established that a
Midé song was not considered fini shed unless it ended with the syilables "ho ho ho ho"
(ibid.: 15).
Page 103
Densmore (ibid.) established that the beliefs and the teachings of the Midewiwin
were the words of the songs, which often functioned as texts for discourses (ibid.). The
songs were only taught to those who paid for the privilege of learning them and the words
were recorded in mnemonics on birch bark. Densmore (ibid.: 16-17) ascertained that
members of the Midewiwin universally understood the images, or mnemonics, of the Midé.
However, she noted that these mnemonics could be used in other circumstances. She (ibid.:
20) also considered the role of the Jiissalcid (known as doctors or jugglers - a different type of
shaman amongst the Algonquians) who were different from the Midé. Densmore recorded a
wide range of songs, the medicinal songs which the Midé, rather than the Jiissalcid used. She
established that, for a charm to work properly, the correct song and the correct medicine
must be used. Therefore the person who bought the medicine from the Midé also bought the
song.
Densmore's (1913) publication continued from her earlier (1910) study of Ojibwa
music but only dealt with mate rial from the White Earth reservation in Minnesota. She relied
upon numerous informants for translations, including Mrs. Mary Warren, Rev. C. Beaulieu,
and Rev J. A. Gilfillian. Both ministers helped with the analysis and translation of the
Chippewa words. Rev. C. Beaulieu was a member of the White Earth Chippewa, while Rev J.
A. Gilfillian had lived on the reservation for twenty-five years.
She discovered that the rhythm was repeated more exactly than the melody. She also
discovered that a number of the songs did not have words and that, if songs had words, they
"suggested rather than expressed" the ide a of the song (ibid.: 2). If the words changed, they
conformed to the music, but the same word could be accented differently in different parts
of the song. She (ibid: 15) asserted that the principal reason for singing a song was directly or
indirectly to seek supernatural aid. However, a large variety of songs existed for different
reasons. She (ibid: 17) clarified that most of the old songs that had a major tonality included
the old "dream songs" used by the Midé, the practise of medicine, the pursuit of war, and the
essential nature of a dream associated with the idea of reliance upon supernatural help.
However, songs that expressed sadness, distress, or uncertainty used a minor tonality (ibid.).
Densmore's (1979[1929]) publication discussed "picture writing", which she (ibid.:
174) defined as "that form of thought writing which seeks to convey ideas by means of
picture signs or marks more or less imitative of the object of the idea". Significance was,
according to Densmore, an essential element of the pictographs and she maintained that they
Page 104
were "alike in that they all express thought, register a fact, or convey a message." She (ibid.)
observed that figures included "crude delineations" of men, animaIs, birds and other
"material objects" and that the Midé used symbols that represented the sky, earth, lakes, and
hills, and sounds including concepts such as "spirit power". Simpler symbolism was used for
messages, including signs that indicated days, direction and duplication of numbers.
Occasionally these were combined with possible ideographs to indicate "progressive action".
Densmore maintained that two types of writing existed: esoteric and non-esoteric.
The esoteric writing could only be understood by initiates, while the non-esoteric writing was
used by everyone to convey information. She argued that examples of esoteric writing
included the Midé records, writings and songs, stories of Winabojo, and drawings used to
work charms. Non-esoteric writing included "totem marks," messages left along a route and
maps used by travellers, illustrations for a narrative, records of time, as well names of
individuals (ibid.: 175). She developed and expanded upon each type of writing using
drawings.
Skinner's (1914) article briefly considered how the Plains Ojibway, known as the
Bungi, employed birch bark. He (ibid.: 317) observed that they used "picture writing", as he
termed it, on birch bark and wood, which in turn reflected this group's origins in the
Woodlands region. Skinner argued that this type of writing was connected with religion and
that the Midewiwin was important. However, he did not describe the images, who used
them, what they may have meant, or how their meanings were transferred from one pers on
to another.
Several very descriptive articles provide little analysis of the images on the scrolls (for
example Cadzow 1926; Kinietz 1940, Reagan 1921 and 1922). Reagan conducted fieldwork
while he acted as lndian Agent at Nett Lake in Minnesota. His (1921) article published some
medicinal recipes made from plants. George Farmer, Ne-bay-day-ke-shi-go-kay, "a medicine
man" of the Bois Fort in Minnesota allowed Reagan to copy them from his notebook.
Reagan (ibid.: 246) stated that the recipes were written in the Chippewa language using
Roman characters. Reagan also (1922: 332) published the words of the songs used by the
Midé, which he found were recorded by the same "medicine man". He (ibid.: 332) observed
that Farmer had used Roman characters, but that the "phonetic values were similar to
French". Reagan published a few translations of the songs, recorded in Farmer's notebook,
stating their meaning, and when in the Midewiwin ceremony the songs were used. At the end
Page 105
of his (1922: 366-369) article, he provided a brief description of the Midewiwin Society in
this region. Both articles are interesting, since it is possible to establish what these songs
meant. However, it would have been useful to know which song was connected to which
mnemonic and whether songs could exist without mnemonics.
Kinietz (1940) unsuccessfully sought birch bark records while conducting fieldwork
among the Chippewa of northwestern Michigan over several seasons. He (ibid.: 38) asserted
that the type of symbols used, the size of the record, and "the excellence of the delineation"
affected the type of record. He argued that "the excellence of the delineation" varied not just
with the purpose of the designs, but with the artists of the different bark images.
Unfortunately, he did not provide any indication as to his methodology or evidence that
could substantiate his claims. However, he distinguished two types of birch bark scrolls
based upon subject matter. The first, understood only by instructed persons called Midé,
were, according to I<inietz (ibid.), "ideographic, esoteric and mnemonic". The second type of
scroll he described as being descriptive in character, and was related to the sign language of
the Indians. Unfortunately, he provided little evidence to substantiate his daims. Otherwise,
he reiterated many of the findings of early ethnographers and preceding researchers
regarding both scrolls and the Midé.
Voegelin (1942: 44), unlike I<inietz (1940), stated that she was successful in her
research. Voegelin (1942: 44) conducted her fieldwork amongst the northern Minnesota,
Wisconsin, and eastern Ontario Ojibway. She maintained that pictographs were used to
convey messages, provide topographic information, and "illustrate narratives". She observed
that the Midé used pictographs to record esoteric history or songs. She (ibid.: 45) established,
based on information from informants, that, in post-contact times, a pointed piece of metal
was used to cut images into the birch bark while in "pre-white times" a wooden or bone
stylus was employed. She was also told that red paint made from blood root often filled in
the outlines of the incised figures (ibid.: 44-45). Voegelin (ibid. 45-46) described and
identified a number of images found on the birch barks scrolls using information gained
from her informants. She observed that when she asked her informants for the terms
"picture, photograph, pictographic representation and so on" (ibid.: 46) that the words given
for each of these terms were numerous and varied considerably between different dialectic
groups, especially on the edges of the Ojibwa and the Ottawa groups.
Page 106
Coleman's (1947) publication examined, discussed at length, and compared the
images from two birch bark scroils as part of her discussion of how the Ojibwa in northern
Minnesota used images in religious contexts. She established that these scroils were read
from left to right (ibid.: 79). One, a Midewiwin scroil, belonged to her informant Charlie
Fox, a Midé who also was the leader of the Midewiwin at Naytahwaush (ibid.: 77). The
second scroil belonged to a Mr Gustavson, an Ojibway from the White Earth reservation.
She observed that the images on the second scroil were larger and more finely delineated, but
that the details of the "faces were lacking and the manitos were not nearly so artisticaily
executed" (ibid.: 80).
Both Blessing's (1963) and Kidd's (1965) papers indicate that archaeologists were
also interested in birch bark scroils. Blessing's (1963) study of a coilection of birch bark
scroils differed from Kidd's (1965) and indicates that he was determined to establish the
meaning of these images on the birch bark scroils. Blessing (ibid.: 91) used a number of
informants who were Midé, sorne of whom were from Madeline Island, Wisconsin and had
since died. His principal informant was Thomas Shingobe, a Midé and member of the Miile
Lacs band whose ancestors originaily came from Madeline Island, down the St. Croix River
and into Minnesota. Blessing (ibid.: 93) argued that the Midé were members of a "cult" that
was "more or less an elaboration of earlier hunting societies which had been formed in the
pressing need for the organized gathering of food". He continued his line of argument by
stating that, as religious activities grew increasingly multifaceted, it became necessary to
devise a method of keeping records. Hence the individuals involved "utilized natural and
conventionalised forms" as mnemonic devices. He posited that the scroils from Minnesota
no longer served as memory aids but as a means of communicating information. These
mnemonic images remained "almost exclusively the property of the originator," but, later,
became part of the larger group. Among the Ojibwa (ibid.: 94), these images became more
complex and "approached the stage of hieroglyphics." He considered the scroils to be old.
Blessing further asserted that the largest number of scroils had been produced in Minnesota.
How he arrived at this conclusion is unclear. Blessing examined scroils from Minnesota,
Wisconsin, Michigan, and Canada, but it remains unclear where the Canadian scroils
originated (1963: 95). He established that the eIder Midé took time and effort in interpreting
the images on the scroils during the instruction of the younger Midé. It was in this manner
Page 107
that the meanings and significance of the images were transmitted from one generation to
another.
Blessing considered the Midé, the Midewiwin, and their activities in Minnesota in
considerable depth: from where and how the Ojibwa had come to settle where they did in
Minnesota, what were guardian spirits to the Ojibwa and the Midé, and how a scroil was
made and inscribed (ibid.: 100-109). Blessing argued that he had accumulated substantiai
quantities of information on the practices of the Midewiwin from two reservations: White
Earth sometime after 1946 and Miile Lacs in 1968. U sing informants, he (ibid 109)
established that the "minor" animal species were associated with lower degrees of the
Midewiwin and that the "rugher species" of animaIs were connected with the higher degrees.
For example, weasels, otters, and mink were associated with the first degrees wrule mytrucal
creatures were associated with the fifth to eighth degrees (ibid.: 109-110). Scroils from
different geographic areas exrubited slight differences.
Blessing (ibid.: 110) established that the scroils in his study consisted of three types:
30% were "instruction" scroils, 50% "ri tuaI" scroils, and 20% were "master" scroils.
Candidates used an "instruction" scroil to prepare for the initiation ceremony (ibid.: 110-
111), while the Midé who conducted the Midewiwin ceremonies employed a "ritual" scroil as
a "weil illustrated edition wruch served as a guide or a reference" (ibid.: 111). The "master
scroil" was cailed an rustoricai document but also indicated the four to eight degrees which
could be obtained by the Midé. Blessing subsequently examined seven scrolls wruch included
ail three of these types. He provided detailed drawings or tracings of seven scroils.
Unfortunately, by using the terms "natural" (ibid. 96) and "invented" to classify sorne
images, he demonstrated a potential problem in rus study since rus own judgement seems to
interferes in rus description.
Kidd's (1965) very descriptive article discussed two scroils discovered witrun an
archaeological context. Kidd wished to make archaeologists aware that the discovery of this
type of object could occur in regions where birch bark was used. P. Ryan accidentaily made
the first find in 1924, reporting it to Kidd in 1959 (ibid.: 480). The first group of birch bark
scroils were found when the water from the fire fighter's hose washed away the soil of a
mound on the north shore of Rainy Lake that covered them. Ryan recailed that it contained
probably around 200 to 300 hundred pieces of birch bark (ibid.: 481). W. T. Daigetty and
Bruce Littlejohn made the second find in 1962 in a cave on Burntside Lake in Quetico
Page 108
Provincial Park. Kldd examined both sites and both groups of bireh bark. He (ibid.: 483)
eoncluded that "seroils" and their assoeiated artefacts eould be found buried in the ground in
places seemingly "out of the way" in the woods, or hidden in caves. He eould not date when
this happened for either group of bireh bark objeets but maintained that arehaeologists
should be aware of the possibility of similar diseoveries in the future.
Chapter III eovers Dewdney's and Kldd's (1962, 1967) examination of the rock
image sites of the Canadian Shield in depth. However, Dewdney (1962: 110 -116) posited
that a eonneetion existed between the images of rock image sites and those made by the
Midewiwin, based upon their similarity. Dewdney eompared a few of the pietograph and
petroglyph images with those of the bireh bark seroils that were known to belong to the
Algonquians of the Great Lakes region. He eoncluded that a study of the depietions of
supernatural beings in the mythology of the Algonquians might aid the identification of these
images. Over ten years later, Vasto kas and Vastokas (1973) suggested that a eonneetion
existed between the engravings of the Peterborough petroglyphs, the images on the bireh
bark seroils, and Algonquian shamans (please see Chapter III for a detailed examination of
their study).
Landes (1968) eonsidered images on bireh bark seroils in her examination of the
Ojibwa religion and the Midewiwin. She (ibid.: 172) stated that only the shaman who ereated
the seroil understood the specifie meaning of the images used. It was the shaman's role, she
(ibid.: 172 - 173) asserted, to teaeh "this explieitly" to someone else "before the notation was
useable"; the shaman's intentions dietated the level of knowledge or seereey. She (ibid.: 224)
deseribed bireh bark seroils and their images, and stated that its intention and, frequently the
availability of bireh bark, determined the size of a seroil. She (ibid.) argued that the
"partieular message" of a seroil, essentiaily, what the pietographs meant, was probably
impossible to establish unless eaeh figure was labeiled or translated by the owner.
Dewdney's (1975) publication is valuable, although it was eriticised by Rogers (1975)
and Veesey (1976). Dewdney (1975: 21-22) posited that six categories of seroils existed based
on the variations in the themes he had identified in ail of the seroils that he examined. This
list included: (a) origin seroils, (b) migration eharts, (e) mas ter ritual seroils, (d) ghost lodge
and sky degree, (e) deviant seroils, and (f) enigmatie seroils. He asserted the importance of
remembering that the principal funetion of the bireh bark seroil was mnemonie. From 1966
to 1972 he undertook to identify and traek down over 150 bireh bark seroils in public and
Page 109
private collections in Britain, Canada, and the United States of America. He noted their
current physical location, type of scroll, its catalogue number, where it came from, who
collected it, its length in inches, number of sections, and a source code. He recorded the
images of sorne of the scrolls using a technique that he described in depth, whereby he traced
the scrolls, or, if a collection was deemed too large, only the "most important specimens"
(ibid.: 7) and sketched the remainder. He did not indicate the process by which he
determined which scrolls were more important. Dewdney (ibid.: 5) distinguished between an
instruction scroll or chart and what he called a song scroll. He (ibid.) argued that an
instruction scroll was a linear series of images to be utilised as memory aids "in rendering
either a sacred Midé song, or a profane personal one." The profane song was dreamed by a
"visionary shaman" for a various purposes, while a Midé composed a sacred song, which
could be sold or traded. However, he believed that the contents of the latter could only be
interpreted by the composer of the scroll or the owner. He (ibid.: 15-18) described in
considerable depth how scrolls were constructed from birch bark, and the possible
techniques used to draw the images on their surface. It remains unclear who made the
scrolls, but Dewdney maintained that the task of stitching of birch bark together to create a
larger scroll was 'traditionally - but not always - women's work." He did not provide any
evidence to support this statement. The stitching cou Id be "highly decorative" (ibid.). He
conducted an in-depth analysis of the styles used to depict the images, using terms such as
"rectilinear, abstraction, condensation, symbol conversion, atrophy, amputation" (ibid.: 18)
to describe the processes by which shapes changed as the scrolls were repeatedly copied.
Unfortunately, his terminology makes it very difficult, even impossible, to comprehend
precisely what information he wished to convey to the reader. His terminology is,
consequently, difficult to use. He also transcribed a number of the scrolls with the aid of
James Redsky from Shoal Lake Reserve, on the western portion of the Lake of the Woods,
relying on an interpreter when Redsky used Ojibwa rather than English.
Rogers (1975: 83) reviewed this book and asserted that Dewdney frequently allowed
his own perspective to colour the way in which he classified the spirit world of the Ojibwa.
Rogers argued that present day Ojibwa "do not" and never did adhere to the Euro-Canadian
worldview. According to Rogers (ibid.), the Ojibwa believe that the same spirits were helpful
in one context while they were injurious in another. He cautioned the reader of this book to
be wary of Dewdney's (1975) interpretations of the significance of the symbols on the
Page 110
/~.,
scrails. Rogers (1975) is considerably harsher than Vecsey (1976) in his review of Dewdney's
(1975) book but both reviewers severely criticised Dewdney's study for numeraus
weaknesses, which they ascribed to his laek of scholarly expertise.
Nelson (1983) diseussed a eoileetion of bireh bark scrails and objeets of unclear
origin. His article pravided a detailed review of previous work undertaken on these scroils by
Hoffman (1891) Sehoolcraft (1854), and Densmore (1979 [1929]). He also examined a
number of serails illustrated in various gailery catalogues. He (1993: 220-221) listed the
serails that Dewdney had not reeorded that were in other publications. Nelson subsequently
used Dewdney's (1975) publication in conjunction with Blessing's (1976, 1977) to identify
and de scribe at length the different types of serails that he had identified using Dewdney's
classification (1975: 21-22).
Vastokas (1984: 425), an art historian, acknowledged the work of her predecessors
but asserted that the interpretation of the images on the birch bark scrails stiil presented "the
most pressing chailenge to scholars" interested in Algonquian art history. She eould have
included archaeologists, anthrapologists, and historians. Vastokas (1984: 435) maintained
that the scrails were the most significant body of visual documents available for the study of
native North Ameriean art history. This paper outlined sorne preliminary and exploratory
observations in interpretation theory and methods whieh she maintained were applicable to
bireh bark serails. She (ibid.: 426) postulated that the style, iconography, and development of
the serails should be eonsidered from an art historical perceptive. Vastokas believed that,
since further fieldwork was impossible (ibid 427), two methods of researeh existed to
interpret these serails: (i) examination of the images depicted on the scrails themselves, and
(ii) external praof supplied by comparison with analogous or related visual systems.
She subsequently outlined each avenue of research and argued that data could be
obtained thraugh detailed examinations of the manner in which the motifs and elements
were organized on the scroils. Art historians, she maintained (ibid.: 428), eould use the
notion of style as a tool. Preliminary observations of the style in which these images were
executed, Vasto kas eontended, highlighted clear differences between the official Midewiwin
scroils and the personal song seroils that belonged to individual members, and these in turn
suggested significant differences in their funetions and meanings. She maintained that these
differences were observable without ailuding to the cultural eontext of each scrail. The most
intrinsic and explicit virtue (ibid.) was their narrative structure in which separate pictorial
Page 111
units were arranged to convey information to the beholder regarding mythological, historical,
or ritualistic events across time and space. This narrative role was crucial in any analytical
procedure aimed at interpreting the "form" or content of each scroil's images; the form
being the physical outline of an image. She maintained that narrative art differed from iconic
representations since, unlike iconic representations, narrative art could communicate events
in time and space. Narrative compositions she (ibid.: 430) defined as multiple motifs
organised in a "more perceptuaily scattered disposition ailuding sequence and movements to
the eye across the surface of pictorial ground." Furthermore, profile images denoted motion,
while frontal views indicated action. Vastokas (ibid.: 431) postulated that narrative art forced
the observer to detect the greater visual complexities of the composition. She concluded that
the birch bark scroils were "narrative visual wholes" to be interpreted in terms of their form
and context. She disagreed with early endeavours to decipher the images, such as Mailory's
(1893) attempts to examine them as writing systems, because these scroils were visual rather
than verbal documents.
Vastokas' (ibid.: 427) second avenue of research, outlined at the beginning of her
paper, was that external proof should be supplied by comparison with analogous or related
visual systems. The Midewiwin scroils play crucial roles as visual religious texts, as mnemonic
aids in the instruction of new members, and as a means to record oral tradition and correct
ceremonial procedure. Essentiaily, they recorded the history, origins, and ritual of the
Midewiwin Society. It was vital, therefore, to understand these scroils within the larger
context of Ojibwa shamanism and the numerous scroils that record the songs of individual
Midé. The song scroils contained, Vasto kas maintained, links with traditional motifs, since
personal scroils habituaily used traditional motifs that derived from personal vision quests.
Advances in the interpretation of the images, Vastokas (ibid.: 437) concluded, could occur if
theories and methods were utilised from a range of fields including art history, symbolic
anthropology, history of religions, and philosophical hermeneutics. Future examinations of
these images, she asserted, would be advantageous if they were undertaken in the context of
a circumpolar shamanic art tradition, since it had a cross-cultural system that was fairly
consistent.
Vennum's (1978) study could be placed in Vastokas' (1984: 427) second line of
research. He drew extensively upon the data and the songs of the Mide collected by Blessing
from the Mille Lacs and White Earth reservations. Vennum argued that larger numbers of
Page 112
Mide songs existed in this coilection than Densmore had coilected. He was interested in the
origin migration scroils of the Midewiwin, and so, he reiterated the legend of the Ojibwa
westward migration. He argued this had to be undertaken because the migration charts
"pravided the necessary dues for the identification and the interpretation of the origin
migration" serails. Hence he pravided a considerable quantity of information about this type
of chart and about understanding its raIe within the Midewiwin. Vennum asserted that this
type of serail was distinguishable fram other serails beeause images behave as mnemonics
for the place names of "real or legendary resting places" used by the Ouer or Bear as they
traveiled westwards (ibid.: 761), while in other serails these places were presented by dots.
Serails were read from left to right, sin ce the Ouer, in the origin-migration story told by the
Midé, traveiled from the east to the west. Vennum argued (ibid.: 753) that the development
of the origin-migration story was linked with the migration of the Ojibwa as they moved
from the east to the west side of Lake Superior before the end of the seventeenth century.
Vennum diseovered that twelve migration charts existed in the Miile Lacs-White
Earth papers coilected and transeribed by Blessing (1963) and his informants. Vennum
(1978: 762-763) identified the place names, and was able to establish precisely where on Lake
Superior they had stopped. He argued that this information "chailenged" Dewdney's (1975:
9) statement about the indigenous knowledge of Lake Superior, while corroborating and
expanding the versions of the migration coilected by Warren (1984 [1885]) and Hoffman
(1891) during the nineteenth century. However, Vennum (1978: 764) could not identify ail
the images as place-names, and conduded that these were probably the "texts of songs". He
established that specifie motifs from sorne song scroils bore a striking resemblance to those
surviving in the song texts, suggesting that regularities also existed in their meaning. He drew
extensively upon Densmore, Schoolcraft, Kohl, Hoffman, and Landes (1968) to identify the
songs and discuss how they might be interpreted. Vennum (1978: 787) argued that these
migration serails may have been performed near the end of the Midewiwin ceremony, based
upon information in the Miile Lac-White Earth papers. He subsequently argued, based on
Blessing (1963: 93-94), Hoffman (1891: 290), and Densmore (1910: 26), that the images, as
mnemonics, "were less generalized and their meanings secretly guarded" (1978: 788).
Vennum (1978: 789) suspected that this could explain why Densmore (1910: 26) had
observed that two Midé may have had the same song, but posited that this occurrence was
possibly a coincidence. Based upon the Mille Lac-White Earth papers, Vennum believed
Page 113
(ibid.) that, in the post-1900 period, shared ownership of the songs began occurring in the
pictographs and the texts, but he was unsure as to whether the songs also shared melodic
similarities. He observed that Densmore had not considered whether her informants
recognised the songs by their texts, their images, or, as he termed them, pictographs, or by
their music. He (1978: 790) concluded firstly, that the Mille Lac-White Earth papers made
possible an understanding of the wide variety of ways in which the same Midé song could be
utilised, and, secondly, that a detailed examination of these songs was an excellent start for
anyone attempting to establish the meaning behind other "abstruse Ojibwa song scrolls."
Finally, he concluded that details found in the migration scrolls were useful for interpreting
Ojibwa pictography and song poetry in general.
Closs' (1986) article is the only one that built upon Dewdney's (1975) work with
birch bark scrolls. It examined the ritual importance and use of numbers, especially the
number 4, for the Ojibway, and in particular for the Midewiwin. He utilised the ethnography
(1968), and Warren (1885). Closs (1986: 181) followed Dewdney's (1975: 12) definition of
pictography, which was used as "a generic term for any form of preliterate art-executed on
any available surface - that is known, or is assumed to have had communicative rather than
decorative or aesthetic intent". He (ibid.: 182) agreed with Dewdney and argued that this
pictography was mnemonic and "did not represent the written word." Closs applied the
information concerning the use of numbers to the drawings of six birch bark scrolls. He
(1986: 189) examined six Midé scrolls that he believed exhibited "graphic properties of the
ritual importance of the number 4" and "illustrate the usage of the number series in the Midé
cult." He argued that five of these scrolls were, according to Dewdney's (1975: 21-22)
classification system "master scrolls," while the sixth was a "possible migration scroll."
Closs' argument is interesting, but it remains unclear whether he examined the
original scrolls, or used drawings published by Densmore in her 1929 publication and
Dewdney in his 1975 publication. It is important to examine the actual scrolls rather than a
drawing or sketch of the scroll, because it is possible that the person who made the drawings
inadvertently made mistakes in copying the images.
Vastokas (1986-87) stated that any investigation of prehistoric and post-contact
native North American art presented new challenges for art historians, because it lacked
written documents, which, she acknowledged, were crucial for historians. New art historians
Page 114
(ibid.: 13) could examine these art works as primary data because they did not recognise the
lack of written documents as an obstacle. New art history allowed academics from different
scholarly areas to draw upon a wide variety of disciplines, themes, and methodologies.
Vastokas, like Rlfkin (1986), argued that new art history assumed the universality of human
visual expressiveness and adopted a cross-cultural perspective in which to explore art from a
point of view held by a particular social group, time period, or culture and instead of a
twentieth century perspective.
Vastokas (1996: 48) developed ideas introduced in her previous article. She argued
that images had been insufficiendy exploited in the writing of Aboriginal history and
suggested that these visual documents should be carefully drawn upon as "aids" when
writing Aboriginal history. She promoted the idea that much of Aboriginal art was produced
intentionally as "history", so that pictorial depictions should be recognised as writing.
Aboriginal artworks, she maintained (ibid.: 49-51) had different purposes in traditional
culture, and, since the arrivaI of the Europeans to North America, new functions were
attached to these objects (also see Phillips 1999).
Vastokas identified the problem as stemming from the viewpoint of European
historians and art historians who held that Aboriginal societies lacked a history despite the
fact that Aboriginal societies often recorded it. She posited that previous academics had
ignored the most important function of pictorial representation in traditional native culture
which was the recording of history. According to these scholars, Aboriginal peoples lived in
an "anonymous and static present". Vastokas argued that this attitude had affected both the
interpretation and presentation of the Aboriginal artworks and Aboriginal history.
According to her, there were four different ways by which Aboriginal art works
"served" the writing of Aboriginal history: firsdy, as "historical documents", in the same
manner as a textual document used by an historian with its own date and characteristics,
enabling the development of historical chronologies (ibid.) (an ide a she had introduced in her
1986-87 article); secondly, as "cultural expressions" that were visual embodiments of
Aboriginal beliefs, values, and society as recognised by practitioners in new art history;
thirdly, as "representations" of events taking place in time and space; and finally as "writing"
(1996: 52-53). She reiterated her idea that Aboriginal art in North America could be
subdivided into two dis crete functional types called iconic and narrative that she had
discussed at length in her 1984 article.
Page 115
Vastokas (1996: 53) supported Boone and Mignolo's (1994) daim that it was
important to get away from the modernist \Yj estern conception of art as something to be
appreciated strictly aestheticaily and to consider seriously the idea of visual representations as
communication. Vastokas (ibid.: 53-54) subsequently developed her argument regarding
Aboriginal art as "writing," stating that there was a good reason why aboriginal people
referred to pictographs and pictorial representations as "writing." She (ibid.: 54) daimed that
for Aboriginal North America the birch bark images, or "birch bark manuscripts," were the
most noteworthy expressions of the relations between image and word.
She described in considerable depth the structure and role of the Midé and of the
Midewiwin Society for the Ojibwa, the role of the birch records for the Midé, the range of
their physical dimensions, the range of images being used "from depictive to symbolic, to
entirely abstract in character" (ibid.: 54). Vastokas cited Boone and Mignolo's (1994) daim
that the terms writing and art, if applied to most of Aboriginal America, were problematic
since, unlike western and other phonological writing systems, the goal in the Americas was
not "visible speech." In further support of her supposition, Vastokas (1996: 57 ) cited Boone
and Mignolo's proposition that Pre-Columbian American art and writing were largely the
same thing.
For Vastokas (1996: 57), the Ojibwa birch bark scroils belong to "semasiographic
writing systems" which communicate ideas independently of language.6 She asserted that, if
scroils could be accepted as examples of "semasiographic writing systems," historians could
obtain an insider's view of native American history. Boone and Mignolo had argued that,
within these systems, a subset of iconic writing systems exist composed of representational
pictographic units that transmit meaning without employing speech. Anyone wanting to read
these pictographic units already knew the general meaning of individu al units and so, could
ascertain the meaning of the different combinations.
However, this notion does not distinguish between the fact that different groups of
people might have the knowledge to use certain images and not others. It is impossible under
Boone and Mignolo's definition that a general vocabulary of pictographic units exists that
can be subdivided into groups which are in turn recognised by the general population as
6 Warkentin (1999:3) argued that historians of writing divide sign systems into two groups: semasiographic language systems (pictography) and phonographic (language based) systems. She asserted that phonographic systems were further subdivided into logographic (where the sign is the equivalent to the word), syllabic (the sign is a syllable) and alphabetic systems.
Page 116
having sorne meanings. Each group of pictographie units is used by either ail or specifie
groups, within that society, for different purposes and reasons. Therefore, sorne of these
pictographie units are not understood by ail.
Boone and Mignolo argued that, not pictorial and conventionalized images, but the
placement and context of the images carried meaning. They (1994: 18) asserted that images
behaved as texts in semasiographic writing systems; however, these writing systems did not
have fuil running texts but a highly conventional set of symbols with meanings similar to
those associated with paintings.
Vastokas (1996: 57) maintained that Boone and Mignolo's (1994) conclusions
regarding Mixtec and Aztec writing systems were pertinent to the pictorial narratives of the
Algonquians, since the scroils were among a variety of visual narratives that historians could
use to examine native American history. Therefore, if the images on the birch bark scroils
were accepted as "semasiographic writing systems," these scroils could be accepted by
historians as one of a number of visual narratives that provide an insider's perspective on
native American history.
The consequences of this claim are interesting, since they mean that an image can be
treated as valid a primary document as is a piece of text. Vasto kas stated that considering the
scroils as documents "merely presented a different set of historical chailenges" (ibid.: 57).
She believed that the "most obvious" problem was "the absence of absolute dating" (ibid.).
She believed that additional research was required into the oral traditions that accompanied
visual documents, and into the contexts in which they were utilised. She acknowledged that,
although information had been lost, a considerable body of information still existed.
Vasto kas also lamented that "very few elders" among the Aboriginal community were "stiil
in touch with the traditions" that created these images (ibid.: 58). Vastokas remained
optimistic that fundamental research into these birch bark scroils could be conducted,
despite these considerable obstacles, especiaily if Native and non-Native historians workcdin
conjunction with each other.
Fulford's (1989, 1990) articles fail into Vastokas' first avenue of research. His
findings demonstrate that use fuI and interesting information can be obtained if a detailed
examination of these images is undertaken. He considered the shapes of the twenty-four
song scroils utilised by members of the Midewiwin from White Earth Reservation in
Minnesota (Fulford 1989). Hoffman coilected them during the late 1880s for the Bureau of
Page 117
American Ethnology. Fulford examined them for patterns of structural variations, and
outlined how three complex symbols or shapes, which he identified as the otter, bear, and
bird, had evolved from clan markers into pictographic markers. It was impossible (ibid.: 134)
to examine precisely how the interactive process influenced the form of the pictographs
because this required comparative observations to be made by potential members of the
Midewiwin as they re-transcribed the record.
Systematic variations were observable ln redundant forms once the White Earth
pictographs were arranged into concordances (indexes). Twenty-four scroils were reproduced
in three concordances, which were published in conjunction with this paper. Fulford (ibid.)
maintained that this method was similar to that used by historicallinguists to reconstruct the
protoforms of a spoken language, because it highlighted the changes introduced by iteration.
The subtle variations existing in redundant pictographs reflected the changes introduced in
the process of recopying the images. Over time, these changes become highlighted and
modulated directly with the rate of iteration, as weil as with the formaI complexity of the
pictographs. A delicate balance existed, Fulford suggested, between those pictographs, cailed
forms, that were too simple to be recognised, and those that were too complex to be stable.
He argued that a certain degree of formaI simplification or standardization assisted
communication. Complexity, he suggested, introduced changes into the system that increased
its adaptability, and enhanced the potential flow of new information.
He used two interrelated indices to evaluate the complexity of each form and image,
and to reduce the potential of projecting ethnocentric conjectures that the complexity of the
symbols encouraged. Two terms were used, "simple" and "complex", to refer to the
symbol's frequency of distribution throughout the sample, and its capacity for
transformation (ibid.: 135). Symbols were termed "complex" by Fulford if they were clearly
distinguishable throughout the sam pIe and had an obvious capacity for transformation, while
"simple" symbols had a homogeneous distribution and were incapable of being transformed.
However, a symbol could be complex because of the frequency of its distribution and its
capacity for transformation, but simple in the wider context of its meaning to Hoffman's
informants.
Fulford (ibid.: 149) discovered that the otter, bear and bird shapes were weil
distributed throughout each of these scroils, and indicated a "high capacity for
transformation". He concluded that the transformation that he found arose as a consequence
Page 118
of the re-transcription of the song scroils. He maintained that the grain of the birch bark
could not entirely explain the syntagmatic and associative relations between the shapes,
which he cailed pictographs, of the White Earth scroils.
Fulford (ibid. 150), based upon Vygotsky, posited that the movement of the stylus
and eye aeross the birch bark scroils was connected to higher psychologie al functions, and
that the formation variations introduced through iteration had potential symbolic
significance. He maintained that Midé symbolism achieved its most powerful exposition if
the relationship between the sequential linguistic elements and the complex symbols was
examined, since the particular transformation features of these symbols could be identified as
the projection of the human form into natural objects.
The patterns revealed in the three concordances, Fulford asserted, provided an
insight into the symbolic world of the Midewiwin. He argued that the propensity of the otter,
bear, and bird symbols to fluctuate as a consequence of iteration into distinct human forms
echoed the unconscious maintenance of earlier symbolic associations. He contended that the
elemental symbols used during the protohistoric period were linked to those used by the
Ojibwa clans of the late nineteenth century through Midé rites. Images (ibid.) that were once
associated with real human faces from the Ojibwa clans during the eighteenth century
subsequently developed into the ghostly figures of the Midé birch bark scroils during the
nineteenth century. He concluded that the Midé were able to distinguish the images of their
ancestors in the symbols of the otter, bear and bird, although the Ojibwa clan system had
disintegrated.
Fulford's study demonstrated that connections existed between different shapes,
which he cailed motifs or symbols, and ethnographie details, although the context of these
meanings had changed.
Fulford (1990) later considered the verbal structure of these Midé chants to establish
how they were pictographicaily encoded. He limited his discussion to seven song scroils, six
of which shared the otter symbol. The seventh scroil was included since it shared several
other formaI characteristics with the six otter scroils. He (ibid.: 127-128) retranscribed and
re-translated the chants that Hoffman had compiled before reworking the entire corpus to
develop glosses to improve the flow, or simply supply better connections between the words
and the pictographs. It was impossible, he maintained, to provide definitive transcriptions
and translations of the Hoffman material due in part to Hoffman's informants' poetic use of
Page 119
languages. Fulford concluded that the reproductions of the scroils were generaily accurate by
comparing Hoffman's published reproductions with the originals in the US National
Anthropological Archives. He subsequendy placed the changes of the seven scroils in a
concordance by typing the revised version of Hoffman's original transcriptions of the chants
and attaching photocopies of the related pictographs to the top of the transcription. This
information he put on file cards in horizontal rows where each row corresponded with the
sequence of pictographs and changes of one song scroil.
Fulford (ibid.: 128-135) believed that the recognition of the taboo that existed on
uttering the otter's name eXplained the relationship between the pictographs of otters and
hum an arms in the song scroils. He argued that the recognition of this taboo was clarified by
chants that accompanied otter pictographs, since they ailuded to incidents in the Midewiwin
creation story but did not directly describe the pictographs themselves. These differences
caused a mismatch to occur between the visual and the verbal forms and transformed the
meanings of the pictographs and the chants. Fulford (ibid.: 135) maintained that the verbal
transformation that occurred was the result of an iterative process that occurred in traditional
Midewiwin chants and mirrored the taboos connected with enunciating the names of "animal
brothers" linked with totemic ancestors.
Both of Fulford's papers complement each other and demonstrate that, with careful
deliberation, these images can be analysed. Both papers arguably support Vastokas' (1996)
argument that native artwork could be used as "historical documents".
Similar shapes to those on birch bark scroils were examined by Fulford (1992: 190)
who considered the ninety-two page account of Vincent Roy, a fur trader of Scottish and
Ojibwa ancestry. Roy administered the American Trader's Post at Vermillion Lake between
1835 and 1839. Roy's account book, which could be securely dated to the 1830s, contained
drawings that strongly resembled those on birch bark scroils, Fulford argued. He identified
forty of the forty-eight different types of the trade goods recorded in the account book and
twelve species of fur-bearing animaIs. He determined which pictographs signified debits and
credits of each account and the identities of sorne of Vincent Roy's customers.
In being able to identify Roy's customers, and that they were not local to the post,
Fulford could indicate that people were willing to travel vast distances to ob tain goods.
Fulford (ibid.: 192-193) established that sorne of Roy's customers used remote posts such as
Roy's, which was outside Schoolcraft's jurisdiction. Schoolcraft was the nearest
Page 120
Superintendent of lndian Affairs in Michigan, and he discouraged Americans from doing
business with Ojibway who sustained ailegiance with the British crown.
Fulford (ibid.: 192-194) believed that Roy was semi-literate but his two years of
schooling while living at La Pointe had taught him numeracy, literacy, and a Calvinistic
attitude to business. These basic reading, writing, and mathematical tools enabled him to deal
with suppliers in St. Louis or New York. However, the pictographs Roy utilised to record his
inventory were familiar to his Ojibwa contemporaries including those used Midewiwin song
scroils (ibid.: 194). Fulford argued that Roy provided evidence of an assimilation process that
occurred amongst the Ojibwa over two generations during the nineteenth century, as the
proliferation of schools encouraged reading, writing, arithmetic, and the Christian religion
among the indigenous populations. He ascertained from Roy's account book that, despite
this process, Ojibwa stiil used traditional names, and considered themselves as living in
harmony with the natural world, although by the 1830s the fur trade had destroyed their
traditional economic, social, and political structures. Fulford's article indicates that the
indigenous people of the Great Lakes area understood and operated within two conceptual
frameworks and confirms that these images, although they appeared on the birch bark
scroils, were also used elsewhere as carriers of meaning and information.
A question exists regarding the relevance of this body of information, especiaily since
this materiallacks written documentation. The Aigonquian-speaking peoples, rather than the
Assiniboine, used a specifie group of images. Unfortunately, knowledge of Assiniboine art,
and of other Plains groups from before the nineteenth century is limited, given that much of
it was made of perishable materials that frequendy failed to survive in the archaeological
record. Greene (2001: 1038) observed that quillwork was the major decorative art form in
the Plains over a long period of time. Yet rock image sites also exist in the Plains region
(Keyser and I(lassen 2001), but the "biographic" style associated with this region is
connected to bison robes and the ledger book paintings used during the late eighteenth and
early nineteenth century (K.eyser 1977 and Greene 2001: 1042).
It is evident that the images, made by the Aigonquian groups and used as a means of
communication, and the techniques used to create them, have shifted over time. However, it
is clear that not ail of these images changed radicaily in appearance and experienced changes . . ln mearung.
Page 121
It appears that three groups of images exist, based upon who used them and their
audience.
The first group of images have not undergone considerable changes in their shape.
The Midé used these images to record their religious and medicinal information, and this
group has been examined by Fulford (1989, 1990), and Vastokas (1984, 1986, 1996).
The second group consists of the early geometric shapes that changed to floral
designs examined by Densmore, Coleman (1947), Penny (1991), Phillips (1984, 1991), and
Whiteford (1986, 1991, 1997). Many of the people who analyse these images invariably use
terms such as 'Thunderbirds', and 'Underwater Panthers,' rather than describe the shape of
the images.
The final group consists of those images that everyone in the Algonquian-speaking
community used. Unfortunately, little evidence exists of the images associated with this
group, but perhaps some of these were used by Roy (Fulford 1992). Nevertheless, Landes
(1968: 172) stated that images were familiar to ail Ojibwa hunters, trappers, and traveilers,
who used them to indicate boundaries, trails, notices of different sorts, and on grave
markers. Coleman (1947: 63) observed that the totem of the deceased individual was placed
on grave markers. Warren (1984 [1885]: 89-90) discussed some images inscribed on a copper
disc that he maintained recorded the number of generations of a family of the Loon totem
who had lived in the La Pointe area. The indentations indicated that eight generations had
lived in this region. Although relatively little is known it is clear that this group of images was
important to the Algonquian speaking peoples as a means of communication
Although relatively little information exists about this last group of images, they must
be considered, because this group of images indicates that everyone within the Algonquian
speaking community used images. It is therefore likely that everyone knew the entire range of
images, and understood how they were utilised and combined. An individu al could perhaps
use one or several images to express an idea, something about an individual, or leave a
message. It appears from Landes (1968) and Coleman (1947) that anyone could use, make,
and understand them. The creator of these images could be an ordinary individual, and did
not need to be a Midé, or any other specialist within the community.
Since ordinary people used images as a means of communication, it is possible that
they were known by everyone, and that an image meant more than just one thing.
Page 122
Unfortunately, little information exists as to what these images looked like, how
many existed, their range, or how they were combined, but it appears from Coleman (1947)
and Landes (1968) that they occurred individually rather than as the image groups of the
birch bark scrolls. The fact that they were used by a wide variety of individuals implies
everyone knew how to use how to use them and that specialists did not exist. Hence, these
images retained their shape much like those used by the Midé regardless of the considerable
changes in the imagery used in beadwork and weaving. However, whether any shifts in
meaning occurred remains impossible to establish.
Conclusion:
It evident that the information needed to contextualise the pictograph sites in the
Lake of the Woods derives from a range of disciplines. There is considerable agreement
about the climatic changes, and the geological history of the region. But the historians and
archaeologists find it difficult to agree who lived there, in part because each discipline uses
different types of evidence to solve their problem. Ethnohistorians (Bishop, Ray, and Smith)
argue that historical documents are useful for tracing ethnicity in the early historie period.
Many archaeologists, who pre fer to equate ethnicity with archaeological cultures, do not
share this opinion. There is sorne agreement that the Sandy Lake materials around the
Winnipeg River are possibly Assiniboine.
With the exception of Meyer and Hamilton (1994), archaeologists tend to assume
that the area was occupied by only one ethnie group at a time. While sorne archaeologists
adamantly as sert that the Ojibwa had already arrived in this region by the early historie
period, historians argue that several groups of people from different ethnie backgrounds,
Cree, Ojibwa, and Assiniboine resided here.
The inadequate archaeological data, in conjunction with the insufficient historical
references to the individual groups who lived in this region prior to contact, at contact, and
du ring the early historie period, inhibits scholars from identifying the terri tories of various
groups. The problems resulting from different names for the same groups could perhaps be
resolved by the careful use of the historical documents and the identification of synonymous
designations.
Page 123
1t also, so far, has proved impossible to ascertain which specific groups of
Algonquians, Cree or Ojibwa, lived in the region between Lake Superior and Lake Winnipeg.
However, both groups shared a similar symbolic system.
From the letter written by La V érendreye, in 1736 scholars know that Siouan
speaking peoples were in the region at the same time as the Algonquian speakers. Since the
early part of the eighteenth century, the Assiniboine were allies of the Cree against the Sioux,
although the Assiniboine are linguistically closely related to the Sioux (DeMallie and Miller
2001: 572).
The first reference to the Assiniboine in the J esuit Relations for 1657-1658 describes
them as living about 100 miles from Lake Nipigon, east of the Lake of the Woods (DeMallie
and Miller 2001: 573). Jean-Baptiste Louis's map, dated 1697, indicates that the Assiniboine
resided north and west of the Lake of the Woods while the Sioux were in the south in the
area now called Minnesota (DeMallie and Miller 2001: 572). 1ndeed, the Lake of the Woods
is referred to as Lake of the Assiniboine. This calls into question Wright's suggestions (1987:
Plate 9) that the Ojibwa, Cree, Algonquian, or Montagnais peoples inhabited the Lake of the
Woods between AD 500 and European contact.
By the early seventeenth century, the Assiniboine resided seasonally on the western
half of the Lake of the Woods, and would do so continually until the middle of the
eighteenth century (Heidenreich 1987: Plate 37). Algonquian-speaking peoples, throughout
this period lived in eastern portion of the Lake of the Woods and gradually took control of
the region as the Assiniboine moved westwards (Heidenreich and Noël 1987: Plate 40).
The question of changing lake levels was considered, since archaeologists in northern
Ontario have traditionally found sites by examining the large expanses of cliff faces along the
shores of lakes and rivers. The graduaI but dramatic increase in the water level starting at the
end of the nineteenth century caused the shape of the Lake of the Woods to change, and
possibly completely or partially submerged pictograph sites. 1t seems likely that pictographs
that are known today were once physically located higher up the cliff faces. Yet, if the water
level had fluctuated as much as seven feet prior to keeping careful records, the positioning of
images could have reflected the differing high water levels at the time they were painted.
1t is very difficult to create a chronological framework regarding the precise ethnie
identify of those who lived in this region in prehistoric times. Historians and archaeologists
have had severe problems in attempting to connect the archaeological and the historical
Page 124
record to determine which indigenous groups resided in a specifie region at specifie rimes. It
is relatively easy to use historical records to determine the ethnic identity of a group of
people from a specific geographical region during the historical period. However, prior to the
arrivaI of the Europeans in this region, such records do not exist. This is caused parrially by
continuous changes in the naming of the people who lived in this region, and the difficulties
of identifying which archaeological complex was associated with which ethnic group.
Osteological information is interesting but, like other archaeological information,
bones do not indicate the ethnic identity of the individuals. However, it cannot be excluded
from a discussion of the identities of individuals. Examination and information regarding
burial practices possibly provides information about the social, religious and cultural
identities of the individuals concerned but not about their ethnic identity. Examination of the
bioanthropological evidence (osteology, human blood group analysis or mitochrondrial
DNA) indicates that considerable advances have been made, but these sciences inform the
archaeologist about the genetic relationships, not about the cultural affiliations, of the groups
involved. Specific traits were shared by the Cree and the Ojibwa.
Many of these images have the same physical appearance despite the passing of time.
Several scholars have demonstrated that similar stylistic and design traits could be shared
with others, and may have belonged to specific groups (Duncan 1991; Lanford 1984; Phillips
1984 and 1999; Sager 1994, 1995 and 1996).
Solid evidence regarding whether the images of the pictograph sites in the Lake of
the Woods region were created by the Cree or the Ojibwa does not exist. It is easier to
conclude that these images were made by the Aigonquians rather than trying to ascertain
whether they were created by the Cree or the Ojibwa. It is possible to posit that they were
probably created by Algonquian-speaking peoples rather than Assiniboine-speaking peoples,
since the rock images in the Lake of the Woods visually strongly resemble those of sites
elsewhere in northwestern Ontario where groups of Algonquian-speaking peoples resided.
lndeed, the images of the rock images created by the Assiniboine are visually very different
from those created by the various groups of Algonquian-speaking peoples in northwestern
Ontario. A specifie group of images were used by the Algonquian-speaking peoples rather
than the Assiniboine.
The physicallocation of the sites in the Lake of the Woods also fits in to the general
Algonquian pattern of the rock image sites in northwestern Ontario. This warrants using the
Page 125
/ '
Algonquian ethnographie data ta interpret the meaning(s) of these images and, hence,
justifies using homologies for eonsidering the meaning of the Lake of the Woods images.
Bence without either the arehaeologieal or the ethnographie proof for warranting the ethnie
origins of this imagery, but based upon eulture-historieal data, 1 argue that a basis exists for
using general Algonquian cultural data to interpret the rock images 1 am studying in this
thesis.
Page 126
Chapter III: A Review of the Literature concerning the Study of Rock Image Sites in the Lake of the Woods in the Canadian Shield prior to 2001.
The collection and interpretation of data are intergenerational processes in which
each new generation amplifies, modifies, and builds upon the work of previous generations
of researchers. A critical evaluation of previous work is an essential part of planning and
executing any new research project. It is important to acknowledge, at the same time, that
most investigators work within their own paradigm. However, this acknowledgement should
not prevent their work from being critically and carefully examined.
The goal of this chapter is ta present and discuss the work undertaken prior to my
own by previous researchers concerned with the pictograph and petroglyph sites in the Lake
of the Woods area. 1 intend to establish which approaches have been the most popular, what
previous researchers have thought of pictograph sites, and how they examined that in which
they were interested. This information is important to examine sin ce it provides a standard of
comparison, a yard stick against which to compare, examine, and contrast the new data that 1
collected. Although this chapter is specifically concerned with the analysis of the pictograph
sites of the Lake of the Woods area, sometimes 1 will make reference to analogous studies of
sites elsewhere in the Canadian Shield.
The literature in this chapter is divided into seven sections. The first discusses the
earliest investigation of these images with specific reference to the Lake of the Woods.
Section two examines the technical literature that has discussed how pictograph sites have
been studied with special emphasis on Ca) subdividing the sites into sections, (b) the
vocabulary used to describe the images and Cc) the development of the techniques used to
record the images. Section three is concerned with the considerable body of literature that
presents the issues involved in the technical analysis of these sites. It is divided into two
parts: Ca) recording and interpreting the pictographs and petroglyphs and (b) the
conservation of the pictographs and the different methods used to analyse them. The last
three sections critique the literature that belongs to one of the five different approaches that
exist in archaeology: culture-historical, contextual, and intuitive, analogical, or homological.
Since these approaches are different, and a clear order exists Csee discussion in Chapter 1), it
is important for the literature belonging to each approach to be examined in sequence.
Therefore, section five considers the literature belonging to the culture-historical approach,
where archaeologists have considered the shape, location, and date of the images. Section six
Page 126
deals with the literature classified as belonging to the contextual approach, which involves
the search for patterns relating different images and combinations of images both between
and within sites. The seventh and final section reviews the literature in which archaeologists
assign meanings to images on the basis of three methodologies which are employed either
together or separately. Since three alternative approaches exist, this section contains three
parts: (i) the intuitive approach, (ii) the analogical approach, and finally (iii) the homological
approach.
Section 1. The earliest investigation of these images
Mackenzie (1793), Schoolcraft (1851-1857), Bell (1879-80), and Lawson (1885)
recorded the presence of images either painted or engraved on rocks in the Canadian Shield
prior to the twentieth century. While examining the geology of Molson's Lake to the west of
Hudson Bay in 1879, Bell (1881: 7c) found a pictograph site on the north side ofPai-Musk
taban Sipi near its mouth. The painted figures existed on a gneiss cliff that "measured up to
100 feet" (ibid.) in the form of "small figures in red ochre, said to have been painted by the
father of the present chief of the colony of lndians which removed a few a years ago from
Little Playgreen Lake to Fisher River, on the west side of Lake Winnipeg". The largest
images were "not more than one foot high" (ibid.), and most of them were not more than
eight inches. According to Bell, among them were "represented[ed] a boat, canoe, tortoise,
bird, deer, oUer, lndian, pipes, etc." He stated that the local lndians described them "as being
much more wonderful than they really are".
The geologist Lawson's (1885) publication was the first to consider the meaning of a
pictograph (DiKp-1) and a petroglyph (DiKo-2) site in the Lake of the Woods. It is difficult
to establish which rock image site was the first to be recorded in the Lake of the Woods area,
but Lawson's detailed descriptions and drawings of DiKp-1, and a nearby petroglyph site,
DiKo-2 probably hold that distinction.
• Site \inde!' diseuss! on
Figure 3.1: Map of the sites that Lawson recorded in 1885.
Page 127
,~
Lawson commented on the groupings of the images at the petroglyph site, DiKo-2 (figure
3.2) and posited that there was "little hope" of establishing any meaning or narrative because
the groups of images, or archaeological sites, were so isolated from each other (ibid.: 656).
· ~ , . ~K .. 1: HU
~ "
'>i 8 I~ 1.
Y r.- Ct, """ 4;fr~"'~
U " 'J< <1-'" · F';, 0 ... C<$ ~~
" 0 :D"t ,.":7l#'M;,S. ... " ~ • ~7
· • b .. t •
Figure 3.2: Lawson's published drawing of the images at DiKo-2, the petroglyph site.
Lawson compared the images that he described in DiKp-l and DiKo-2 to engravings
on large rocks found in the Alagoas, Brazil, described by John Brenner in a previous issue of
the American Naturalist (ibid.). Lawson also noted the similarity of the images at DiKo-2
with the painted images at the pictograph site (DiKp-l) painted on hard greenstone nearby
(figure 3.3).
Figure 3.3: Lawson's published drawing of the images at DiKp-1, the pictograph site from 1885.
Lawson concluded that the similarity between the engraved images and the painted
images suggested that they were "closely related in authorship." He was obviously interested
Page 128
in who might have created them, sin ce he concluded his paper with: "[1]he Indians of
present day have no traditions about these inscriptions beyond the supposition that they
must have been made by the "old people" of long ago" (ibid.: 657). However, if the
pictograph site in 2001 is compared with how Lawson drew it in 1885, it is clear that he did
not see or record the red images to the right and above the ones he published. The
photograph of DiKp-1 below demonstrates that what Lawson saw differed from what
existed in 2001. It is possible that the other images were painted later.
Figure 3.4: Annotated photograph of DiKp-1 taken in June 2001.
Section 2. More recent literature.
More recent examinations of rock image sites are published in a variety of places.
Sorne of these publications have summarised the discoveries of pictograph and petroglyph
sites in the Canadian Shield (Brenner 1994; Conway, T. 1979; Dewdney 1963, 1965, 1977,
1979a; Molyneaux 1977, 1980a; Nute 1948; Wellmann 1979a). A few have considered issues
such as tourism (Arsenault 1996) and political and ideological conflicts (Arsenault 1997;
Nelson and Hinshelwood 1998). Others have discussed trips that involved searching for,
finding or visiting, sometimes recording sites in the Canadian Shield, and also speculating on
the meanings of images found (Cameron and Cameron 1979; Dewdney 1958, I<nowlton
referenced rock image sites as being widespread throughout the Canadian Shield within a larger
discussion of sorne other related topie. One of this group, Dickason (1972: 13), used part of
DhKm-3 as her example of the pictographs of the Canadian Shield. Several archaeologists were
Page 129
determined to find the oidest pictographs or petroglyphs in the Canadian Shieid (pohorecky
& Jones 1968a; Steinbring 1993, 1999).
During the 1970s and the 1980s, archaeologists concerned with pictograph sites were
interested in two related issues: the selection of a place in the lands cape and the orientation
of the site. Reid (1980) observed from the information on the pictograph sites in the West
Patricia region north of the Lake of the Woods that pictograph sites tended to occur on
straight shorelines, or coastlines that were not broken into parts by bays, islands or points
and that the creators of these sites preferred rock faces that faced in a south-to-east
direction. Rajnovich (1980b: 34) developed this ide a further in her discussion of pictograph
sites in Cuttle Lake, arguing that straight shorelines were predominately selected "to increase
the spectacular effect" of the images because these shorelines enabled the images to be
visible both from nearby and from afar. She argued that, based on her examination of sites
from the Rainy River region (south of the Lake of the Woods), south-to-east facing rock
walls were selected as light conditions here were best for creating and seeing paintings. The
"rising and noontime sunsrune reflected from the water" beside the cliff face caused
"dancing light sparkles over the work heightening its dramatic value" (ibid.). She asserted
that cliff faces which faced west-to-north only experienced "shadows or the weakest sunset
light." She assumed that pictograph sites always occurred on a cliff face immediately beside a
body of water close enough to reflect light on any images painted upon the rock's surface.
Licence reports are useful for finding sites and establisrung the range of work
conducted at different sites in the Lake of the Woods (for example: Cameron 1982,
Molyneaux 1978, 1979, and 1981). Cameron (1989) described fieldwork activities in the Lake
of the Woods where she recorded and described the images at four petroglyph sites:
Kennedy Island, called DjKp-4, Sunset Channel, called DiKo-2, Tranquil Channel, called
DjKp-3, and Mud Portage, called DkKr-4. She referenced these sites only by their local
names and did not use their Borden numbers. Such practise makes trac king what sort of
archaeological fieldwork was conducted at each site and by whom difficult for subsequent
archaeologists.
Page 130
• Sites _der m!IeU!llsi61l
• Picmgra,ph shes • Petmglypb !J'tes
Figure 3.5: The four sites that Cameron recorded in 1978.
Molyneaux (1978) wrote in his report to the Ontario Heritage Foundation that his
intention was to create a photographie record of two petroglyph sites when the water level
was abnormally low. He noticed that Lawson (1885) had described the lake levels as varying
over a range of 10 feet (ibid.: 1) but that the lake's flow and therefore its level was now
artificially controlled. Molyneaux, like Cameron (1989), did not use Borden numbers, only
local names: Kennedy Island, or DjKp-4, and Sunset Channel, or DiKo-2. Molyneaux (1979)
for his subsequent study produced a detai!ed photographie record of six of the seven
pictograph sites in Whitefish Bay, in the south eastern portion of the Lake of the Woods:
DiKm-3, DiKm-4, DiKm-l, DhKm-3, DhKm-4, and DhKm-5. Molyneaux hoped to use
photography to record the images systematically, accurately, and efficiently given the
"difficult field conditions" (ibid.: 4) and also to minimise human error and interpretation.
Other reports by Molyneaux discuss sites in areas neighbouring the Lake of the
Woods (Molyneaux 1981, 1982, 1985a; 1985b). Both his 1974 and 1975 publications resulted
from projects conducted under the auspices of Trent University and the Canadian
Conservation Institute. The Trent University team headed by Molyneaux examined eleven
pictograph sites and twelve rock image sites in Northwestern Ontario. Eight of these sites
existed in the Lake of the Woods, including DiKm-3, DhKm-5, DhKm-l, DhKm-4, DhKn-
1, DiKo-l, DjKn-l, and the now extinct lichenoglyph site called DhKn-2. However, the
team considered only one Lake of the Woods site, DiKp-l on Cliff Island, in the 1975 study.
Still other studies only mentioned or briefly described pictograph and petroglyph
sites in the Lake of the Woods (Fox 1974; Reid 1976 and 1977). Reid surveyed numerous
archaeological sites, among which were fifteen pictograph sites. These sites were examined as
part of an inventory that was conducted at the request of the Regional Lands Co-ordinator,
Northwestern Region, Ministry of Natural Resources (1976: ü). The goal of this
Page 131
archaeological project was to coliect data for the Lake of the Woods General Land Use Plan
and its principal concern was the "dense st current and potential development within the
northern half of the Planning Area" (ibid.). A number of specific archaeological sites were
tested at the request of the Senior Lands Planner. A total of seventy-one archaeological sites
were recorded in May 1975. Ali the pictograph sites were aliotted Borden numbers and
assessed as to their level of "interpretability" to the public as weli as their level of
"representability" (ibid.: 13-15). What these terms meant is unclear. Reid provided the
precise geographical location of, and specified how to reach each site. He also postulated
their "cultural affiliation," essentialiy their age, and sought to determine whether the site had
any religious significance to the local indigenous population. The physical condition of each
site was evaluated, and Reid established whether the general public knew of its existence and
whether it had been discussed in print. He did not describe the images but photographed and
determined the size each site.
Fox (1974: 3) described only DhKm-1, a pictograph site calied Devil's Hole, in the
southern part of Whitefish Bay, Lake of the Woods. He observed that bundles of clothing
were stuffed into a cleft in the rock surface of this site. An informant told him of rumours
that these bundles had been "placed there by local native people attempting to affect a cure
for a sick child." He considered that perhaps the offerings of these bundles were closely
connected to the "offering tree" which Ritzenthaler (1970) had described as existing among
the southern Chippewa. Fox found similar items of clothing such as buttons, buckles, and
suspender components at this site which he presumed to be evidence of earlier offerings. He
also found lead sinkers, a corroded metal bowl, and numerous coins (1974: 3.). The presence
of these objects was recorded and they were left where they had been found. He speculated
that the clothing bundles had not been disturbed since they were left as offerings. Birch bark
scrolis were not found at this site, although fragments of birch bark with stitching punctures
were found one to two metres back in the large cleft at this site. Fox posited that these
fragments might have been part of a birch bark container used to hold food offerings and
sent these fragments to the Canadian Conservation Institute for preservation. Before Fox
finished recording this site, ali the adjacent clefts in the rock were checked and photographs
were taken (ibid.: 4).
Page 132
Figure 3.6: Fox's photograph of DhKm-1 in 1974.
Lambert (n.d.) used the same procedures as Reid had in 1975 (1976) to record eight
pictograph sites (DkKn-7, DkKn-6, DjKn-l, DhKm-5, DjKn-2, DhKm-18, DhKm-4, and
DhKm-l) in the Lake of the Woods. This report was never published. Ali the pictograph
sites he recorded in the Lake of the Woods were photographed, recorded using direct 'dot
for-dot' tracing (ibid.: 20) and described in considerable depth. These sites were discussed in
the findings of the third Northwestern Ontario Rock Art Project, conducted in 1985-1986.
This project considered some of the pictograph sites in the Lake of the Woods. The other
reports dealt with pictograph sites elsewhere in Northwestern Ontario (Lambert 1983, 1985).
Lambert (n.d. 201) argued that classification of the images from the pictographs examined in
the 1985-1986 project permitted him to distinguish between sub-areas of the Winnipeg River
drainage basin and the Bloodvein River drainage basin (ibid.: ii). The abstract of his third
survey indicates that he viewed these images not as 'art' but as a method of communication.
Lambert (n.d.: 201) argued that the largest number of animal creatures occurred in
the Bloodvein River Region, north of the Winnipeg River drainage basin. He asserted that
his results indicated that sympathetic hunting magic was an important function of the rock
images in the Bloodvein drainage basin, since more images of creatures were present in this
region. In the Winnipeg River region, images of canoes and anthropomorphs occurred more
Page 133
often. He (ibid.: 199-200) observed that there were more hand prints, and pairs of handprints
in the Winnipeg River region than in the Bloodvein River drainage basin.
For two sites in the Bloodvein River region, Lambert (ibid.: 202) posited that a
connection existed between the orientation of the site and its function. Almost ail of the
images at both sites were hand prints and both had similar orientations. He suggested that
both sites "signposted a possible avenue, by way of a portage" to another lake (ibid.: 202).
The pictograph sites he cailed the Rushing River sites, DjKn-1 and DjKn-2, in the Lake of
the Woods region acted in a similar manner, since they both used to exist on rapids or fails
prior to the raising of the Lake of the Woods water level (ibid.). He argued that pictograph
sites had "a function ta guide the traveiler" especiaily when they occurred in narrow
"situations" since those site areas were difficult to avoid. This meant for Lambert (ibid.) that
these sites could be interpreted as "message boards, maps or navigation aids." Unfortunately
he was unclear whether the message was to "guide or identify, or warn of fast water or
malevolent spirits".
He (ibid.: 204) argued that the images he examined were related to various secular
activities such as hunting, fishing, and horticulture. Specifie images he argued (ibid.: 204-
205), although unstated, were connected to hunting ungulates and sturgeon, others indicated
a life-threatening disaster connected to horticulture, and that sorne images were connected to
the Midewiwin. It is evident, upon close examination of Lambert's discussion of DhKm-1
and DhKm-4, that he has included and treated the images which comprise the site DhKm-1
as part and parcel of DhKm-4.
Page 134
Faœl-f)2A ~~ $'
t \ el ~If
VI 1'.4/sm, -/"
5ITE#!J2A "()eyih /fal" P
os' Whitd'lsh l>~ Lake "f th<r bods
Q ",.
'l.f·~J""f"'. 1 i , \) i ) ."."
1 ~~ :S"
.,.t.rf • ...r N=Jf fi2tJ(Z" Dl auf JtlI1 IS; IJ.tJ
Figure 3.7: Dewdney's 15 July 1960 field record of Site # 92A, now called DhKm-1 and DhKm-4.
This may have occurred because Dewdney, in 1960, called DhKm-4 Site # 92 and
DhKm-1 Site # 92A. It is clear from his field drawings that the images on Faces II and III of
Site # 92A are part of the pictograph site subsequently classified as DhKm-1. Face I of Site
# 92A is now called DhKm-4.
Pictograph and petroglyph sites were frequently recorded as part of larger
archaeological surveys concerned with other types of archaeological sites in the Lake of the
Woods. One notable example in the Lake of the Woods is Pastershank's (1989) report, in
which she considered pictograph sites as separate units of analysis, therefore not to be
analysed in conjunction with the archaeological sites examined at the same time. These
studies, aimed to establish that sites existed and to describe them, did not analyse them in
detail. This treatment occurred elsewhere in the Canadian Shield (Conway 1975, 1984;
Dawson 1973; C. Halverson 1988; Hill 1982, 1983; Lemaître 1995; Smith 1981). Occasionally
archaeological surveys were conducted for the sole purpose of surveying for pictographs and
petroglyphs but not for "conventional" archaeological sites that could be excavated (Lambert
1983, 1985, and n.d.; Rusak 1992a; Scott 1980; Pelshea 1980; Pelleck 1981; Rajnovich 1980b,
1981 b). The quantity of literature published in other government reports indicates that
archaeologists were concerned with these images (Conway n. d. c; Friend 1983; Jones 1966,
1981a; Petch 1991; Pohorecky and Jones 1968b; Rajnovich 1980c, 1980d, 1981c; Rusak 1991,
1992b; Steinbring and Iwacha 1982; Steinbring 1998; Tassé 1977a). These sites would have
Page 135
been recorded and discussed at sorne length in the reports required by the relevant
government bodies to inventory and analyse the archaeological sites in a given region.
Section 3. Techniques used to record the pictographs.
Subsequent methods used to record and analyse rock images are derived from
Dewdney's work. Indeed, Dewdney is rightly considered to be the "father of Canadian rock
art studies" since he was the first person to undertake a comprehensive study of these
images. The initial method for determining the size of sites appears to have emerged as
Dewdney (Dewdney and I<idd 1962) developed rus techniques for sketching and measuring
pictograph sites.
The technical lite rature is divided into three parts: Ca) subdividing the sites into
sections; (b) describing the images; and Cc) recording the images.
a) Subdividing sites into sections
It is crucial to understand the problems involved in Dewdney's use of the terms 'face'
and sometimes 'panel'. If the area covered with ochre is large, sites are invariably subdivided
into smaller components to aid recording, photography, and description. The initial method
for determining the size of a site emerged as Dewdney developed rus techniques for
sketcrung and measuring pictograph sites.
5itfE #10' 5uns"r ChclrJlleILake of thRWoodr Ailg./Z'S9
(perroglyplzs of 5" .. 1171 sbawn ds~wh.-n:)
rwa>'/!f
Figure 3.8: A field recording by Dewdney of two pictograph sites in the Lake of the Woods, Site # 70 Sunset Channel (DiKp-l) and Site # 69 Picture Rock Island (DhKm-3).
Page 136
Dewdney's sketches of the rock features of the different parts of a new site probably
enabled him to identify the physical location of various groups of images (ibid.). Yet
Dewdney did not identify different parts of a 'face' as 'panels' if they had different
orientations to each other. It is possible that this practise developed as a result of other
fieldworkers beginning to take compass bearings of rock faces. It was observed that the
surface of the rock used for paintings faces the rising or the setting sun or in a southerly
direction (ibid.: 6). Hence the pictograph sites that Dewdney examined faced towards the
east, the west, and the south. Indeed, Dewdney (ibid.) stated that he had "seen only three
sites on wruch the sun never shines" where lichen grew, but he did not state in wruch
direction the surface of the rock was faced.
Difficulties exist with the meaning of the terms 'face' and 'panel' since researchers
often do not clarify their meanings. Unlike Pastershank (1989), who used these terms but did
not define them, Rusak (1992: 1) defined a 'Face' as a single figure or group of figures on a
common rock plane. She argued that the "F" must have its first letter capitalised to
distinguish it from its common meaning. It appears that she used the concept of panels to
subdivide large sites, especially if two areas containing pictographs were physically separated
on the same face. These panels were numbered sequentially if the panels had different
orientations.
It seems likely that the concepts of faces and panels were used to divide the rock face
into surfaces that could be discussed and analysed easily. In rus unpublished report (n.d.: 21),
Lambert developed the ide a of splitting sites even further into manageable parts, if cliff faces
with paintings were "sufficiently large to warrant supplementing the standard Face and panel
designations with seriaI, horizontal data points." Lambert maintained that seriaI horizontal
points were useful if separations appeared on the rock face. He (ibid.) applied it at Bloodvein
1 Pictograph, EiKs-l, in the Bloodvein drainage basin, but maintained that trus technique
would have been useful in the 1982 and 1984 Northwestern Ontario Rock Art Project
(Lambert 1983, 1985) studies. However, he acknowledged that he had not discovered many
sites where there were "continuous paintings for long, relatively flat, clifflengths" (ibid.: 23).
While archaeologists use this technique to subdivide pictograph sites for the purpose
of making them easier to describe and quantify, it can cause problems for subsequent
researchers, since precise rules do not exist concerning how trus technique should be
employed. Twelve pictograph sites were divided into parts when they were first recorded.
Page 137
This group includes DiKm-3, DhKm-3, DhKm-1, DhKm-4, DiKn-1, DhKn-1, and DjKll-1.
Each site will be examined in sequence. It is clear that both Dewdney and Pastershank
invariably subdivided sites into parts if they were the first individuals to examine a site. It is
therefore easy to subdivide this group into two smaller groups according to whether
Dewdney or Pastershank recorded the site. Dewdney examined DiKm-3, DhKm-3, DhKm-
1, DhKm-4, DiKll-1, DhKll-1, and DjKll-1, while Pastershank examined DgKl-19 and
DgKl-17 first. However, Pastershank contradicted Dewdney's divisions of DgKl-1 and
DgKl-2 when she examined them in 1989. DhKm-5 will also be examined because, rather
than being subdivided, it grew in size.
DiKm-3:
Dewdney used the crack visible in the middle of the field drawing (figure 3.9) and
photograph (figure 3.10) below to differentiate between the two parts, or Faces, of the site
thfT rv.sl.dlJt2 of Q 'Pf'4!:P" t'rlH-"Fou"d hli!t""i~ now il! f1I", ROMdfrnogroph{c:coIJochon
faC', l
"
ettiUSTOPHIR DIWON..,
Figure 3.9: Dewdney's field drawing of Site # 28, undertaken on 15 August 1953.
Dewdney's first Face is the left side of the rock tripe growing vertically across the
pictographs, which is evident on the extreme left hand side of the photograph below.
Page 138
Figure 3.10: Face 2 and 3 of Site # 28 by Dewdney in 1953.
Unfortunately, it is unclear whether Reid, who examined the site in 1975, and Molyneaux,
who photographed and examined it in 1979, used Dewdney's subdivisions in their analysis.
DhKn-l:
Dewdney subdivided this site into Face l and II probably using the large vertical
crevice that exits in the cliff face that separates the two clusters of paintings in figure 3.11.
r", fU1'~" '---
IJ'q~
18"'''
a6<W12'waftzr Auy,Ij.l5'i/{J .
!
Sire #/OG 06abikonNofTi:JW5, /VOFtt7@a'of ObobikonlJO'/{. S'outil of AU!I71?atl Pel7il7su(<1, Lake oft~ W'oodr
Figure 3.11: Dewdney's field drawing of Site # 106 undertaken on August 19 1960.
l t is not clear whether Reid in 1975 and Molyneaux who examined it in 1974 and
Page 139
1981 followed Dewdney's method of dividing this site into two parts.
DhKm-l and DhKm-4:
Dewdney's field drawing of Site #92 A and Site # 92 indicates that the two sites
DhKm-1 and DhKm-4 were considered one site (see figure 3.7). Dewdney found and
recorded both in 1960, classifying them as one site. He recorded only half of the encire site,
possibly because he did not think that smears of ochre were paincings. However, why
Dewdney only recorded the images from part of DhKm-1, which he called Site # 92,
remains unclear. Dewdney's Face l is DhKm-4, while Faces II and III are part of DhKm-1.
DhKm-4 is clearly not immediately beside DhKm-1 as Dewdney's field drawings (figure 3.7)
of both sites imply. However, why Dewdney only recorded the images from part of DhKm-
1, which he called Site # 92, is unclear. The sites are more than a few metres apart from each
other as the annotated photograph in figure 3.12 below indicates.
Figure 3.12: The Close Proximity of DhKm-4 to DhKm-l.
Dewdney missed an image now classified as a smear as indicated by the annotated
photograph and Dewdney's photograph of the site below.
Page 140
Figure 3.13: Annotated photograph of DhKm-1 in 2001 by Colson according to Dewdney's divisions.
Lambert, in his (1985) unpublished study of sorne of the pictograph sites, made the
mistake of considering sorne of the paintings of DhKm-1 as part of DhKm-4, despite Reid
(1976) in 1975 reporting that the two sites were separate entities with their own Borden
numbers. Why this confusion existed for Lambert is difficult to determine, since he would
have known that Reid had already designated DhKm-1 as a separate entity from DhKm-4.
His own records show that Lambert (n.d.) knew of Dewdney's 1973 publication and had
probably seen Dewdney's 1960 field drawings of DhKm-4 and DhKm-1. The images of
Lambert's Face II of DhKm-4 are images of DhKm-1 that he recorded, if one examines the
annotated photograph below in figure 3.15. Lambert recorded the images that he deemed
were the site. He applied a technique of classification developed during his field survey of
sorne pictograph sites in the Upper Severn region of Northwestern Ontario in 1982 (1983).
Page 141
Figure 3.14: Annotated photograph by Colson of DhKm-1 taken in 2001 according to Lambert's divisions.
DhKm-3:
Dewdney was the first person to record this site. He subdivided it into three sections
as seen below in rus field drawing (figure 3.16).
#69 P,cTure Rgd; /,/<7nd WI)!fefish Boy,Lake of the Woods Aug855'
16,50 Fow$
CHili STOPHtR otWONEY
Figure 3.15: Dewdney's field drawing of Site # 69 undertaken on 8 August 1959.
It is unclear whether Molyneaux in 1974 and 1979, Reid in 1975, or Pelshea in 1978
paid any attention to Dewdney's subdivisions of the paintings.
Page 142
DiKn-1:
According to the two field drawings below (figures 3.16 and 3.17), Dewdney divided
this site, initially called Site # 91, into five parts. Several of these parts were subdivided again .
• rth<:ilsttip of Ha'{A:r Penil/sale; '/letfn e/ltranœ toWlJtlefishfJar,ta/œoftil<?/Iiâ"ds :j
lB·
\ 'tl ;i' [~:~~I-::~ lB'
,7:1..170
Face 1
[ '\,
Facclb Fac,z Il/:'
CH~U STOPHER DEWONtV
Figure 3.16: Part 1 of Dewdney's field drawing of Site # 91A and B, undertaken on 15 July 1960.
wntinued)
/t(!'1'1IttÇ ,;'own"
OruVII wc/ur kw/
or; cr /u/,/ 15; INO
{ace V
'"
,,' ./,,,"'" "",du·
50' _ ~ __
Ct11\1 STOPHi'R DEWDNEY
Figure 3.17: Part 2 of Dewdney's field drawing of Site # 91A and B, undertaken on 15 July 1960.
Page 143
DjKn-1:
Dewdney recorded this site as Site # 29 in 1958. From examining two photographs
of trus site, rus field notes in New World Arcruves at the Royal Ontario Museum, and
Dewdneyand Kidd's publication (Dewdney and Kidd 1962: 27 and 1967: 29) it seems that it
contained two components.
Figure 3.18: Setting with offerings and C.D., Site # 29 by Dewdney.
The second part of the site was published in Dewdney and Kidd's publication
(Dewdney and Kidd 1962: 27 and 1967: 29).
Figure 3.19: Photograph of part of DjKn-1 called Site # 29 taken by Dewdney in 1959.
Unfortunately, it remains unclear whether Molyneaux in 1974, Reid in 1975, or
Pelshea in 1978 used Dewdney's method of dividing the site into two parts. However,
Page 144
Lambert (n.d.) considered the site as two separate parts, recording two parts of this site as
Face l and Face II. Rusak in 1993 expanded the site to include a new section which she
called Face III. This was the location of a hand print that existed higher up the same cliff as
part of the same site but not near the water's edge. Unfortunately, l was unable to ob tain
coptes of Rusak's photographs, so l have provided my photograph of this image (figure
3.20).
Figure 3.20: Part of DjKn-1, July 2001.
Dewdney recorded and examined the following two sites ln August 1964.
Pastershank challenged his subdivisions in 1989. Both sites are close to each other, but differ
in terms of size, and in what has occurred since Dewdney first, then Pastershank, examined
them.
DgKl-1:
The water below this site is shallow, and the site is partially protected from the main
channel by two of the three small, neighbouring islands. It lies on a south facing spit of land
on the main travel channel, between red marker S56 and S60 at the northern end of
Sabaskong Bay. Dewdney identified two groups of images which he called Face l and II of
Site # 197 (figure 3.21).
Page 145
/lit orm d' 5a&/$konq i3aq, Cqk~ orthe Wood~
Sd!(~ #('/7 uod#198
"<" ... :, '!.i.... r'"
-. ~1 q~~~~~~~~~.'~~'~_~'~_~~" __ ' __ ~_~A_
'\ CMfUSTOPHIft lIlWf)NW
Figure 3.21: Dewdney's 1964 field drawing of Site # 197 (DgK1-1) and # 198 (DgK1-2).
Pastershank in 1989 (1989: 51) contradicted his method of subdividing the images. She
found images on a surface to the left of his Face 1 and, hence, renamed his Face 1 as her Face
II and his Face II as her Face III. The 'new' images to the left of her Face II were cal1ed Face
1. The site was clearly subdivided as indicated in the photograph in figure 3.22, into three
parts, cal1ed Faces, based upon the physical structure of the cliff face.
Figure 3.22: Photograph of DgK1-1 in July 2001 with Pastershank's 1989 subdivisions.
Page 146
DgKl-2:
Pastershank disagreed with Dewdney's subdivision of DgKl-2 when she recorded it.
Her divisions, unlike Dewdney's of 1964, did not depend on the identification of specifie
images, which she termed motifs, of DgKl-2.
Figure 3.23: DgKl-2 in 1989 hy Pastershank.
DgKl-2 is the larger of the two sites that Dewdney examined. The decisions
Pastershank made in subdividing it in 1989, when she recorded the site again with her crew,
are harder to understand and replicate than with DgKl-1. Dewdney recorded the part of
DgKl-2 called Face 1 by Pastershank in 1964, but since then the number of paintings present
at this site had increased in number. The image in figure 3.24 is the image that Dewdney
recorded at DgI<J-1 although it was published 90 degrees counter clockwise in Dewdney's
1962 publication with Kidd.
Page 147
Sabaskong Bay, Site #198
Figure 3.24: Dewdney's published images from this site.
l also discovered four additional and obviously new images at DgKl-2 in 2001 (figure
3.25). The presence of these images reflects the issue that Pastershank's subdivisions of the
site may have affected her description of the site. Therefore, it was important to replicate
Pastershank's subdivisions of DgKl-2.
Figure 3.25: Three of the four new images at DgK1-2 in July 2001.
Unfortunately, almost ail the images Pastershank identified, except for two
recognisable handprints and sorne parallel lines, were very pale and next to impossible to
distinguish. The disappearance or fading of the images recorded by Pastershank in 1989
could be a consequence of the exfoliation of the images, possibly in conjunction with the
deposition of mineraIs. Mineral deposition often causes a white deposit to form on images
painted using ochre.
Page 148
The new paintings found in 2001, in figure 3.25 above, were produced using modern
paint instead of ochre. The new paintings including an type of creature, a six pointed star and
a vertical crescent with its points facing right, were both superimposed on older paintings
and painted directly on the rock surface beside paintings Pastershank (1989:54-63) had
described, recorded, and photographed. The fourth new image was created not on any of the
designated "Faces" but was placed beside the area called a "site" by Pastershank. This image
is radically different from any of the new images since it is hollow rather than solid. Large
quantities of offerings also were discovered in the 2001 field season. .
Figure 3.26: The new image at DgKl-2 not painted upon the older ochre images in July 2001.
Pastershank (1989) designated Face I as being at right angles to the rest of the site on
the right hand side of figure 3.27. The large overhang was divided into Faces V, VI, and VII.
The rock underneath the overhang was divided into Faces II, III, and IV.
Page 149
Figure 3.27: Annotation by Coison in 2004 of DgKl-2 taken by Pastershank in 1989.
Face 1, first recorded by Dewdney in 1964, is relatively easy to relocate from its clear
images and Pastershank's published field recording in figure 3.28 below.
Figure 3.28: Field recording of detail of Face l, DgKl-2, by Pastershank in 1989.
Page 150
Figure 3.29: Photograph of Face 1 by Pastershank in 1989.
However, the remaining Faces are not allocated to the surface of the rock in a way
that can be identified easily, since Pastershank described their location on the rock surface in
relation to specifie morphs that she identified. Unfortunately, by 2001, these morphs were no
longer recognisable. Consequendy, it was difficult to determine the precise location of each
Face. Her published report designated Face II as the area of rock underneath the overhang
where she observed "an overpowering red ochre smear" (ibid.: 55). The image below is her
field recording of Face II (ibid.: 57). Pastershank argued that, when she recorded this "Face",
the clearest morph was a "zoomorphic bird (Morph 4)" found at the bottom corner on the
north side of the panel (ibid.).
Page 151
~ ...
Figure 3.30: Annotation by Colson in 2004 ofphotograph of Face II by Pastershank in 1989.
Figure 3.31: Field drawing with detail of Morph 4 at DgK1-2 by Pastershank in 1989.
Unfortunately, by 2001 this morph, which she called a "thunderbird," was not as
bright or distinct as Pastershank had described it. Consequendy, it was difficult to establish
its precise location. Pastershank argued that the base of this bird image, was "in sorne danger,
as it is sitting on the top of the middle of the roof-laminating caverns" shown in figure 3.31.
Page 152
The bird image she is concerned with is the arca of concentrated dots in figure 3.31. She
asserted that distinct handprints, called Morphs 5, 6, 8, and 10, were also on Face II. Three
"finger strokes" were recorded as existing in the central area of the red ochre smear cailed
Morph 7 (ibid.: 59). Pastershank found fresh tobacco left as an offering, in a smail crack
fifty-three centimetres below Morph 10. She stated that trus part of the site suffered from the
largest quantity of rock spalling. There was crumbling over the whole site and smail colonies
of lichen were growing there. She argued that mineraI seepage was evident both above and at
the sides of the red smear and throughout the he ad of the bird morph (Morph 4).
10cm
+ 1m22ahw 2m30aw.
Figure 3.32: Field recording of detail of Face III by Pastershank in 1989.
Pastershank maintained that Face III was located twenty-eight centimetres left of
Morph 10, and forty-seven centimetres below it (see figure 3.32). The only recognisable
morph on trus Face was a "possible hand print" called Morph 11 (ibid.: 59). However, by
2001 it was difficult to identify the morphs that Pastershank had used in 1989 to subdivide
the site. Pastershank (ibid.) located Face IV farther south, i.e. below ail of the Faces that she
had already designated as protected by the roof of the shelter from any "weathering agents."
Page 153
10cm
Figure 3.33: Field recording of Face IV of DgK1-2 by Pastershank in 1989.
Several morphs were identified, including the outline of an image she described as an
"upside-down man with his elbow bent on the right as if to place one's hand on this rup"
(ibid.) (figures 3.33 and 3.34). She maintained that the depiction of trus image was indicative
of "its importance and possibly predicts the death of some-one prestigious" but,
unfortunately, did not provide any evidence to substantiate her conclusion. Pastershank
identified another morph, a smear that apparendy measured "the size of a handprint" to the
left of the upside down stick figure. She identified nine handprints as existing at trus site and
maintained (ibid.: 63), as Lambert (1985:123-125) had done, that the hand was "an indicator
of death of a warrior." She argued that farther left of trus smear was a similar sized but
"indistinguishable" smear of red ochre. Pastershank photographed the same part of the site
where Morph 12 was found. However, by 2001 this morph was no longer distinguishable.
Page 154
Figure 3.34: DgKl-2, Detail of Face IV by Pastershank in 1989.
Faces V, VI, VII were stated to exist on the overhang created by the rock shelter.
These faces were contiguous and the morphs were adjacent to each other on the curved
overhang. According to Pastershank, Faces V and VI had one hand print each, called
Morphs 14 and 15 respectively, while Face VII had a horizontal undulating red line
measuring twenty-seven centimetres, called Morph 16 (ibid.) (figure 3.35).
Figure 3.35: Pastershank's Field recordings of Faces V, VI, and VII in 1989.
Pastershank observed that ail three faces suffered from mineraI deposition which she
argued was indicative of heavy water leaching that was caused by precipitation and lichen
growing above the red images.
It was difficult to utilise the arbitrary framework that Pastershank used to divide
DgI<1-2 into manageable units. Pastershank's subdivisions superimposed on the photographs
in figure 3.36 do not foilow the physical structure of the rock but depend on the
identification of specific morphs. Furthermore, problems exist for subsequent archaeologists
Page 155
Slnce the images fade and decay. The deterioration of the morphs underneath the rock
shelter made it impossible to identify the precise boundaries between Faces II, III, and IV.
Figure 3.36: Using Pastershank's (1989) subdivisions of rock surface to divide the paintings in 2001 into Faces V, VI, and VII.
Some of the difficulties in establishing the precise location of Pastershank's morphs
may be the results of the techniques by which they were reproduced for the final
archaeological report. The published field recordings are the final product of the whole
Page 156
~ .. 1
process to which the field recordings were subjected, once they had been taken to the
archaeological laboratory. The field copies of the sites were subsequendy traced onto acid
free albene paper in the laboratory. The field copies were reduced considerably in size using a
process called photo-mechanical transfer, so that they could be readily published using 8.5 x
11 inch paper. Therefore, it is possible that tiny details, which enabled Pastershank to readily
identify morphs, became difficult to determine.
Pastershank interpreted DgKl-2 as follows:
The environmental setting and the extensive ochre red paintings have made this site overpowering. Death may be the subject of Face IV as seen in the form of an upside down man, positioned over the third cavern, perhaps the death of an important person. A total of nine handprints were recorded at this pictograph site. The interpretation of the hand being an indicator of death may be valid (Lambert 1985: 123-125). To make this site even more spectacular is the placement of the powerful spiritual figure, the Thunderbird, over the middle cavern.
Pastershank (1989: 63)
Pastershank clearly considered the location of the images in this cavern-like setting as highly
significant. Her naming this site "Three Caverns Pictograph Site" (ibid.: 54) seems to reflect
this fa ct. Yet, only two caverns existed at this site since the second "entrance" is in reality a
hole in the wall of the larger of the two caverns. Pastershank (ibid.: 70) identified this site,
and the three other sites in Sabaskong Bay, as "tentatively called 'Boreal Forest' Aigonkian."
She argued that they were dated circa 1600, which she maintained was "characteristic" of the
majority of the rock image sites in the Lake of the Woods. This is an interesting conclusion
but she does not provide any evidence to support it.
DgKl-17:
Pastershank recorded the site in 1989 and subdivided it into three parts designated
Faces l, II, and III. Face l did not exist nearby Face II and III, which are beside each other.
Pastershank (1989: 63) described Faces II and III as "approximately seven metres away from
the eastern start of the bedrock cliff and Face l is an additional 5.4 metres away just from a
southward bend." She stated that the images at this site occurred in two "components"
(ibid.: 67). The first group, or component, if her word is used, are in the photograph below
in figure 3.37.
Page 157
Figure 3.37: Face 1 of DgK1-17 by Pastershank in 1989.
The images in this part of the site are quite faint, although she described them in
sorne detail. She asserted that this part of the site had "paired horned anthropomorphs, and
one hand print" (ibid.). Pastershank divided the images of her second component at this site
into two parts which she called Face II and Face III, although they are immediately beside
each other. Unfortunately, a photograph does not exist of this part of the site, nevertheless
her field recording in figure 3.38 provides sorne indication of the images present in
Pastershank's Face II and Face III.
Page 158
Figure 3.38: DgK1-17 - Pastershank's Face II and II in 1989.
No-one recorded this site between 1989 and 2001.
DgK1-19:
Pastershank (ibid.: 67) described this site as "only 18 metres west of DgKI-17" but,
because the images had "different styles, elevations, and hues of pigmentation," this site was
"not contemporaneous" with DgKl-17. She (ibid.) asserted that this site could be divided
into two smal1er parts since sorne of the images existed on a "perpendicular, south-facing
hiatus ofbedrock," as shown in the field recording in figure 3.39.
!il!' (,
Figure 3.39: Pastershank divided this Face into two parts called Panel i and ii in 1989.
Page 159
. ,r-
DhKm-5:
It appears that this site was considered as a single entity and not subdivided into
several parts by any of the archaeologists who recorded it including Dewdney in 1960, Reid
in 1975, Molyneaux in 1975 and 1979, Pelshea in 1978, and Lambert in 1985.
Figure 3.40: DhKm-5 in June 2001.
However, when l visited this site in the 1991 with Rusak, an archaeologist who at the
time was conducting fieldwork in the White Otter Lake/Turtle River system (1992a), she
pointed out that DhKm-5, in figure 3.40, consisted of not only the rock face recorded by
Dewdney and his subsequent colleagues, but an additional part in figure 3.41 that existed to
the left of the rock surface that was already termed DhKm-5 .
Page 160
Figure 3.41: The new part of DhKm-5.
Clearly, if a site could be subdivided into smailer units, faces and panels it could also
be increased.
b) Terminology
Pastershank's (1989) identification of specific morphs at DgI(J-2 inadvertendy
caused her descriptions to become blurred with low-Ievel interpretations. Identifying the
images was a crucial part of Pastershank's description of the sites and her subdivisions
enabled her description of each site to have a structure and become manageable. The
boundaries for each Face were dependant on her ability to recognise specific images that she
labeiled as morphs. Yet, by 2001 many of the images that she had described in 1989 were
impossible to identify.
The vocabulary used to describe the images of pictograph sites is important, since it
enables the archaeologist to classify the images, and to create an inventory of ail the images
at each site. The vocabulary used to describe each image influences how it is examined, both
as a separate entity and within a group. The words used to describe an image can and do
influence how the shape and ultimately the style is considered. Archaeologists are interested
in style because different modes of depiction are often equated with different ethnic groups,
while similarities are interpreted as evidence of those interactions.
Page 161
Pastershank's (1989) classification of all of the readily identifiable morphs on each
Face enabled her ta group them by type, and by whether they were hollow or not. In doing
this, she followed the technique of analysing these ochre images advocated by Dewdney
(Dewdney and Kidd 1962, 1967), Vastokas and Vastokas (1973), Lambert (1983, 1985 and
n.d.) and Rajnovich (1980b, 1980c, 1980d, 1981a, 1989, and 1994). Pastershank (1989: 70),
who was trained by Rajnovich, determined in her archaeological survey of four pictograph
sites in Sabaskong Bay that 76 identifiable morphs were found on 15 Faces. DgKI-1 had 26
of ail of the identifiable morphs found at these four sites, while DgKI-2 had 17, DgKI-17 had
30, and DgKI-19 had 3 morphs. Unfortunately, she did not describe the shapes of these
images. The names she gave to different morphs were also problematic.
Pastershank's verbal descriptions of each type of morph cause problems because they
imply meaning to the reader according to their physical shape. As Dewdney (1979b) had
noted earlier, this problem was widespread amongst archaeologists and others interested in
these images. He observed that numerous difficulties ensued when archaeologists described
visual images verbaily and admitted that his own work suffered from these flaws and
problems (1979b: 326). In one case, he had argued that at over half the sites he had recorded
images that bore:
no recognizable likeness to any known form and l designate them as abstractions. Many of them are single strokes occurring in groups or series that suggest tally marks. The remainder range from simple to relatively complex forms.
The other half of the symbols subdivided roughly into five groups: miscellaneous man-made objects, hand-prints, other human subject matter, animaIs, and composite-presumably mythological-creatures.
Dewdney and Kidd (1967: 18)
Dewdney (1979b) realised that verbal descriptions of images were prone to subjective
variation (ibid.: 326-328). He (ibid.) subsequently suggested that a series of graphic and visual
models could provide the level of rigour needed to classify images because words were
insufficient. Dewdney argued, drawing upon a number of examples from the literature on
rock image sites from both the Canadian Shield and elsewhere in North America, that he and
his colleagues had been careless "in assigning captions to our illustrations" (ibid.: 330) in the
publications of their research on rock image sites.
He (ibid.: 333-336) also questioned the manner ln which his coileagues had
reproduced rock image sites. He harshly criticised the practice of examining these images one
at a time, often extracting a single image and connecting it with similar images from another
Page 162
~ ..
group. He insisted that images which formed a group should be examined in conjunction
with each other as weil as in relation to their original location, and that a site must be studied
in relation to its local environment. Not doing so was the equivalent of treating sites and
images at "the level of an artifact picked up in a ploughed field."
Unfortunately, Dewdney's (1979b) hopes were not realised quickly. Sorne
archaeologists conducting field surveys in the Boreal Forest have continued to define the
images at rock image sites over-narrowly. For example, Haywood (1983) wrote that thirty
figure," and "fish." Both the perception and value judgements of the archaeologist have
affected the description of these images. Dewdney (1979b: 336-337) concluded that those
concerned with rock image sites should develop new recording standards independent from
the conventional methods of attributing meaning to images.
(c) The development of the techniques used to record the images
The development of how the images were recorded commenced with Dewdney in
the 1950s (Dewdney and Kidd 1962). After he had sketched the images, Dewdney measured
the distances between the physical places on the rock surface where the images existed and
designated them as individual parts using Roman numerals (1962: 8-9). He took colour
photographs, and recorded the compass bearing of the site, the depth of the water nearby,
and the height of the cliff (ibid.: 8-9). At first, a three-inch grid was lightly chalked on to the
rock to ensure an accurate scale copy of the images at each site (ibid.). He washed the chalk
off afterward. This grid enabled Dewdney to determine the location of a key point on a rock
face. The distances to key points in a painting were measured from string co-ordinates across
the rock face. The use of thin ]apanese rice paper employed for block printing, which
became transparent when it was wet, soon superseded this technique, as visible in the
photograph below.
Page 163
Figure 3.42: Dewdney working with a colleague on an improvised platform at Site #108.
Sometimes Dewdney built physical structures beside the site to enable him to record
them. The photograph in figure 3.42 demonstrates that Dewdney used a log probably
attached by rope to a tree growing above the cliff face at DhKo-1. Two rectangular shaped
grey areas exist beside and in the middle of the images. The one in the middle of the
paintings immediately in front of the standing individual is probably rice paper. The nature
of the second rectangular shape is unknown and appears perhaps to have been applied to
coyer something up.
Dewdney used a sponge roller to wet the paper, which also caused it to adhere to the
rock surface. The images below and visible through the paper were subsequently recorded on
the rice paper using Conté chalk allowing him to record all but the faintest of images.
Dewdney also made notes directly on the paper regarding cracks and the height of the
images above the water level. This development permitted Dewdney to reduce the time
taken in recording these images. However, as Dewdney noticed, the rice paper both
obscured the recorder's view of the fainter images and lighter details of the stronger coloured
motifs, especially if the stronger images were superimposed on larger, fainter images.
Dewdney's site numbering sequence reflected the order in which he recorded the site rather
than anything else. He argued that, as he recorded, his accuracy increased "due to practice"
(ibid.: 9).
Pelshea (1979: 24) expanded the range of colours used so that red recorded ochre
and green was used for lichen, frost spalls, and rock level changes. He also started to replace
Page 164
or substitute the Conté chalk with Dixon Lumber crayons since they did not cause the wet
rice paper to tear.
Jones (1968: 110) believed that the images should be outlined using felt pens, rather
than duplicated using a method called "the solid-line technique." The latter resulted in the
"character and the distinctiveness" of rock images in different geographical regions being
omitted. According to Jones (ibid.: 110), solid-line recording was the most common and
widespread technique for recording pictographs. It was no t, however, a good method for
"conveying actual appearance of painted symbols, and, as importantly, the char acter (his italics]
and distinctiveness of the rock art of a particular area." Jones was concerned that key aspects
of the images of different regions were being ignored and simplified, since that made "the
symbols more subject to gratuitous classification" and "interpretation." He subsequently
advocated that the best alternative was to record every aspect of these "symbols and their
settings that set them off from other styles" (ibid.: 111). Evidently, according to Jones a
recording technique had to render any distinctions in the style of the images evident in the
field recordings.
According to Jones (ibid.), Dewdney initiated the method of recording whereby the
edges of images were outlined. Jones either crosshatched or shaded the interiors of the
images to indicate gaps in the pigments of faded areas. Pictographs were recorded on
transparent plastic using different coloured pens to differentiate between the pictographs and
lichen encroachment and damage to the rock surface. Jones advocated that notes should be
made on the plastic where spalling, differential fading of the images, and unusual features
existed. Later in the laboratory, a stencil copy was made of this field recording using a light
celluloid paper and inking equipment. The resulting image was checked against photographs
so that the colours could be verified, ensuring a good reproduction of the original image. The
technique of Jones' recording as many precise details of these images as possible likely led to
the development of the dot-for-dot technique.
Questions regarding image reproduction had already been raised in the minds of
several archaeologists. Since Dewdney was aware that rice paper prevented faint paintings
from being discerned, he quickly adopted Jones' (1968) and Pohorecky's (1968) refinement
of the technique, using Saran Wrap (thin clear clinging plastic usually used for wrapping le ft
over food) that was electrified with a brush, and grease or felt pens for tracing the images. It
is possible that Pohorecky (1968) advocated Saran Wrap because of its tendencies to stick or
Page 165
"ding" to surfaces. This new recording technique did not obscure the images, enabling
recorders to salvage faint images invisible under the wet rice paper. The water required to
adhere the rice paper to the rock surface possibly also reduced future visibility of the
pictographs, since it left mineraI deposits on the surface of the site. However, Steinbring and
Elias (1968a) questioned Dewdney's rapid adoption of this technique. They argued that,
sometimes, it was difficult to record a site using Saran Wrap, and that, in these instances,
field researchers should record the images using metre grids. The representational advantage
of the dot-for-dot recording technique is evident, given that a number of researchers on
pictograph sites in Northwestern Ontario (see C. Halverson 1988; Lambert 1983, 1985, n. d;
Rusak 1992a), including Pastershank (1989) used it for several decades. EssentiaIly each dot
of ochre visible on the rock surface was recorded by a dot, using a red felt tip pen, upon
clear plastic that was held by sticky tape to the rock face. Lambert (n.d.: 20) asserted that this
technique enabled the "duplication of the cultural, as weIl as the equaIly important non
cultural information." He argued that recording both the ochre and the colour-coded "linear
and areal symbolics" conveyed "contextual associations for both types of information." This
type of record, he asserted, enabled archaeologists to provide more detailed tmage
descriptions and analyses of specific paintings. Lambert (ibid.) argued that the plastic
recording was a good idea and, since it used a 1:1 ratio, rendered "an almost exact, complete
reproduction." He maintained that this record was important for preservation purposes
because the pictographs that he examined suffered from erosion.
However, changes in pigment could not be replicated easily using the dot-for-dot
technique. Superimposed images were recorded with dots positioned close to each other if
they were dark, or, if the ochre was faint, relatively far apart. The relative distancing and
spacing of the dots appears to have been at the discretion of the recorder. Rusak advocated
that the 2 millimetre thick plastic be cut to size, and fastened to the rock face with tape.
Whatever was visible below Rusak reproduced immediately above it on the plastic using
waterproof felt-tip pens (Rusak 1992a). Rusak, as weIl as Lambert, did not use Saran Wrap
but did not inform the reader why it was not used. Red denoted ochre and blue, minerai
precipitation and patination. Any organic matter, such as lichen, rock tripe and moss
encroachment, was recorded using a green felt tip pen, while a black felt tip pen recorded
cracks and fissures on the rock surface (ibid) (figure 3.43). Pohorecky and Jones (1966: 104)
coded their felt tip pens according to colours on the MunseIl soil colour chart.
Page 166
Figure 3.43: The plastic being taped to the surface of the cliff face and the mineraI precipitation and patination being recorded by a black pen by Rusak in 1991.
Pastershank did not darify which colours were used to denote which features for recording
the pictograph sites in Sabaskong Bay, Lake of the Woods, but it appears from her 1989
report that she used the dot-for-dot recording technique. It is probable that Pastershank
(1989) used the same colours as Rusak (1992a), because Pastershank trained Rusak as a crew
member for the archaeological survey during which the four pictographs sites in Sabaskong
Bay were recorded. Field recordings, created using this technique, were once again traced
onto acid free albene paper in the laboratory and reduced in size considerably so that they
could readily be printed on 8.5 x 11 inch paper by a process called photo-mechanical transfer
(Rusak 1992a: 23).
This process of physically recording these images is problematic despite Pohorecky
and Jones's (1966) daim that this technique required "no artistic skill" and allowed relatively
few opportunities for the recorder to make mistakes. Both the field recording process and
the creation of a duplicate in a laboratory are laborious. Mistakes can accur if the recorder or
a copier is a novice, since the image produced could reflect the impression of the images in
their minds rather than being an accurate reproduction of the images. The recording of faint
and superimposed images is an especially difficult and possibly subjective activity.
Superimposition of images occurs when images are painted one top of another at different
Page 167
times and consequendy the colour of the image varies in intensity. During the intervals
between repainting, these images could have faded, weathered, or become partially covered
in lichen, rock tripe, or mineraI deposits. Photography provides a detailed record of the
images themselves. Dewdney (Dewdney and Kidd 1962:8-9) took colour photographs of the
sites he described. Only Molyneaux (1979) considered the use of the photographie record as
the sole means of documenting the pictograph sites of the Lake of the Woods. Others
continued to use black and white and colour photographs and slides as additional means of
recording the images (see Pastershank 1989 and Reid 1975). Still other researchers working
on sites elsewhere in the Canadian Shield maintained that photography was too problematic
because photographs were often "distorted" as a consequence of "many factors". They did
not list these factors, other than to state that one of them was the distance of the camera
from the site. It was for these unstated reasons that Pohorecky and Jones considered detailed
field recordings and tracing of the images more important than photographs, whieh,
nevertheless, were important for checking details of the images in question. Pohorecky and
Jones (1967: 305) argued that photography "allowed too much artistic licence" and "too litde
accurate draughtsmanship."
Section 4: T echnical analysis
(a) Technical issues involved in recording and interpreting rock image sites
Sorne archaeologists have discussed the technical issues involved in recording and
interpreting pictographs and petroglyphs (Brand 1979; Chisholm 1982; Pelshea 1979;
Pohorecky and Jones 1966, 1967; Pohorecky 1981; Pufahl 1990; Steinbring and Elias 1968a).
Pufahl (1990) considered that computer enhancement of the images from the Eagle Lake
area was a good idea because he had been forbidden, by the EIders and the Chief of the
Eagle Lake Reserve nearby to touch or trace the images at each local rock image site. Yet
Pufahl was aware that both the EIders and the Chief were concerned about the severe
physical deterioration of these sites because they were sacred to the local indigenous
community and "routinely receive[d] offerings of tobacco" (ibid. 19). Consequendy
photographs were the oilly means allowed by which these sites could be recorded.
Pufahl's initial efforts at using photography were thwarted because of slow shutter
speed, insufficient naturallight, and the need to take all the photographs from a boat. Pufahl
resolved to use polarizing filters with both his camera lens and the flash unit in conjunction
Page 168
with graphically enhancing the images on the computer. He proposed replacing those pixels
where the lichen had encroached upon red images with matching red ochres from the same
area (ibid.: 20). This meant that the image became completely visible instead of partially
obscured. He was aware that potential problems could occur if this process substantially
distorted these images. Pufahl was not only enhancing these images but also manipulating
them. In "replacing" the grey or non-red bits of images, he perhaps unconsciously changed
their shapes. Therefore his daim that his was a radical breakthrough technique for recording
images is slightly misleading.
(b) Conservation and analytical techniques
The concern of sorne researchers lay solely in the conservation of pictograph sites
(Tassé 1979); others desired to conserve petroglyph sites, particularly the petroglyphs in
Petroglyph Provincial Park in southern Ontario (Bahn, Bednarik and Steinbring 1996;
Wainwright, Sears and Michalski 1997). Molyneaux (1974) and a te am from the Department
of Anthropology at Trent University visited twelve pictograph sites in Northwestern Ontario
du ring the 1974 field season to document them and determine the general condition of each
site using "previously defined weathering variables" (ibid.: 1). The reader of this report is not
provided with any means of determining what these variables were, but they dearly were
important enough to warrant a detailed description of the sites for conservation and
protection purposes. Occasionally he described the images that he found at each of the sites.
Seven of the twelve pictograph sites covered in the 1974 study were in the Lake of the
Woods region. The team examined the following Lake of the Woods sites: DiKm-4, DhKm-
5, DhKm-4, DhKm-1, DhKn-1, DhKn-2, DhKo-1, DjKn-1, and DiKn-1. All of the sites
discussed were referenced using Selwyn Dewdney's numbers and their Borden numbers.
A year later, Molyneaux and his colleagues (1975) from the Canadian Conservation
Institute and the Department of Anthropology, Trent University, sought to document in
detail fifteen pictograph sites in the Canadian Shield in terms of physical conditions such as
moi sture access, seepage deposit formation, seepage deposit flaking, exfoliation, and
biological growth. They assessed each site to establish its physical condition with regard to
these criteria. Topographical maps illustrating site locations and scale drawings of the
pictographs were also recorded for each site. One site, DiKp-1, called Sunset Channel in
figure 3.44, in the Lake of the Woods was induded in this study.
Page 169
Figure 3.44: DiKp-1, Sunset Channel, 25th July 1975
The site DiKp-1 was assessed as follows:
The most significant factor affecting the condition of the site appears to be the heaving undercutting of the face by a fracture. The resulting instability will ultimately lead to the destruction of the site when the slab fractures away from the cliff face. Other weathering factors have a moderate effect, although a progressive decrease in pigment density appears to have occurred. The deterioration of the lower sections of the morphs 1 c and 1 e may be related to their proportional1y greater accessibility to wave splash and high water submersion.
(Molyneaux 1975: 4)
Other studies were concerned with the conservation of the paints used ln
pictographs, the conservation of pictograph sites in the Canadian Shield, and the different
ways of examining them (Myers and Taylor 1974; Taylor, Myers and Wainwright 1974, 1975;
Wainwright 1990, 1997; Wainwright and Taylor 1977).
Steinbring and Elias (1968a) advocated establishing the pH level of the mineraI
washes on the surface of pictographs to as certain whether the acidity of the soil above the
site affected the images below. They maintained that the acidity of the soil could affect the
visibility of the images.
Goods (1992) analysed the paint of one morph from the Rice River Pictograph Site in
Manitoba to determine the micrographical, elemental and mineralogical composition, the
constituents, and the paint binder. He discovered the presence of hematite, quartz, potassium
oxide, and aluminium oxide. He argued that organic matter was not detectable either in the
vehicle or the binder.
Page 170
Two additional studies were conducted to discover the composition of the paint, so
as to aid the conservation of the rock image sites in the Canadian Shield. Establishing the
nature of the paint aided conservation since the archaeologists would know what to expect in
the way of deterioration. The discovery of the paint recipes and the organic binders could
also have important consequences for archaeologists investigating pictographs world-wide
regarding their conservation, meaning, and the motives behind their creation. Archaeologists
involved in these experiments argued that the results demonstrated that the artists who
created these paints probably thought about what they were doing (Buisson et. al. 1989;
Steinbring and Simpson 1986; Steinbring, Danziger and Callaghan 1987; Steinbring and
Callaghan 1985). Approximately half of these authors, with the exception of Grant (1967),
Rogers (1962), Rajnovich (1980b), and Dewdney (1970c) considered the petroglyph sites in
the Lake of the Woods.
Grant (1967: 145) considered the style of these images within this geographic region of
North America as "Northern Woodland" which "continued westward" into the prairies of
western Canada as far as the Columbia-Fraser Plateau (ibid.). He maintained that some
"simple abstract elements" existed but that they were subordinate to the naturalistic style
(ibid: 20) and that this style of painting was:
confined to the regions dominated by a nomadic hunting economy. The paintings are mainly simple, rather erude representations of men and animais. In sorne areas the drawings are done in red alone, though black and white were sometimes added.
(Grant 1967: 20)
Other images included: "men in canoes, moose, elk and buffalo," Ojibwa mythological
creatures such as the "water panther," and the "thunderbird," animaIs such as deer and
Page 173
mountain sheep, fish and buffalo, an abstract element which was a semicircle with rays, and
"human forms" of different types (ibid.). The pictographs were "painted (naturalistic)" (ibid:
20), while the petroglyphs existed on horizontal surfaces in Ontario and northern Minnesota
(ibid: 149) and had "crudely pecked animaIs, thunderbirds, and humans" which Grant
asserted were similar to those found at unnamed examples of petroglyph sites in southern
Minnesota. Much of his discussion in his chapter on the Northern Woodland rock image
sites reflects his consultation with Dewdney who was recording sites at the time that this
work was written and published. lndeed, he clearly stated in his acknowledgements (ibid.:
viii) that he could only have discussed the Canadian work after consulting with Dewdney,
who had been actively involved in surveying and recording sites in the Canadian Shield.
Grant clearly recognised a wide range of images. Unfortunately the vocabulary he used to
describe these images implies sorne meaning to the reader. He did not indicate the precise
technique he used to establish the identity of an image. Furthermore, he did not consider
local stylistic traits specifie to different regions in the Canadian Shield.
A site was deemed to exist if Dewdney recognised a clearly defined image existing
there. Dewdney obviously had thought careful1y about how to de scribe these images so that
they could be classified into groups. He classified images according to the categories in the
pie chart in figure 3.45 published in both editions of the book that he co-authored with Kidd
(Dewdney and Kidd 1962: 18, 1967: 18).
Figure 3.45: The table drawn by Dewdney to illustrate how he c1assified the images at these sites.
Page 174
~ ..
Unfortunately, Dewdney's description of these types of images inadvertedly implied their
meaning to either himself or to others. Dewdney probably classified the images using art
historical terms laden with Western cultural baggage such as "naturalism" in his 1962 and
1967 book with Kidd. This was probably because he had been trained as an artist and
illustrator. He stated (ibid.) that his own work suffered from these flaws and problems.
Dewdney (1979: 326) later realised that verbal descriptions of images were prone to
subjective variation and he argued that others were equally guilty of this problem. He
subsequendy suggested that a series of graphic and visual models could provide the level of
rigour needed to classify images because words were not sufficient.
When he commenced searching for pictographs, Dewdney felt that the presence of
red iron ochre on the rock surface was insufficient to designate a physical location in the
landscape as a site. Smears of ochre were "not shapes," from Dewdney's viewpoint in the
1962 publication with Kidd, since they were not recognisable symbols. Older images that had
decayed to the point of becorning amorphous shapes were also excluded as shapes. For
Dewdney, a shape must have a defined edge and be "recognisable."
Images such as blobs and smears, which could have been recognisable images blurred
by time, exfoliation, the accumulation of white mineraI deposits, and by the growth of rock
tripe and lichens, he ignored or left undescribed. Indeed, Dewdney (1975a:4) acknowledged
that he omitted the images that he considered "ambiguous" "when in doubt, leave it out."
Smears differ from blobs since they are usually not only larger than blobs but they could
have been placed on the rock's surface intentionally. Images could fade becoming blobs, but
if Dewdney's approach was used, places with only smears and blobs were not important and
not recorded.
Perhaps the lack of smears, washes, or shapes without clearly defined edges in his
field drawings reflects the fact that Dewdney did not consider these shapes as images. Faint
pale images probably would not be present on his field recordings using paint and rice paper
since the rice paper would probably obscure them when placed on top of them. Perhaps
Dewdney's perspective regarding smears stems from his training as an artist accustomed to
paintings and art objects created in the ]udeo-Christian perspective. However, the index of
his 1967 edition of his publication with Kidd indicates that "smearing of paint on rock" was
considered important since this term was included and it had not appeared in the index or in
the text of the previous publication (1967: 190). Indeed, Dewdney (ibid.: 107) wrote in the
Page 175
text of this edition that "the vague hints of frequent over painting suggest that this was a
long over-used site". The second edition of his publication with Kidd indicates that he had
changed his opinion as to the value of the smear as an image. Dewdney (1979: 326) later
realised that verbal descriptions of images were prone to subjective variation and he argued
that others were equally guilty of this problem.
Rajnovich (1981 b: 286) acknowledged areas of ochre as washes or smears. lndeed,
she argued that a style of image existed in the Canadian Shield region whereby the edge of
the image was painted first and the shape was subsequently filled in with an ochre wash.
Rusak (1992a: 124) observed that only two sites, DhJx-9 and DgJx-2, of the seventeen
pictograph sites that were examined during her 1991 field season in White Otter Lake
(twenty-seven sites are located in this lake) in 1991, did not possess an "ochre wash or
smear". She stated that
the location of washes and smears along prominent naturalledges, in small cubby holes, along noticeable cracks and crevices on the rock surface, indicates the ochre's intentional placement. It is therefore hypothesized that these washes and smears designate the suitable location to leave offerings for the powerful Manitous within [the rock].
(Rusak 1992a: 124) Smears and ochre washes were obviously important images for Rusak smce she
argued that their presence confirmed that these places were important in the world view of
the Aigonquian-speaking peoples. Rusak, like Pastershank (1989), viewed them as individual
images in themselves while Rajnovich (1981 b) perceived washes as a part of an image. Two
years after Reid, Pelshea in 1978, photographed and recorded the images of DiKm-4, DjKn-
1, and DhKm-3 in the Lake of the Woods. Unfortunately, it is impossible to determine what
he and Reid (1976, 1977) thought of the stylistic qualities of the images. Molyneaux (1979)
did not consider the stylistic aspects of the six pictograph sites that he considered. This is
because he (ibid.: 2) argued that "regional interpretations based upon the style typologies and
style areas and supported by ethnographie ethnologies are expedient but largely
indefensible." Rajnovich (1981 b) had discovered several styles of images from either her
involvement in or reading the results of a large number of field surveys in Northwestern
Ontario. lndeed, Dewdney (Dewdney and Kidd 1967: 176) had speculated that more than
one style of paintings existed in the Shield region and that only detailed studies would reveal
the regional styles.
Page 176
Rajnovich (1891 b: 286) maintained that four styles of images existed in Deer Lake:
"open or outlined morphs", "open or outlined morphs subsequently filled in with ochre
wash", "finger lined, or stock morphs" and "closed morphs". She argued that the lines of the
first style of image could signify "internaI organs," a stylistic feature frequently used by Cree
and Ojibwa artists during the twentieth century. She (ibid.: 286) claimed that the images at
the Deer Lake pictograph sites were "distinctive" since a large number of sites occurred in a
relatively small area with "the complete range of morph styles known to Shield rock art". She
(ibid.: 282) claimed that these sites were the sorne of most northerly sites to have been
discovered in Ontario.
Her system of classification of these images indicates that the images and the centres
of the images were not the same. Fortunately she provided a few examples of each style of
image so it is possible to compare the images she recorded with other examples elsewhere.
She reviewed work by her colleagues (Smith 1981 and Pelshea 1980) who also had conducted
research in the West Patricia district, and Dewdney, and concluded that geographical patterns
did not exist regarding the distribution of the morphs in northwestern Ontario. However,
she argued (1981 b: 286) that "open style morphs" were usually found north and south of an
imaginary line between the Lake of the Woods and Lake St. Joseph. She argued that the most
common style of image was the "open and closed style".
Rajnovich (ibid. 287) maintained that the images at Deer Lake were "abstract" rather
than "realistic" based on Dewdney's (Dewdney and Kidd 1967: 19) use of these terms. The
term 'abstract' was more appropriate because she argued that the creators of these images
were more concerned with their meaning rather than their physical shape.
It is interesting to observe that, although Pastershank (1989), Rusak (1992b), Smith
(1981), and Pelleck (1981) each recorded at least one pictograph site, none of them discussed
the possible style of the images.
Rajnovich (1980b and 1981 b) also considered the idea that images occurred in
groups. She observed that occasionally single images occurred at sites such as those in Deer
Lake (1981 b: 287), at the Pukamo Island, the Cuttle Lake, and the Jackfish Lake sites (1980b:
34). She inferred that the presence of repeatedly occurring images indicated that the
"pictograph artists did not perceive the Images as single, isolated figures but drew the
morphs within the framework of a "story"" (ibid.). The presence of a single shape image
Page 177
implied "an object, thought or character" whereas if two images of the same type were
present, then a relationship existed between the two shapes.
Conway (n.d. a. and n. d. b) considered three morphs, from various pictograph sites
in Northeastern Ontario, which occurred in conjunction with each other: an open armed
man, his canine companion, and an animal pelt. Another morph, which Conway identified as
a beaver, existed above the man and his animal companion. He identified these images as
specific star constellations. Rajnovich (1980b: 35) argued that Conway's (1978) identification
of Orion and Canis Major (see discussion page 199ff) at different sites across the Canadian
Shield was problematic. Bear images, she asserted, either occurred alone or in pairs as she
had observed in the pictograph sites of both Rainy Lake and Lake of the Woods. Pairs of
images also existed at Cuttle Lake (Rajnovich 1980d). She did not discuss canine images
although Conway had identified a canine image existing with a human figure.
She also observed that images described as a "canoe-with-passenger" motif existed
throughout the Canadian Shield but that two styles of this shape existed at Pukamo Island
and Jackfish Lake, two sites in the Rainy Lake region. She (ibid.) argued that the image from
the site on Pukamo Island had "stick" passengers while a similar image on the pictograph site
at Jackfish Lake, had "open armed passengers." Unfortunately Rajnovich did not provide any
photographs of either of the images under discussion. She (ibid.) posited that the differences
occurred since the passengers in the canoe were men in one canoe and Maymaygwayshi in
the other. Rajnovich (1980b: 35) maintained that the different style of canoe reflected either
different cultural affiliations of those drawing the se images or time periods in which the
images were drawn.
Rajnovich was clearly concerned with the style of the images in the Canadian Shield,
but she did not pay any attention to Dewdney's advice (Dewdney and Kidd 1967: 176) about
resolving the problem of dating pictograph sites. She asserted that a correlation existed
between Blackduck ceramics and images of hand prints in the Rainy River region. She
established that Blackduck ceramics were found in association with one pictograph site. This
information, in conjunction with information from her informant that sorne of the images
were created by the Midé, led Rajnovich to posit that these images dated to the Blackduck
period and provided a date for the origins of Midewiwin Society. Pastershank (1989: 71), like
Rajnovich (1980b: 35), observed that paired images could be important since pairs of images
were present at each of the pictograph sites she examined in Sabaskong Bay.
Page 178
At two pictograph sites where the images, he believed, conformed ta Grant's (1967)
designation of the Northern Woodland style, Pohorecky (1981) categorised and assigned to
different groups ail the images deemed to be human figures. However, he argued that the
standard categories of classification available for examining images could not be applied to
the "hum an figures" at both sites. Pohorecky did not indicate how he concluded that any of
the figures were human rather than supernatural. He (1981: 12) was principaily interested in
the "human" figures at the two sites on Smith Narrows between Hickson and Maribel1 Lakes
in northern Saskatchewan. He believed that these represented the northwestern tip of this
style, the epicentre of which occurred in Northwestern Ontario. Grant (1967: 147) had stated
that the large concentrations of pictographs had been described in Ontario, specificaily, in
the Lake of the Woods area. The criteria Pohorecky used were whether the figures had
upraised arms, arms hanging down, or one arm up and one arm down or whether the images
were hum an faces with no hands (1981: 13). He stated that the figures in the third category
figures, and turtles. Reid (ibid.: 246) observed that none of these sites had images which were
deemed as "typically Archaic projectile points, or atl-atls", which Lothson (1976: 31) had
used to date the] effers Petroglyphs in Minnesota, but a number of similar animal and human
forms were present. He reasoned that the excavation of the Meek Site, DjKp-3, which had a
habitation site as well as petroglyphs, provided sufficient evidence for testing two
hypotheses: (a) that the majority of the rock image sites on the Lake of the Woods were
"temporally placed" in the Middle and Late Woodland period and were "a product" of the
Algonquian speaking peoples, and (b) that the petroglyphs at DjKp-3 (Tranquil Channel)
were created by the Cree during the tenth to the fifteen centuries AD.
Reid maintained that these petroglyphs indicated an Algonquian cultural affiliation of
these sites. Based on two radiocarbon samples, this archaeological site belonging to the
Selkirk culture was dated to 935 AD. and 1420 A D (ibid.: 252). Reid (ibid.: 253) reasoned
that majority of the Lake of the Woods sites were interrelated, based on their shared imagery.
He argued that the date from the habitation site supported his hypothesis that the Lake of
the Woods sites were relatively recent Algonquian ones, because the fertility images at
Tranquil Channel were connected to the nearby archaeological site.
If the ]effers Petroglyph site and DrKr-4 were connected, they belonged to "an
extremely long cultural" period and that the relationship between the petroglyph and
pictograph sites in the Lake of the Woods must be considered (Reid 1979: 247). He noted
that the Canadian sites shared similar animal and human forms with the American site. Reid
reasoned that several images at petroglyph sites DjKp-3, DiKp-2, DrKr-4, DiKp-4 and
DkKq-24, in the Lake of the Woods, could be identified as fertility symbols, stick figures,
and turtle shaped images. He concluded that, since the petroglyph sites in the Lake of the
Page 189
Woods shared the same images, they were probably connected to each other as weIl as to the
pictograph sites. Reid (ibid.: 247-250) asserted that the offerings left at DiKm-1, DhKm-1,
DiKp-1, DjKl-2, DkKn-7, DkKr-4, DhKm-3, and DhKo-1 were additional evidence of a
connection between the pictograph sites and DiKo-2, a petroglyph site. These offerings
were, for Reid, an indicator of religious importance and provided a sense of "the sacred" for
local indigenous communities.
One particular morph, the key factor for Reid (ibid.: 250), demonstrated a link
between these sites and the belief system of the Midé of the modern Ojibwa. This morph
was identified by Redsky, a Midé practitioner from Shoal Lake (part of the Lake of the
Woods) as a "paisq", which he roughly translated as a "bird of omen". This image was
identified at two of the petroglyph sites: DiKo-2 and DjKp-3, three times on the Massacre
birch bark scroIl in figure 3.46 and on an Ojibwa birch bark scroIl caIled a "migration chart"
held in the National Museum of Canada.
Figure 3.46: The Massacre Scroll photographed in 1993 by J. Rusak.
Reid (1979) has taken a different approach to dating the images of the petroglyph
sites since he did not rely on absolute dating techniques. He argued that none of the
petroglyph sites from the Lake of the Woods had shapes that resembled the images at the
]effers Petroglyph Site in Minnesota dated to the Archaic period. Steinbring and his
coIleagues (Steinbring and CaIlaghan 1985; Steinbring, Danziger and CaIlaghan 1987)
subsequently advanced an Archaic date for the Algonquian occupation and creation of
DkI<r-4 (Mud Portage), one of the sites that Reid had considered. They argued that DrI<r-4
was connected to the ]effers Petroglyph site since Archaic projectile points and atl-atls were
depicted at this site.
Page 190
Two articles debated about how a method might be developed so that images from
different sites could be related to each other. Whelan's (1983) study built upon the work of
Maurer and Whelan (1977). They sought to develop a method applicable to other forms of
"pictography within this region" (ibid.: 196), by which sites and paintings could be classified
and catalogued according to the subject matter of each image. Each image, termed data, was
placed into different levels: class, subclass, and attribute (ibid.).
Maurer and Whelan observed that problems existed in trying to conduct this type of
analysis since the descriptions of the images were invariably affected by the personal
interpretation of the observer. They discovered that difficulties existed in developing a
format that was relatively adaptable to a computer programme: how could an archaeologist
de termine and identify what constituted a class, a subclass, and their respective attributes,
and how could they identify specific symbol types and names (Maurer and Whelan 1977:
199).
Maurer and Whelan were determined to deal with the increasing quantity of data and
to clearly describe and classify these images without imposing meaning upon them during the
culture-historical approach. They understood the issues of unintentionally imposing meaning
to an image in the course of describing its attributes.
Section 6: Contextual approach:
This approach will be used, only after the application of the culture-historical
approach, to relate various types of empirical information to the site where the images were
found. It considers a broad set of associations and relations among the images themselves,
and between the images and their physical settings. This approach involves a search for
patterns relating different images, and combinations of images both between and within sites
Archaeologists inevitably describe the immediate physical landscape. Petroglyphs
exist on the surface of a rock while pictographs exist on rock surfaces either on rock walls or
in caves. It was traditionally believed that pictographs exist at the base of vertical granitic
rock walls, either immediately beside the water, or near the water's surface. Pohorecky (1968:
103) stated that:
the micro-environment of the rock paintings is easily categorised. The paintings occur either singly, or, more commonly, in small groupings on fiat, vertical and sloping rock faces or cliffs just above the surface of sorne body of water, and can be reached usually from sitting or standing position in a canoe, and sometimes by stepping onto ledges just back from the water. AlI paintings are visible from the vantage point of a canoe. This
Page 191
kind of setting, where paintings lie near the edge of the water in summer or ice in winter, is itself one of the most distinctive features of rock paintings throughout the entire area.
Yet, pictograph sites are not always found immediately near water. This statement is
supported by the pictograph site DhKn-1, first recorded by Dewdney as Site # 106 on 19th
August 1960 in figure 3.11. It is on the southwest facing shore of Aulneau Penin sula
immediately opposite Wisakode Island and north of Rabbit Island. It lies beside one of the
channels that can be used to enter Obaibikon Bay to the north. The paintings here can be
reached by standing on a narrow, long rock ledge immediately behind the large boulders and
the birch trees visible in figure 3.47. These boulders are a few metres from the water's edge,
where various grasses, trees, and short plants grow.
Figure 3.47: Part of DhKn-l beside Obabikon Channel.
Two studies from neighbouring regions exemplify the contextual approach because
the authors sought to ascertain how rock image sites were related to other features in their
vicinity (Noble 1968; Sweetman 1955). Noble described and discussed the pits, cairns, and
petroglyph sites found in close proximity to each other at Rock Lake, in Algonquin Park in
southern Ontario. He (ibid.: 63) contended that the rock cairns behind the rock lined pits
were not definitely connected with each other, although they were related in some manner.
The pits were interpreted as '''vision pits'" (ibid.: 62), while the cairns were interpreted, based
upon comparison with cairns at Red Sucker Point, on Lake Superior, as possibly being
""tobacco-drops", dedication cairns erected to a guardian spirit after a successful vision"
Page 192
(ibid.: 63). The petroglyphs were interpreted as being both archaeological and artistic in
nature and "of probable socio-religious import" (ibid.). He asserted that they were not as
complex, numerous, or detailed as those elsewhere in the Canadian Shield, discussed by
Sweetman (1955) and Dewdney and Kidd (1962). Noble also concluded that it was
impossible to infer the cultural affiliations of the people who used the pits.
Sweetman (1955) described how the images at the Peterborough Petroglyph site were
mapped and recorded. The images, called "glyphs" (ibid.: 101), were recorded using plaster
moulds, drawn, and photographed. Test pits were dug near the petroglyphs to establish
whether the people who had created the glyphs had left their tools nearby, or camped in its
vicinity while they created the images (ibid.: 101-104). Nothing but tree stumps or windfalls
were found in the test pits and these could not provide any information about the creation of
these images. An archaeological site of an Iroquoian village, which he called the Quakenbush
site was found nearby (ibid.: 104). Sweetman argued that it was unlikely that the inhabitants
of this village had created the petroglyphs, but he did not indicate why he arrived at this
conclusion. He (ibid.: 108) concluded that it was impossible to arrive at any clear conclusions
regarding the age and meaning of the images.
Arsenault and his colleagues have examined in considerable detail eight pictograph
sites in Quebec (Arsenault 1997; Arsenault, Gagnon, Martijn and Watchman 1995; Arsenault
and Gagnon 1998). Arsenault (1998) examined in considerable detai! the sacred space, and
physicallocality of eight pictograph sites in northern Quebec. They are in different regions of
the province: one is in the James Bay area, two are in Abitibi area, one is in the
Témiscamingue, two in the Outaouais region, one in Mauricie, and one in Haute-Côte-Nord.
Nevertheless, he drew considerably upon ethnographic data from western Ontario. He
argued from the substantial ethnographic record of the Ojibwa, that rock image sites in
Quebec were sacred places used by shamans.
He drew on ethnographic data collected from throughout Canadian Shield rather
than just specific regions in Quebec. This is surprising, especially since the Algonquian
groups of the western Canadian Shield probably had different cultural experiences from
those in Quebec; although these images were created by Algonquian-speaking peoples, they
were created by different groups. It is therefore possible that there were specific differences
in the religious practices and perhaps the use of images between the Ojibwa, living farther
west, and the local groups of Algonquian speaking peoples living in these regions of Quebec.
Page 193
Arsenault considered that pictograph sites may have acoustic properties that explain
why some rock faces rather than others were utilised. Archaeologists, elsewhere in the world,
have investigated whether site selection occurred as a consequence of the special sound
effects that could have been produced in specific spaces (Rainbird 2003, Waller 1993).
Both Dewdney and Kidd (1967) were obviously concerned with identifying what
constituted a site. Yet, in the same breath as they (ibid.: 6-8) delineated the properties of a
"typical site," they stated that it was difficult to generalise about the properties of a typical
site. Evidently, the visual landscape was not considered important for understanding these
images. Nor is much information recorded as ta how, precisely, sites were located. It appears
that Dewdney relied upon information provided by inhabitants of the areas where he sought
to find rock art sites (ibid.: 7-8).
Section 7: Literature concerned with meaning
The three remaining approaches are aU concerned with establishing the meaning of
these images. For a detailed discussion of each approach see Chapter I.
(a) Intuitive approach:
Archaeologists who have adopted the intuitive approach to explain rock image sites
in the Canadian Shield include Granzberg and Steinbring (1995), Lanoue (1989), Pettipas
(1982), Steinbring (1976, 1979b, 1982), Steinbring and Elias (1967, 1968b), Steinbring,
Granzberg and Lanteigne (1995), Steinbring and Steinbring (1978) and Vickery (1991). Only a
few of them have referred to sites in the Lake of the Woods.
Steinbring and Steinbring (1978) postulate that the images found at Machin Point,
DjKr-1, in the northwest part of the Lake of the Woods, were created as a consequence of
shamanism, based on a study of shamanism in connection with a petroglyph site in British
Columbia by Walker in 1977, and ethnographic data from the Winnipeg River watershed
(Steinbring and Steinbring 1978: 13). Both archaeologists speculated that a "template effect"
existed in the distribution of the glyphs of DjIv-1, which "seemed to be closely related to"
the naturallines evident in the rock upon which the images existed (ibid.: 13). The authors
concluded that the natural lines guided the creation of the images in a manner related to
Algonquian scapulimancy.
Page 194
Some of the images were identified and discussed according to the identification of
both archaeologists rather than merely described. According to the authors, careful scrutiny
of the images indicated that "vandalism" had occurred at this site. The green paint in some
of the "authentic-appearing" images, which had been created using a variety of pecking
techniques, was considered as evidence of vandalism. The authors maintained that they were
able to deduce that "fraudulent glyphs," made by short pecked lines, also existed at this site.
Low-level interpretation is embedded in the descriptions of these images, such as the
"dragonfly" petroglyph discussed by Steinbring and Steinbring (ibid.).
Steinbring (1982) later drew upon a variety of ethnographic sources to substantiate
his arguments that the images found in pictographs and petroglyphs in the Canadian Shield
existed as a consequence of Ojibwa and Algonquian shamanistic activities and dreams. The
petroglyphs of the Lake of the Woods sites were only briefly mentioned. The goal of his
paper was to argue that the art of contemporary native artists, with Algonquian backgrounds,
had the same origins as the crea tors of the petroglyphs, created during the Archaic Period.
He overlooked demonstrating precisely how this was possible, since he neither provided
evidence that he had consulted or interviewed contemporary native artists, nor suggested
who these peoples might be.
Steinbring (1976) speculated about the origins of the copper technology and lunar
cultism in the ancient Great Lakes region. He (ibid.: 164) argued that creseent shaped forms,
which he eailed lunar motifs, were present at rock image sites and on bireh bark seroils
created by the Ojibwa. He asserted that substantial ethnographie details existed concerning
lunar symbolism in the ethnology of the Great Lakes Algonquians, but provided no
references for his readers. He subsequently identified erescent shaped motifs at the
Peterborough Petroglyph site and disagreed with the Vastokases' (1973) identification of
these images as boats. He argued that some of them were probably not boats with people in
them but eould "actuaily represent copper erescentie symbols" (Steinbring 1976:. 164). He
subsequently argued that similar images are found at a number of sites in the Lake of the
Woods and elsewhere in Northwestern Ontario and eastern Manitoba (ibid.: 164-165). For
eaeh example given of a erescent shape at a site, he disagreed with its interpretation by
previous archaeologists, arguing that "many crescents have probably been misinterpreted as
boats or eanoes, the most obvious of these being the "upside down" ones (ibid.: 156).
Page 195
Granzberg and Steinbring (1995) classified glyphs by type usmg the intuitive
approach. They argued that line, tree, and circle motifs were "core symbols with a
transcultural base line of archetypal usage" based on examples taken from sites in the
American Southwest and from northern Manitoba and Northwestern Ontario (1995: 52).
They postulated that their data suggested, despite the lack of ethnographic references, that a
certain level of trans-culturally valid meaning existed, allowing the authors to identify distinct
types. However, both authors maintained that considerable research was necessary before
any categories could be securely identified (ibid.).
Lanoue's (1989) study considered the petroglyphs of the Lake of the Woods. He and
Richard Doble, from the University of Winnipeg, exarnined and recorded an unknown
number of petroglyph sites in 1981. Lanoue asserted that these rock image sites were
"stylistically Algonkian". He argued that these images, which could not be placed "in the
tradition of 'great art'" (ibid.: 7), were "the product of similar intellectual processes and
formally the same everywhere at all times" (ibid.: 9). Lanoue drew (ibid.: 42) upon
Steinbring's work (Steinbring et. al. 1987) to postulate that the petroglyphs were at least
5,000, and possibly even 9,000, years old. He maintained that his own examination was
flawed because he did not have the full body of data available to him since the full repertoire
of images in the Lake of the Woods was not yet established. He posited this was because (a)
not all the petroglyph sites had been found and (b) sorne were covered up with water and
therefore not visible.
Using CUITent knowledge concerning hunter-gatherer and tribal societies, Lanoue
proposed that, if the style used to create the images was related both to their meaning and
content, useful information regarding the "central paradox of territoriality" might be
obtained. He maintained that the Ojibwa material from Rama Reserve, northwest of
Toronto, should be used to illustrate his answer rather than analyse his data. He was
particularly interested in the animal images, referring to them as "symbolic mediators of a
paradox which characterises territorial hunting and gathering societies" (ibid.: 64). Images
were stylised since their meaning occurred as a consequence of its association with different
"symbols and symbol sets" instead of being a point of reference (ibid.: 67). Two sub-styles
were identified in the depiction of the animal images that were not easily recognisable. He
claimed that those naturalistic petroglyphs that were "stylized through simplification"
represented the paradox of location and mobility (ibid.: 68), while those animal images that
Page 196
were simple in design but distinguishable from the first group because traits had been added
that were "clearly not natural" (ibid.: 69), represented the intellectual paradox. The second
group had arisen as a consequence of the creation of boundaries which separated human
beings from animaIs, on the one hand, and the required dependency on the world of animaIs
to create a social world. Therefore, if specifie design traits indicated that an image was not
normal, or "natural," it meant that the authors of the images referred to ideas about the
supernatural rather than natural beings. Lanoue maintained that the people who occupied the
Lake of the Woods during the historie period were "preoccupied" with questions of
territoriality and desired to situate themselves not just using "a social and spatial grid" but
also by involving "abstract expression."
(b) Analogical Approach:
Cailahan (1998) has applied the neuropsychological "model", developed by Lewis
Williams and Dowson (1988) to this geographical region of North America but only to sites
in Minnesota (Cailahan 1998). The rarity of this approach is interesting, given its popularity
elsewhere in the world.
(c) Homological Approach:
This is the most popular approach for archaeologists interested in determining the
meaning of rock image sites since meaning was drawn by comparing archaeological data with
related historie and ethnographie cultures (Conway n.d. a, n. d. b; Conway and Conway 1989,
1980c, 1987; Tassé 1977b; Turner 1979; Vastokas and Vastokas 1973).
Lipsett (1970) compared the images of specifie pictographs with images created and
used by the Ojibwa, as discussed in Hoffman's (1891) report on the Midewiwin Society and
the different types of religious practitioners among the Ojibwa. Molyneaux (1987) took a
similar approach to the images he found at several pictograph sites in Northern Ontario,
including DhKo-l and DhKm-5 in Whitefish Bay, in the Lake of the Woods. Dewdney
(1970a) developed the idea that by comparing the style of rock images and those made by the
Midewiwin on birch bark scroils, it might be possible to distinguish between historie and
precontact images. He and Kidd (1967: 171) had argued that the images on the birch bark
Page 197
scroils would probably provide clues to the meaning of the images on the rocks ln the
Canadian Shield.
Dewdney and I<idd's (1962, republished 1967) study was arguably the first
comprehensive investigation of rock image sites of the Canadian Shield. Dewdney examined
twenty pictograph and petroglyph sites in the Lake of the Woods. He also posited a
connection between the images on rocks and the Midewiwin, based on the similarities in
morphs executed at rock image sites and on the birch bark scroils (1967: 106-110).
Numerous sites were described in varying degrees of detail (ibid.: 22). Unfortunately, detailed
interpretations of each site were not provided, although much of this information was
retained in Dewdney's private archive, which is now kept by the Royal Ontario Museum.
The age of the images was estimated from their style. Dewdney and I<idd speculated
as to why the sites were painted and who the artists were (ibid.: 16-20, 102-116). They
hypothesised that sorne of these sites depicted articles of European origin such as "an early
European fort" and "the f1ag" (ibid.: 48). The distinction between explicit "description" and
low-level "interpretation" is blurred in Dewdney and I<idd's descriptions of the different
images at each site. An example of this exists for site # 106 on the shores of Obaibikon
Channel, on Sabaskong Bay, now cailed DhKn-1, in the Lake of the Woods (figure 3.11).
Dewdney wrote about this site:
Speeding up the channel from Sabaskong Bay we caught a glimpse of red through the trees high up on the east shore. On shore, expecting to find another example of iron stains, we were happily astorushed to discover the paintings shown here: two serpentine figures one with antlers the other with horns, symmetrically facing a large turtle. To the left, rather crudely painted on very rough granite, was a serpent fifteen feet long, with open mouth, ears and three large flippers - a veritable Ogopogo.
A deep cleft between the ledge we stood on and the rock was almost filled with dirt and rubble. Lying on the ground was an ancient, weathered overcoat, and various rags that had rotted beyond recognition.
(Dewdney and Kidd 1962:43)
The "veritable Ogopogo" is visible in figure 3.47 while his "serpentine figures" are visible in
figure 3.48 below.
Page 198
Figure 3.48: The serpents and the turtles described by Dewdney in 1960.
Dewdney often classified the images using Western art historical terms such as
"naturalism", perhaps because he had been trained as an artist and illustrator. His cultural
background clearly influenced rus verbal descriptions of the different sites. Another example
relates to a pictograph site called Painted Rock in Lac La Croix:
For sheer naturalism there are no other paintings of moose that l have seen in the Shield country ta compare with the three on this site. Ail are surely by the same hand, as is the little antelope-or deer. Unique, too, are the pipe-smoking figures: one beside an hour-glass figure and tracks, the other not far from the initiaIs "L.R. 1781".
(Dewdneyand Kidd 1967: 26-27)
Dewdney and Kidd each dealt with separate components of this study. Dewdney
located, recorded, and described the pictograph sites and occasionally obtained sorne
ethnographie details (ibid.: 11-14), while Kidd considered and examined the related
anthropological data (ibid.: 159-177). However, trus split between the work undertaken by
Dewdney and Kidd is misleading: it may simply reflect academic traditions at the time their
book was written. Yet, throughout, Dewdney speculated as to the authors of these images,
why the images existed, their meaning, and the paint ingredients, prior ta discussing the
different sites and their respective paintings. Kidd (ibid.: 159-177) summarised the rustory of
the interest in these images, the archaeology and the early rustory, and considered the
Algonquian-speaking peoples in the Canadian Srueld. He, like Dewdney, speculated as to
why the images were created, the possible authors of these images, the paint's constituents,
Page 199
and the techniques of applying these images, as well as different possibilities for establishing
their meanings. Neither I<idd nor Dewdney considered that the meanings of these images
might be polysemic, despite I<idd's discussion of the Midewiwin Society and its members
(ibid.: 167-168).
The second (1967) edition contains considerably more information, but suffers from
the same problems. Dewdney and I<idd's study represents the first major step in the
homological analysis of these images. Although Dewdney was not an archaeologist, his work
on these images was immensely influential.
Dewdney (1970b) used ethnographic evidence concerning shamanic practises among
the Ojibwa to postulate the reasons for the creation of pictograph sites. He illustrated his
arguments using individual images from four sites in Whitefish Bay, Lake of the Woods:
DiKm-3, DhKm-3, DhKm-S, and DiKm-4. He drew upon information regarding the
worldview and religious practices of the Algonquian-speaking peoples collected by
anthropologists, and from his numerous conversations with indigenous informants, including
Norval Morrisseau, the Ojibwa artist, to try to determine how and why these paintings were
created, and what they meant (ibid.).
Dewdney argued that the pictographs, like the other images created by the
Algonquians on wood, birch bark, and hide were all forms of expression (ibid.: 22) that he
subdivided into three categories: secular, tutorial, and visionary. The images on the birch
bark were deemed secular and possibly included those le ft at portages for others, to give
practical information such as an individual's 'signature', a representation of his totem or his
name" (ibid.). He maintained this mark or image would have also existed on personal objects
that belonged to individuals. Yet, since Algonquian society was collective, an individual
would only have left a "mark" for territorial reasons. Tutorial images were defined as those
that were used between a shaman and his clients or those used as mnemonics on the birch
bark scrolls (ibid. 23). Dewdney deemed sorne images as visionary, based upon his interviews
with older generations of Ojibwa.
Dewdney (1970b) asserted that the description and classification of images grouped
together by their physical shape was problematic, since images should not be examined
separately if they were originally part of a larger group from a site. This daim occurs in an
article that seems to have been part of a larger research project that he was undertaking prior
Page 200
r-..
~ ..
to ms death. It seems evident from reading ms research and field notes3 that Dewdney may
have been synthesising the information he had collected. He had noted that twenty-six
percent of the sites that he had recorded were dry sites and seventy-four sites were wet sites.
What he meant by the terms "wet" and "dry" is undear. It is also dear that he sought to
identify ail the different variations of each type of image he had recorded. He provided the
number and gave the name of the pictograph site, if it had one, where each image existed.
What precisely Dewdney intended to do with this analysis remains unknown.
He was also interested in whether the type of image he had identified at each site
existed only at certain orientations, or in certain parts of a pictograph site and whether tms
changed according to where in the Canadian Smeld the site was physically situated. This
work is possibly related to some ideas that Dewdney (197 Sa: 1) expressed in an
Archaeological Newsletter that perhaps understanding how the stylistic variations were
distributed throughout a region provided chronological dues to dating the images of the
Shield region. He argued that he initially posited that using a computer provided the most
objective means of sorting and correlating the images which he had collected over the course
of ms research. He posited that he had approximately 3,000 figures which he termed morphs
(ibid.: 2). He realised that although he had used a computer that "an inadequate typology,
loosely defined coding values and inconsistencies (Plus too many errors)" led mm to
condude that higher standards were required in the initial phase of a study (ibid.). He
concluded ms short piece by reminding his colleagues that it was important to be as
meticulous and objective as possible and that he was guilty of being too subjective when
recording and describing ambiguous images (ibid.: 4).
It is dear from ms notes, held by the ROM, that he was conducting sorne systematic
analysis using a computer with colleagues at the University of Western Ontario. Another line
of research that concerned mm in ms notes is mentioned in an ROM Archaeological
Newsletter dated 1971. He discussed his observed correlations between specific pictograph
sites and "vision pits" which were special places that he believed were chosen by shamans
for their vision quests (ibid: 3). He posited that DhI<m-3 was possibly one such example of a
place that a shaman had chosen as a "shaman's retreat," especially given its physical
characteristics: the images were beside a bench "large enough to pitch a tent on, and far
3 Dewdney's field notebooks, field drawings, slides and photographs are held by the Department of Anthropology, Royal Ontario Museum, in Toronto Ontario, Canada.
Page 201
above the water" (ibid.: 4). However, he concluded that further research was required to
substantiate his opinion that pictographs sites were probably connected with "vision pits"
(ibid.). Clearly severallines of research were being explored and developed for some purpose
that is now unclear and difficult to establish.
Vastokas and Vastokas (1973) attempted to understand the mearung of the
Peterborough Petroglyph Site. These images were pecked from crystalline lime stone, one of
the softest rocks in the Canadian Shield, and ground out using gneiss hammerstones (ibid.:
14-18). The petroglyphs were speculatively dated, based on available data, to the end of the
Woodland period, between A. D. 900 and 1400 and attributed to prehistoric Aigonquians
(ibid.: 20-27). The Vastokases postulated that a connection existed between the engraved
images on the rock faces and the images on birch bark scroils created by the members of the
Midewiwin Society (ibid.: 40-47). They surmised, from ethnohistorical and anthropological
sources, that the images were located where individual shamans believed they could consult
the resident spirits (ibid.: 47-54) and that unusuaily shaped large rock outcrops, rocky hiils
and boulders with crevices and holes were the "dwelling-places of the manitous and mythical
creatures such as the Maymaygwayshi" (ibid.: 48).
Vasto kas and Vastokas discussed at considerable length how the Algonquian universe
was organised and how the physical locations of the rock images in the landscape were key
loci, since these places were where the different worlds co-existed and hence these places
were "the seat[s] of visionary experience and super natural counsel" (ibid.: 54). They (ibid.:
29-40) discussed the nature of Algonquian hunting and gathering practices and the different
objects upon which similar images appeared, and maintained that the manner in which
shamanism, Algonquian cosmology, manitous, vision quests, and guardian spirits were
interrelated provided the information required to understand the petroglyphs. They
postulated that a detailed understanding of Aigonquian use of pictographs to convey
information would aid modern acadernics to understand these images.
Images were analysed once they had been classified by type and placed in different
broad groups according to subject matter (ibid.: 55-129). The Vastokases noted that,
although over nine hundred glyphs existed at the site, only three hundred were readily visible.
The others were deemed "vague and their identity is difficult to de termine owing to the
degree of overlap and superimposition, wear and erosion of the limestone surface" (ibid.:
55). They subsequently asserted that the variety and extent of the images' meaning was only
Page 202
accessible once detailed analysis of this site was completed (ibid.). Interpretation and
discussion of the possible meanings of each group could occur once the identifiable images
had been classified by shape and subject matter, and with reference to ethnographic and
ethnohistoric data. Three years earlier, Dewdney (1970b) had asserted that the description
and classification of images that were grouped by type (their physical shape) was problematic,
since images should not be examined separately according to their physical shape; especially
if they were originally part of larger group of images which were different.
It was possible that the meanings of the different images were interrelated. Based on
the changes in the styles of the images, the Vastokases (ibid.: 131-134) concluded that the site
had probably been used repeatedly over a long period of time. They speculated that these
stylistic changes were possibly connected to technological, cultural, or temporal factors. They
(ibid.: 137) asserted that the images of the Peterborough Petroglyph site should only be
compared with other petroglyph rather than pictograph sites to establish stylistic affinities.
However, this statement is problematic, because it is possible that the crea tors of these
images did not differentiate between images ground into stone and those painted on stone.
The Vastokases concluded that the rock itself, the landscape, was crucial to the meaning for
these images and to understanding why it was used for these images. Although this study is
over thirty years old, many of its conclusions remain important to the analysis of these
images today.
Both of Conway's (n.d. a. and n. d. b) papers examined sites in northeastern Ontario.
He (n.d.a) was concerned specifically with three morphs, from various pictograph sites in
northern Canada, which occurred in conjunction with each other: an open armed man, his
canine companion, and an animal pelt. Another morph, which Conway identified as a beaver,
the pelt, existed above the man and his animal companion. He identified these images as
specific star constellations. The open armed man was identified as Orion since the paintings
had anatomical attributes that were similar to those of the constellation. The position of
canine companion, identified in the pictograph sites as either a dog or wolf, concurred with
that of Canis Major. The bilateral symmetry and location of the beaver matched that of
Gemini to Orion (ibid: 1). Conway (ibid. 1-2) drew upon ethnographic information collected
by Schoolcraft, Speck and some other unnamed individuals, and from his own informants, to
identify specific constellations as animaIs and supernatural heroes. Schoolcraft was important
for Conway, because he was particularly interested in native astronomy and cosmographic
Page 203
~ ..
information. Conway Qbid.: 3) maintained, based on unknown sources, that native
astronomy depicted at pictograph sites had "considerable antiquity" since according to him
such paintings had ceased to be created after 1650 A. D. by Ojibwa shamans and medieine
people. He asserted that few sites showed historie scenes involving Europeans, but provided
little evidence to substantiate this daim. He argued that the style changes and presence of
constellation groups in the pictograph sites indicated that native cosmographie information
had existed since the prehistoric era. Fourteen constellations were identified, but Conway
was specifically interested in the Orion trio whieh he identified as existing on a birch bark
scroll made for Schoolcraft by a Lake Superior shaman as weIl as in some unnamed
"examples of Ojibwa rock art sites" (ibid.: 4).
Conway (ibid.: 4-5) asserted that the "Orion trio of paintings" was explicable using
information from popular native folklore. However, he argued that it was possible to put
forth another interpretation of these images, since the constellation paintings behaved as a
mechanism by which seasonal changes were indicated to the local indigenous population.
Orion was interpreted as a celestial marker for "a shift from late fall to winter camps" (ibid.:
5). The depiction of constellations at pictograph sites indicated that these sites were "sky
maps" which enabled information about seasonal migrations through the landscape to be
transferred from one generation to another. Conway conduded that rock image sites should
be considered as "one of the few permanent markers" in the lands cape that provided
knowledge to the transitory Ojibwa hunter-gathers of this region.
Another paper by Conway (n. d. b) is perhaps a later draft of the paper previously
discussed (n. d. a) as its content is similar. In this second paper, he hypothesised that specific
images were based upon the constellations recognised by prehistoric Algonquians. These
images originated from sites in an area occupied by the historie Ojibwa, Cree and other
Algonquian bands (ibid.: 5). Conway argued that this combination of morphs was first
identified at a site by himself in Matagamasi Lake, in northem Ontario (ibid.: 7). He
subsequently briefly discussed Dewdney's published works, the Vastokas and Vastokas
(1973) publication, as weIl as various site reports from throughout Ontario in an attempt to
establish whether this triad of images existed. Conway maintained it was easier to identify
whether these three images existed together if the painted figures were "not as dustered"
(ibid.: 8). This triad of images existed in only two clusters in the Temagami stylistic and
Shield Edge stylistic areas (ibid.: 9). It was absent elsewhere in the Upper Great Lakes area
Page 204
and north central Ontario. Unfortunately, Conway did not define what constituted the
Temagami or the Shield Edge styles. Two types of canines were identified to exist on their
own or with a human figure: a dog and a wolf. Two examples of the dog morph were
identified at two sites in the Lake of the Woods: DiKo-2 and DhKm-3. Wolf morphs were
identified at one undisclosed site in the Lake of the Woods. But Conway did not establish
how wolf and dog morphs were different from each other.
Conway's conclusion had four points. Firstly, two distinct canine morphs existed in
the rock images sites in Ontario, which Conway identified as a wolf and a dog. Nevertheless
both images could denote "the same Algonkian concept" (ibid.: 12). Secondly, both morphs
had comparable limited geographical distributions, namely the northeastern edge of the
Canadian Shield in Ontario and north and west part of the Lake of the Woods in north
western Ontario. Thirdly, dog canine morphs often occurred with hum an figures while wolf
canine morphs existed with human figures that had their arms open. Conway's fourth
conclusion was that "an Algonkian triad" existed at the majority of sites that had the wolf
morphs, and consisted of "a human figure or shaman figure with outstretched arms, a pelt or
a cross" (ibid.: 12). Conway's conclusion is interesting since it demonstrates that his own
Eurocentric cultural perspective was influential in his choice of images to examine, despite
his utilising the pertinent ethnographie evidence from the region where the sites existed.
Fol1owing Hodder (1987), Rajnovich (1994: 160) argued that it was possible to arrive
at sorne conclusions that could not "be absolutely proven" (Rajnovich after Hodder 1987: 6)
but were plausible. Rajnovich (1989, 1994) repeatedly drew on the Lake of the Woods as a
source of examples for her discussion of rock image sites in the Canadian Shield. She
postulated that archaeological, ethnographie, and anthropological evidence should be
combined with linguistic and oral evidence to establish the link between the different types
of Ojibwa shamans, Midé and Jiissakid, to obtain an understanding of these images. She
(1994: 19-22) asserted these images were signs, probably mostly polysemie metaphors, since
Algonquian languages are metaphorical. She (ibid.: 22-39) discussed the Algonquian world
view, the practice and use of indigenous plant medicines, and the different types of religious
practitioners. Rajnovich (ibid.: 29) maintained that only the Midé and Jiissakid, two of three
types of religious practitioners, rather than the Wabeno, created rock images.
She appears to base her arguments on two rock image sites, providing a field drawing
of each site: Dewdney's field drawing of Painted Rock Island, DhKo-1, in the Lake of the
Page 205
Woods and a petroglyph site in Uskik Lake in Saskatchewan. She (ibid.: 41-55) debated the
age of these images and concluded that it was only possible to use relative dating techniques
and to draw upon ethnographie, and related archaeological data from nearby peoples to posit
the age of individual sites.
She (ibid.: 57-63) related the images present on the Qjibwa birch bark song scroils to
those on the surface of rocks by drawing on the work of Garrick Mailory (1880, 1881 and
1884), Tanner Games 1956), Hoffman (1888, 1891, 1896), Kohl 1985 [1860], and Densmore
(1910, 1913, 1926, 1979), as weil as sorne unpublished studies in the Smithsonian
Institution). Rajnovich (1994: 57-63) postulated that the metaphorical nature of the
Algonquian languages provide clues to the interpretation of the paintings and engravings.
She (ibid.: 65-66) argued that the pictographs were standardised in the sense that other
people could recognise the images and interpret them but that different and legitimate
meanings could be attached to specifie symbols.
To demonstrate why and how these images were polysemie, Rajnovich first discussed
the importance of the cliff faces and then a range of images with reference to ethnographie
sources and informants (ibid.: 66- 143). Each type of image was analysed and interpreted
after it had been grouped into a broad shape type. Meaning was ascribed to an individual
shape through references to data from ethnographie and ethnohistoric materials, as weil as
from informants.
She ignored Dewdney's (1970b) stricture that images should not be interpreted
individuaily from the other images with which they exist. Her practice of first dividing images
into broad groups also may cause problems. This activity is probably related to the manner in
which rock image sites are treated by archaeologists accustomed to coilecting data according
to archaeological conventions and guidelines. It is surprising that her discussion of these
images did not draw on the conclusions about these images that she had published earlier in
government reports (for example Rajnovich 1980b, 1981b).
Rajnovich (ibid.: 145) observed that a common explanation for the presence of these
images was "hunting-magic", (ibid.: 145- 157) and discussed at great length what this could
mean with reference to the Algonquians of the Canadian Shield. She subsequendy argued,
based upon ethnographie data, that rock images appeared on rocks where manitous lived
(ibid.: 160), and that these sites occurred beside water because that was where the sky, earth,
water, underground, and underwater meet. Places su ch as deep lakes, whirlpools, bases of
Page 206
.~.
lakeside cliffs, caves, and crevices were the locations used to cross from one world into
another (ibid.: 35). For Coleman (1937:34) these places were where the manitous lived and
where the manitous and the medicine people could meet.
Rajnovich asserted that paintings were done principaily using red ochre because this
mineraI was "a powerful medicine itself and used in sorne cures" (ibid.). Specific images were
painted because they referred to particular experiences of the medical practitioners of the
Algonquians, the Midé, since they created the images on the birch bark scroils. The images
were painted on rock surfaces, where these medical practitioner people wanted "to leave
lessons" (ibid.: 161). Rajnovich (ibid.: 163) postulated that the paintings in this region were
"intimately" connected to Algonquian beliefs, legends, and songs. She concluded that the
best approach for those wishing to establish the meanings of these images was for
information from the Algonquian people's history and philosophy to be combined with
ethnographic data. Although Rajnovich's (1994) study has sorne flaws, it remains the most
recent and sustained contribution to research to date on the meaning of rock image sites of
the Canadian Shield.
Sorne archaeologists have used Rajnovich's approach to analyse specific rock image
sites from other parts of the Canadian Shield, but not from the Lake of the Woods
(Hamilton 2000; Pettipas 1991 b). They have explored the concept that different bodies of
ethnographic information could aid archaeologists in answering specific questions regarding
these images. It is important that the archaeologist also acknowledges the problems of using
ethnographic evidence from present day informants to interpret images from the
archaeological record. Jones (1981b: 46-47) maintained that ethnographic information from
Northern Saulteaux and the Cree groups aided the "empirical concerns of archaeologists"
seeking to establish the binder used to create the paint to make pictographs. Wheeler (1977a,
n. d.) argued that attention should be paid to studies of the structure of the Cree and Ojibwa
languages where the May May Quah Sao (often cailed Maymaygwayshi), who were spirit
helpers, were associated with stones in the sacred stories told by Algonquian-speaking
peoples (n.d.: 1). The word "stone" belonged, Wheeler asserted, "in an animate grammatical
category in Algonkian languages" that was "linguisticaily distinguished as that which interacts
with man and that which does not" (ibid.). Wheeler (n.d.: 5) recorded a rock image site on
the Semple River near Oxford House in northeastern Manitoba including the legend
connected to this site by informants. Information existed, Wheeler noted, regarding the
Page 207
relationship of these spirit helpers and rock image sites ln the ethnographie and
anthropologicalliterature such as such as that produced by Dewdney and Kidd (1962; 1967)
and Landes (1968), while others (Stevens 1971) had recorded stories that connected these
spirit helpers to rock image sites.
Two more in-depth studies of two large rock image sites must be mentioned since
they are found in the Canadian Shield, although neither site is in close proximity ta the Lake
of the Woods (Conway and Conway 1990b; Vasto kas and Vastokas 1973). Conway and
Conway were archaeologists based in the Sault Ste Marie region of Ontario and, so, this
study reflects the ethnographic and historical details of this area. Bath archaeologists asserted
that the paintings at the Agawa rock images site on the north shore of Lake Superior were
created by "Ojibwa Indian shaman artists" (ibid.: 7) who in the indigenous communities were
the link between the human world and the spirit worlds. They (ibid.) argued that Agawa was
"an exceptional site" with several components.
Conway and Conway (1990b) drew upon information from a variety of sources
including Schoolcraft's published date for the images, and from several Ojibwa elders,
including (a) Fred Pine, who Conway identified as a "noted Ojibwa tribal elder" born in 1897
and Shingwauk's great-grandson (ibid.: 19), (b) Chief Norma Fox from the Cockburn Island
band of Lake Huron (ibid.: 21), and (c) Bill Sheskekwin who lived in the Dog Lake area of
the Upper reaches of the Michipicoten River. Conway and Conway argued that their
informants enabled them to establish that several individuals had painted the images at
Agawa, including: a 17,h century Amikwa band leader called Myeengun (the Wolf) and a 19th
century shaman called Shingwauk, or 'The White Pine', from Lake Superior.
Conway and Conway argued that, because the water level of Lake Superior was
higher than the painted are as of the cliff over three thousand years ago, the paintings had to
be younger. They estimated that the most visible paintings were created during the last five
hundred years. Conway and Conway (ibid.: 7-8) posited that the images were created for
religious reasons and were the "end products of religious experiences such as vision quests,
group ceremonies, and acknowledgement of spiritual assistance," and did not directly
chronicle episodes from the pasto
The images are examined in seventeen "panels" (ibid.: 15), and as one hundred and
seventeen individual images (ibid.: 8, 15). Conway and Conway asserted (ibid.: 15) that the
panels were "numbered from north to south (left to right)" if an individuallooking at the site
Page 208
was in a canoe in front of the cliff face. They established the meanings of the pictographs of
Agawa Bay with reference to ethnographic and ethnohistoric data as weil as information
from Fred Pine, Chief Norma Fox, and Bill Sheskekwin. The images on each panel were
described and identified in considerable depth and each panel was interpreted. It is, however,
unclear whether the division of the Agawa rock image site was based on information gained
from their indigenous informants or not.
The examination of pictograph and petroglyph sites has developed and advanced
over the last hundred years. Considerable time, effort, and money have been spent on the
development of the techniques to physicaily record and date them, and many have tried to
interpret the se images. People want to find out how to discover more of them, satisfactorily
record the images, and preserve them for posterity. Yet, the most important goal is to
establish what these images may mean.
Different people have clearly taken different approaches, although it is clear that the
homological approach is more popular amongst those determined to establish the meaning
of these images. The current understanding of these sites in the Lake of the Woods is that
they were created by shamans, ritual specialists, in key places in the landscapes where the sky,
the underground, the underwater, and the earth meet and where the traveling soul of the
shaman could cross from one world to another.
Conclusion
The goal of this chapter was to present and discuss the work undertaken prior to my
own by previous researchers concerned with the pictograph and petroglyph sites in the Lake
of the Woods area. l determined to establish the most popular philosophical approach, what
previous researchers have thought of pictograph sites, how they examined what they were
interested in, and to understand the origins and the reasons behind the terminology used to
identify the images found. This critical review of the literature demonstrated that a large body
of data ranging from conservation of the paints used in pictographs and the different ways of
examining them (e.g. Myers and Taylor 1974; Wainwright 1990, 1997), how to establish the
age and style of specific pictograph and petroglyph sites in the Lake of the Woods (e.g. Reid
1979 and Steinbring 1977), the context of the physical surroundings of the images (e.g.
Noble 1968), to whether particular images only exist together and how to establish their
meaning (e.g. Cailahan 1989; Conway and Conway 1990b). Much of the literature was
Page 209
obviously concerned with the sites of the Lake of the Woods but it is important to consider
analogous studies of sites elsewhere in the Canadian Shield. Sorne of the literature is
descriptive (e.g. Lawson 1885), and sorne is concerned with technical issues, but much of it
was taken using several of the five approaches that exist in archaeology: culture-rustorical,
contextual, intuitive, analogical, and homological. As I argued in Chapter I, it is important to
initialiy use the culture-rustorical approach, then the contextual approach, and, if one is
concerned to establish the meanings of the images then either the intuitive, the analogical or
the homological approach should be used. Several archaeologists have used only the culture
historical approach (e.g. Grant 1967; Reid 1979; Steinbring, Danziger and Caliaghan 1987;
Steinbring and Caliaghan 1985). Others utilised the contextual approach (e.g. Arsenault 1997;
Pohorecky 1968; Noble 1968). Many were determined to establish the meaning of the images
they found. The homological approach is the most popular approach for archaeologists (e.g.
Dewdney and Kidd 1962; Rajnovich 1994, Vastokas and Vastokas 1973), wrule the intuitive
approach is the second most popular approach (e.g. Granzberg and Steinbring 1995; Lanoue
1989). Only one archaeologist utilised the analogical approach (Caliahan 1989).
A considerable quantity of fieldwork and research has been conducted on the
pictographs of the Lake of the Woods, and my field work in 2001 should contribute to trus
larger body of information. It was for trus reason that prior to conducting my own fieldwork
I summarised ali of the information available for each of the pictograph sites pertaining to it:
alternative names; whether photographs were taken; how was it recorded; measurements if
taken; location of field recordings and published works making reference to the site; the
name of each researcher who examined it, with the dates and rus profession; rock type;
location of the sites; whether offerings had been le ft, and whether any statements had been
made as to its possible cultural affiliations (Trus information is in Appendix 1.). Ali of this
information was required so that I knew who had conducted research, what type of analysis
they had undertaken, and at wruch pictograph sites in the Lake of the Woods.
Previous researchers have established several key pieces of information about
pictograph sites. Pictographs occur on cliff faces while petroglyphs occur on horizontal
surfaces throughout the Canadian Shield, both in close association with bodies of water.
Images in caves are relatively rare and therefore little information has been accumulated
about these sites. Previous researchers agree that the paint used was made from hematite
mixed with an organic binder wruch cannot be identified. It is also agreed that these images
Page 210
occur in places that usuaily have ail of the foilowing five physical features: rock benches
(sometimes cailed shelves), overhangs, large cracks, large crevices, and holes. Absolute dating
techniques cannot be utilised, therefore, relative dating techniques are usuaily used to
ascertain the age of these images. Sorne researchers have recorded the quantities of lichen
and rock tripe growth and encroachment, exfoliation, and the accumulation of white mineraI
deposits present at each site. This information is often compared with the orientations of
these sites. Pictograph sites on cliff faces tend to face south, east, and occasionaily west. Very
few sites face north. Previous researchers have described the images they found, and
attempted to organise systems of stylistic classification to aid in establishing local and
regional patterns in the Canadian Shield. But there is no consensus about which classification
scheme is best used to describe the images. It is acknowledged that the description of these
images is highly subjective, since the vocabulary used influences their interpretation, and each
researcher tends to use their own system. Site description has also influenced the manner in
which a site can be subdivided into parts, facilitating its description but causing problems for
subsequent archaeologists. Researchers have observed that pairs of the same image often
occur at the same site.
Ali researchers agree that it is unclear which ethnic group created these images, but
they agree that the images were created by the Algonquian-speaking peoples. Consensus
exists regarding the importance of the sites to the current local indigenous peoples, who
consider these places as important and integral to their own world view. Objects left at sites
as "offerings" have always been recorded by previous researchers. Consensus exists
regarding the polysemic nature of these images, and that they were used to convey
information to other people. Ali researchers concur that the images of the pictograph sites
are similar to those on birch bark scroils made by the Midé ritual specialists. Hence, everyone
has posited that a detailed investigation of the images on birch bark scroils, the ethnographic
record, and the pictographs might provide sorne answers as to the meanings of the images
on the rock faces.
Unfortunately, l was only able to examine the pictograph sites in 2001 because the
water level was far too high for me to survey and examine the petroglyph sites. It was equaily
important for my research to foilow the three levels outlined in Chapter 1. Therefore, l used
the culture-historical approach, then the contextual approach, and finaily the most pertinent
of ail the approaches that should be used to establish the meaning of these images. Each
Page 211
subsequent chapter de aIs with and discusses the data that l collected according to this
sequence. However, the next chapter, Chapter 4, presents the basis for making the decisions
that l made while conducting field work and how l identified and classified images that were
found. The goal of my fieldwork was to systematically collect detailed information about
rock image sites over a portion of the Canadian Shield in order to better understand the
nature of the evidence that was available. Each of the subsequent chapters will deal with the
archaeological data according to each of the different philosophical approaches that must be
applied to this data in sequence as outlined in Chapter I.
However, l do not use the intuitive approach, sin ce l think that it is important ta
articulate to other researchers how my conclusions were reached. l do not think that readers
should be reliant on the whims of the author, and they should be able to understand
precisely how an interpretation was reached. l think it is important for discussions and
arriving at conclusions to use vocabulary that is easy to understand. If an archaeologist is
determined to establish the meaning of these images in a more rigorous and persuasive
fashion, the analogical or homogical approaches should be adopted (see Table 1 in Chapter
I). Chapter V presents how the culture-historical approach is applied to this data, and seeks
to establish the total range of the images. Chapter VI presents the conclusions that are drawn
once the contextual approach is described, and considers whether combinatory rules exist
regarding the relation of the different sorts of images to one another. Chapter VII uses the
most applicable approach concerned with meanings to consider how to establish the
meaning of these images, and whether these images are related to others outside the rock
images sites.
Page 212
Chapter IV: My Work on the Rock Image Sites.
This chapter presents the ground work required to collect the data I needed on each
of these sites. It presents the basis for my decision making while conducting field work, and
how l identified and classified the images found. Through fieldwork, I systematically
collected detailed information about rock image sites over a portion of the Canadian Shield
in order to better understand the nature of the available evidence. This required not only re
recording the known sites, standardizing and verifying current knowledge, but also locating
as many unknown sites as possible. This data will enable me to achieve the four goals of my
thesis to (a) identify the possible vocabulary of images, (b) determine whether combinatory
rules exist, (c) reconstitute the life history of each site, and (d) ascertain whether the images
can be related to other images outside the rock image sites to determine if this can provide
information about the meaning(s) of the rock images.
Therefore, in 2001, l conducted three and a half months of fieldwork on the
pictograph sites in the Lake of the Woods area located between North latitudes 49° 00" - 49°
50" and West longitudes 93° 53" - 95° 09". I implemented a newly developed method of
collecting survey and field data on the pictograph sites in the hope that perhaps new
pictograph sites could be found. l based my research on my experience, gained conducting
fieldwork in the Canadian Shield region since 1990; and working on conventional
archaeological sites in the Canadian Shield and the United Kingdom since 1989. I undertook
this fieldwork with an assistant, Melinda Bell; a volunteer, Elizabeth Campbell, who also
provided valuable photographie advice and assistance; and a team of approximately four to
five professional divers, from Sunset Diving, in Kenora. The divers examined specifie
underwater cliff faces that we suspected of being completely or partially submerged sites.
They also examined the base of sites that either had large quantities of offerings or were near
sites with large quantities of offerings. l did not conduct ethnographie research or use
informants for information regarding both the name of a shape and its meaning for the
reasons discussed in the introduction of this thesis.
This type of archaeological fieldwork is unconventional for two reasons. It must be
conducted from a boat. In addition, pictograph sites cannot be excavated in the same
manner as conventional archaeological sites. Fieldwork commenced in the middle of the
eastern portion of Lake of the Woods, since l rented a boat from the private marina owned
by Tomahawk Lodge, in the township of Sioux Narrows. l chose this marina because its
Page 213
owners have excellent business relationships with other private marinas throughout the Lake.
Long distances between the sites and Tomahawk Lodge necessitated the purchase of
additional gas from these other marinas.
l rented a boat so that l could start fieldwork each day with the maximum of gas (50
litres), and a 30 horse power motor free of mechanical problems. l am accustomed to
conducting fieldwork using sorne form of boat. l knew the importance of keeping the boat
where someone had the required mechanical skills to repair a boat engine. Most of my
previous fieldwork in the Canadian Shield involved canoes as transport to run rapids and line
rivers (move upstream against the current), since we moved through river systems and lakes
in remote parts of the bush usually reached by float plane. Many fieldwork contracts required
portaging from one lake to another to get from one site to another. Carrying an engine and
sufficient gas, portaging field equipment, field gear, and foodstuff was not feasible for a small
field crew who had to travel and work over a region extending a considerable distance in the
bush. Usually, l did not cross very large expanses of open water typical of the Lake of the
Woods. It is impossible to survey the Lake of the Woods in a single field season using a
canoe. However, occasionally the water was so shallow that a motor boat could not get close
enough for a detailed examination of a few the sites so a canoe had to be used to examine
them.
l used knowledge that l had obtained regarding the edible plants readily available in
the bush; the possible number and nature of rapids, especially waterfalls; information on
local water currents; types of portages; and as many detailed maps of the region as possible.
My previous experience had occasionally involved using a radio to maintain daily contact
with the nearest remote Ontario Ministry of Natural Resources outposts. However, this time
l could use a telephone rather than a radio. This was important, since there was always the
danger of being caught in a fore st fire, something to which Northwestern Ontario is prone.
Forest fires occur without warning and are very dangerous, especially if the winds cause the
flames to spread rapidly. Fortunately, during 2001, there was little danger from fore st fires in
the Lake of the Woods. However, this region did experience a large tornado which hit
unexpectedly and caused large numbers of older white pines to be ripped from the ground.
During my 2001 field season, l had a cell phone which occasionally functioned, while
working, so that l could find out the latest marine and weather report whilst travelling by
boat. l did not use a global positioning unit because l was told by the regional archaeologist
Page 214
based in Thunder Bay that l must learn how to travel through the bush without one before l
. could learn how to use one. l did not use a guide from one of the local reserves.
Conducting fieldwork in the Lake of the Woods differed from working in smaillakes
and rivers elsewhere in the Canadian Shield. l could use highly detailed marine charts. l
needed to acquire a detailed understanding of the many different local weather patterns, and
highly variable winds. l developed the ability to relate information from marine maps to a
landscape, the configuration of which changed as the level of the Lake rose or fell. In many
cases, islets morphed into reefs, muskeg became floating islands, and shallows became bays
or inlets. l had to rely upon my ability to navigate through a landscape that looked the same
regardless of where l was, be aware of rapidly changing weather and water conditions, and
know the location of marinas, while managing one or two small boats and a crew that varied
from two to eight. l employed equipment ranging from digital cameras, anchors, ropes of
various types, paddles, boat lights, an electronic depth finder, life jackets, permits to travel by
water into the USA, measuring tapes, food supplies, fresh water, a cell phone, ail weather
gear, to pens and pencils.
We worked from nine a.m. to nine p.m. for six to seven days a week, depending on
the weather, through June, July, August and half of September. l decided not to camp using
tents since l had to undertake digital analysis of my data while in the field. l had one
assistant, and one volunteer on a tight budget and a large geographical area that had to be
covered in a short period of time. Therefore l decided to stay in a small log building with
basic facilities including full internet, hydro, and running water. This enabled me to set up a
small computer laboratory on site.
l chose the Lake of the Woods area, in Northwestern Ontario, because the rock
Image sites in this area appeared representative of the data resulting from archaeological
surveys conducted elsewhere on rock images of the Canadian Shield. l could not find one
previously known pictograph site called DjKn-2, despite numerous attempts to do so.
Nevertheless, l re-recorded twenty-two known sites, and five additional sites were discovered
for a total of twenty-seven sites.
Two of these new sites constitute a new type of rock image site, since the paintings
were found inside caves, not on cliff faces. Pictographs in caves should not be considered
unusual, or an oddity, but rare. Another pictograph site, DcJh-45, found in Confederation
Park in downtown Thunder Bay in 2001 (Bell pers. comm. 2001), exists in a cave within a
Page 215
serÏcs of granite cliffs that border a dry riverbed (Ross pers. comm. October 2001). Given
the occurrence of these three sites beside water, it is possible to argue that this type of site
follows the rule of thumb that pictograph sites occur in conjunction with water.
Only nineteen of the twenty-two known sites had Borden numbers, possibly because
previous archaeologists had neglected to ob tain Borden numbers for three sites. Dewdney
recorded all three, but in 1960 and 1967. For many years, one of the sites was called the
pictograph site on Picture Rock Point, or Dewdney's Site # 95, while the neighbouring
pictograph site in Portage Bay, also first recorded by Dewdney in 1960, was called Site # 94.
However, the third of these 'known' sites without a Borden number had been forgotten,
since Dewdney did not publish any information about it, although he found and recorded it
in 1967. l discovered trus when l read his field notes and examined rus slides at the Royal
Ontario Museum.
l initially had intended to include petroglyph sites in my survey of the Lake of the
Woods. However, with the exception of one site at Machin Point in Shoal Lake in the
northwestern portion of the Lake of the Woods, the extremely high water levels covered all
sites to a depth of approximately 75 centimetres, rende ring this goal impossible. The water
level did not drop fast enough over the period of my field season to consider recording these
Images.
Based on my reVlew of the work by previous archaeologists on pictograph sites
discussed in Chapter 3, l designed two forms, A and B, which l used in the field.
l deeided not to use the established technique of subdividing pictograph sites into
segments although it has been used by Pastershank (1989), Dewdney (Dewdney and Kidd
1962 and 1967), and Lambert (n.d). This practice subdivides sites into parts and inadvertendy
causes the images to be placed into groups in the culture-historical approach. Analysts must
complete the culture-historical approach, then the contextual approach prior to establishing
the meanings of these images.
l also discovered insurmountable difficulties in obtaining sorne desirable information.
l removed these categories from Form A: (a) prehistoric and historie setdements in vieinity
of the site, (b) present-day population with date, Cc) prehistoric and historie populations with
dates. l also omitted several categories from Form B. l removed the category "small island
visible" because of the possibility that islands occur near sites created, as a consequence of
the construction of the dams at the turn of the twentieth century construction and the
Page 216
artificiaily raised water level present in 2001. l eliminated the category "water reflection on
rock surface" because l realised that it was too subjective or dependant on factors beyond
the control of the individu al archaeologist. l cut the category cailed "rock fail visible"
because rock fails could have occurred in the previous few years and not in any time period
related to the creation of the images. l excluded a section cailed "mineral offerings" since it
was difficult to de termine whether smail rocks existed on the site naturaily or had been
purposely left there. l merged the category entitled "exfoliation" with "general notes", which
describes the physical state of the rock surface upon which the images exist. l removed the
categories "rocks present at the top of cliff', "rocks present on top of cave", rocks present
on cliff shelves/ shelf', and "rocks present in crevices" because the presence of rocks only
indicates that the surfaces of larger rocks are experiencing exfoliation and wearing away over
time.
The individual pictograph sites associated with this lake vary in geographical
concentration, located in the eastern and northern parts rather than the western and the
southern portions of the Lake of the Woods. Sorne of the southern portion are not in the
Canadian Shield. Most of the known pictographs occur at the base of vertical granitic rock
wails either immediately beside the water or several metres from the water's surface. It is
impossible to date these sites using absolute chronometric techniques. Local indigenous
groups, who daim these sites belong to them, do not ailow removal of pigment samples.
This is unfortunate because analysis of the images for their age and for the paint composition
would be valuable.
Section 1: Numbers of images
Three hundred and eighteen 1mages were recorded. The smailest number at a
pictograph site is one and the largest number is thirty-seven (DhKm-5). Three sites have only
one image: DhKm-19, DiKm-50, and DkKn-6. The average number of images at a site is
eleven. The most common shape is a 'blob,' and seventy-three different types were identified
at these twenty-seven sites. Seven other types of blobs were also connected together or
occurred attached to lines or with curves. The second most popular shape is a 'creature,'
which is an animal-like shape, and there are twenty-three of them, ail slightly different from
each other (see for example Appendix 6).
Page 217
Section 2: Defining sites
The presence of the water in and nearby the caves raises the question of how and
whether previous archaeologists recorded the water level and considered whether it may have
changed. The problem of changing water levels relates to the quandary of how to define a
'site'. Clearly, from the discussion in Chapter III, the definition of the size of a site affects
how researchers have recorded the images, taken photographs, and created written
descriptions. Nevertheless, sorne previous archaeologists also presumed that the water level
beside these sites did not experience radical, dramatic changes. They invariably recorded the
water level and noted that the water levels have fluctuated over time occasionally affecting
the recording of sites. Archaeologists concerned with these sites invariably indicated the
drainage system on which the pictograph sites is situated and the type of rock upon which
these images exist. Dramatic shifts in the water levels recorded on the sites have resulted
from large-scale alterations to the drainage system of Northwestern Ontario. The Lake of the
Woods is part of the Winnipeg River Primary Watershed Division. This watershed stretches
from the height of land near Thunder Bay, east of Rainy River into the Lake of the Woods
and runs northwards towards the Winnipeg River. Dams are located on the mouth of the
Rainy River and he ad of the Winnipeg River, at the far north end of the Lake. l discussed the
effect of fluctuating water levels caused by these dams in relation to the water levels in sorne
depth in Chapter II. Changes in the water levels of the Lake of the Woods are indicated on
the rock surfaces. The water level in the summer of 2001 [the year that fieldwork was carried
out] was approximately 75 centimetres higher than the norm. This meant that the
watermarks on the cliffs on the sites recorded in 2001 were the highest yet recorded.
Section 3: What methods were used?
(a) Examination of maps from different periods
To find new sites and to re-examine old ones, l modified the techniques traditionally
used by archaeologists. l needed to know the history of the investigation of each site, and
whether previous archaeologists had investigated such matters as whether offerings existed at
these sites. However, before recording the sites, they had to be located - a serious challenge
given that only a dot on al: 40 000 map indicated a site. Ali the while, l attempted a
preliminary analysis of the data. l quickly learnt that the most recent charts were inevitably
Page 218
'idealizations;' tourists and fishing boats who relied on them were often found marooned in
the middle of nowhere complaining about 'uncharted reefs' that had mysteriously appeared.
Furthermore, the scenery in this region generally looks the same, so one easily becomes lost
in the middle of Lake of the Woods.
Archaeologists working in the twenty-first century should realise that they use maps
made during the second part of the twentieth century. Due to changes in the water level, the
shorelines on these maps are different from those observed and recorded on maps made by
different people travelling through the Lake of the Woods sin ce La Vérendreye arrived there
to establish Fort Charles (Burpee 1927). The twenty-first century archaeologist searches for
rock images in a twenty-first century landscape but most of these images were created during
previous centuries, when the landscape was possibly physically quite different. It is possible
that islands in that landscape once formed part of the main shorelines but became islands as
a substantial rise in water level flooded low lying lands. Sites could have been fartheraway
from the water's edge. Rock images may be found in caves that are no longer immediately
beside water. Yet, archaeologists view pictograph sites that are not immediately adjacent to
water as anomalies.
The archaeologist can resolve this problem if he remembers that the water levels
have risen and the shape of that Lake have changed. Therefore, archaeologists should
examine the maps created by White and Meyer between 1913 and 1916, (1916) and make a
visual comparison between them and the modern 1:8 000, 1 :250 000, and 1 :40 000 maps of
the region. The detailed literature review of the Canadian Shield established that pictograph
sites are invariably found on pro minent rock faces, prominent outcrops, and along the shores
of islands, but not in areas of muskeg. Archaeologists traditionally believe pictographs to be
found in places that are orientated east or west, on granitic rocks, and always in association
with water. In locating new sites, 1 kept these key points in mind and examined maps of
1 :250 000, 1 :80 000 and 1 :40 000 for changes in elevation and for geologïcal information. 1
identified every part of the lake where the contour lines were close together on maps, since
these might represent steep cliff faces rather than where reefs, muskeg and swamps existed. 1
established the effect of changes in depth of the water, the nature of the bottom of the
different parts of the lake (whether it was rock or mud), and graduaI changes in water depth
and compared the shorelines of the landscape. The comparison of the different maps
enabled me to estimate how the shape of the land probably had changed when the water
Page 219
levels were raised. l needed to carefully examine modern maps to determine the relative
proximity of the elevation lines along the shore edges of each and every landform. Rising
water flooded shallow areas, submerging low-profile archaeological sites, and low-Iying rock
images sites.
Surveying the shorelines and cliff faces revealed many rock faces as potential site
locations, like those depicted in figure 4.1 below, in the area north of the Aulneau Penin sula
and south of the Barrier Islands. The physical features associated with sites invariably include
ledges, overhangs, or large cracks or crevices. Previous archaeologists clearly established the
association between these features and images.
Figure 4.1: Examples of cliff faces examined for pictographs during 2001.
Archaeologists in northern Ontario traditionally have found sites by examining large
expanses of cliff faces on rocky shores of lakes and rivers, where the rock surface faced east
or west and was immediately adjacent to water, or the rock surface plunged direcrly into
water. They also relied upon information gained from local informants. Dewdney's notes,
held by the Royal Ontario Museum, show clearly that he relied upon informants to find sites
in the Lake of the Woods and throughout the Canadian Srueld.
Two new sites were discovered near already known sites, indicating that previous
archaeologists did not always systematically examine the vicinity of known sites. DhKm-20,
in figure 4.2 below, a new site found in 2001 is located on the opposite headland from where
Dewdney recorded DhKm-18 in 1960. DiKm-49 was found in 2001 in a cave berund DiKm-
3, wruch Dewdney recorded in 1953.
Page 220
Figure 4.2: DhKm-18 (an "old" site) and DhKm-20 (a "new" site).
Attempting to find new sites in this landscape involved using the services of a team
of professional divers. The divers examined thirteen specifie cliff areas that we suspected
were completely or partial1y submerged sites. The list is as fol1ows:
1) the rock face at the base of and around DgKl-2 in Sabaskong Bay.
2) the rock face at the base of and around DgKl-1 in Sabaskong Bay
3) the rock face at the base of and around DgKl-17 in Sabaskong Bay
4) the rock face at the base of and around DgKl-19 in Sabaskong Bay
5) the base and the length of the cliff face from just south of DhKm-4, including
DhKm-l and north of this site in the southern part of Whitefish Bay.
6) the entire length of the base of the rock face at the base of the headland near the
entrance of Atikaminike Bayon the eastern side ofWhitefish Bay.
7) the base of the rock face upon which DhKm-3 exists
8) the rock face at the base of and around DiKn-l on Hayters Peninsula
9) the entire rock face south of and at the base of the cliff on which DiKm-3 exists
10) the base of the cave where DiKm-49 exists
11) the entire rock face south and north of DiI<Jn-51 and the base of the cave cal1ed
DiI<Jn-50 on Horseshoe Island
Page 221
In other cases, the divers also wanted to investigate sorne submerged cliff faces
which they thought had potential for pictographs, so I could deterrnine whether they were
sites or not. The map in figure 4.3 indicates where underwater investigations took place.
Aubeau hoin1mla Wh·.
1h'Im~
Pit:tognph Sitefi Areas investigated underwmer
Figure 4.3: A map of where the diving took place.
Several of these divers, who dove professionally in the Lake of the Woods, had
encountered sorne of these rock faces while working. Diving in this region is important. The
Lake of the Woods is used as place to train and certify divers for diving under the ice in the
win ter, since sorne areas of the Lake do not have strong currents. Sorne of the Lake is free of
ice during the winter because of fast currents, including sorne areas where pictographs are
located or are covered only by thin ice. The divers conducted this reconnaissance work under
the terrns of my archaeologicallicence issued by the governrnent of Ontario. A diver below
the boat, beside the cliff face, used an underwater camera pointed at the rock surface. While
Page 222
~ ..
viewing a TV screen of what was on the rock surface below the water level, l indicated to
another diver at the surface where l wanted the camera pointed. The te am a1so searched
underwater for artefacts, caverns, piles of rocks, or anything not geological in nature at the
base of each cliff face.
Although we checked places where we or the divers thought sites might exist, we
found none.
The divers inspected the base of the cliff face of eight known pictograph sites:
DhKm-4, DhKm-5, DiKm-3, DgKl-2, DgKl-1, DgKl-17, and DgKl-19. Offerings above
water existed at all of these sites, ranging from tobacco at DhKm-4, DgKl-1, and DiKm-3, to
clothes, smallliquor botdes and tobacco at DgKl-2 and DhKm-1. Occasionally, we noted a
striking difference between the offerings left at the site and those found below it. The next
chapter presents and discusses these offerings.
(b) Methods for collecting data
We used exacdy the same methods for recording data both at known sites and at
newly discovered sites. l devised the forms after reading numerous field reports of rock
image sites and the guidelines published for the creations of collections of architectural
drawings (Porter 1994). Pictograph and petrog1yph sites are large three-dimensional
structures in the landscape, like buildings. Therefore, studying how arcrutects collected,
recorded and archived information became a logical step. l used Form A to record
information on the physica1 structure of the rock surface and the information required
according to the terms of my archaeological licence from the Ontario Government
(Appendix 3). l devised Form B to record the images as a whole unit rather than separately
(Appendix 3). Confidentiality issues arose as a consequence of the licence requirements by
the Government of Ontario to conduct archaeological surveys or fieldwork in the Province.
l am responsib1e for keeping the particulars of the locations of the site(s) found under this
licence confidential and not allowed to publish or release to the public, in any format, the
particulars of the locations of these site(s). l must supply, however, detailed information
concerning the locations of site(s) to the Ontario Ministry of Tourism, Culture and
Recreation on Archaeological Site Record Forms or the Archaeo1ogical Update Forms. Prior
to recording each site, either l or my assistant 1eft tobacco, a common practice among
archaeologists in this region.
Page 223
Form A recorded the location and the physical context of each site. The Borden
number must be included, likewise, the names of the recording archaeologists and the date of
record. Form A contains similar information ta that required by the official Borden form
each archaeologist licensed by the Ontario government submits after the completion of
fieldwork. The foilowing information must be included for ail sites: the Borden number,
recorder's name, date, preferred site name, other names, name of township, concession,
county or district, location, access, and settlements. l developed both forms prior to
conducting fieldwork, but modified them after physicaily examining the first few sites.
If an official name does not exist, the archaeologist should try to ascertain from
inhabitants of the region if a local name exists. Two sites, DiKm-SO and DiKm-Sl, are on
the southwestern shores of an island thought by local inhabitants to resemble a horseshoe
and, therefore, cailed Horseshoe Island. However, trus name, like that of Tamarack Island to
the north of it, is not recorded on the 1 :40 000 map.
l will present the information obtained about DgKl-2, in the photograph in figure
4.4, to illustrate the information coilected in Forms A and B.
Figure 4.4: Overview of DgKl-2 taken in 2001.
DgKl-2 was examined several times. The first date indicates the first examination; the
second, the divers' expedition to determine whether offerings existed below the surface of
the water, and the trurd date indicates when W. A. Ross, the regional archaeologist and l
examined the paintings and the offerings. He wished to see for rumself the new paintings and
Page 224
the large quantity of offerings present at this site in comparison to the other sites found in
the Lake of the Woods.
Borden Number DgK1-2 Recorder M. Bell & A. Coison Date IJuly 4/01, August 5/01, August 24/01 Place Name The Three Caverns Pictograph Site
(Pastershank 1989) Alternate Site Names None Previous Site Names Dewdney's Site # 198 (August 20d 1964)
We recorded the current place name as well as prevlous place names to enable
archaeologists to track each site by name or Borden number. The Borden number was
provided for a known site. Until l obtained a Borden number from the Ontario Ministry of
Culture and Communications, l left this section blank when recording each of the new sites.
The next categories enable the archaeologist to indicate the precise geographic
location of the archaeological site in question. The spatial co-ordinates category records
longitude and latitude. After fieldwork at the end of the day, l added additional data: NTS
grid location, and site elevation.
Figure 4.5: Physicallocation of DgK1-2 in Sabaskong Bay, Lake of the Woods.
Page 225
The "geographical notes" section enables the recording of further information about
the topographie setting of the area. The archaeologist here identifies the local name, if one
exists, of the place where a site exists.
Geographie Notes It lies on a rock of land whieh is on the main route to Sabaskong Indian Reserve IR 35D on Crow Portage Bay, main channel from Splitrock Narrows to Sabaskong Bay. It is opposite island with a private eamp (no name) and near the turning to be taken to the Nestor Falls government dock.
Information about present day nearby settlements must be gathered so that the
archaeologist can if, she so wishes, consider whether sites always occur in the vicinity of
specifie types of settlements. The word 'near' implies that site is within 6 miles, or
5.21385744 nautical miles, or 9.6506404 kilometres by boat from either the nearest
settlement or reserve. The journey would only take up to 15 to 20 minutes by boat to qualify
as 'near'.
Present Day Nearby Settlements It is near the entrance to the harbour of Indian Reserve 35D and the township of Nester Falls
It is important to state whether the landmass is a peninsula, an island, a narrows, or
mainland and to record its name.
ILand Masses IPeninsula - Sabaskong Peninsula
Since it is possible that not ail the pictographs can be recorded from a boat, l
included the space "Access" to record the method of visiting these sites. DgKI-2 was one of
several sites that could be recorded partially by standing on the rock ledge immediately in
front of sorne of the images.
Access
The archaeologist must supply information concerning the regional drainage of the
water system in the geographical region in question.
Regional Drainage Water Systems (direction of Crow Portage to Sabaskong Bay to Little flow) Traverse Bay to Big Narrows to Kenora to
Winnipeg River to Nelson River System to Arctic Ocean.
1 think that the archaeologist should attempt to determine the uses of the adjacent
body of water and speculate about the period of said use in years. The uses of the body of
water can relate to the presence of fish, waterfowl, ungulates and how the land in the vicinity
of the site is currently used for recreational purposes. This commentator stated that the water
has been used for travel (communication), the transport of people and goods, recreation and
fishing (both personal and commercial). A time frame of three hundred years plus was given
for the period of use of this body of water, sin ce this covers the historie period from when
European observers first arrived in this region.
Uses of Body ofWater Communication, transportation, fishing, recreation, wild rice?
Period of use for body of water (in years) 300+
The presence of water on the number of sides must be noted at the site to allow the
archaeologist to indicate whether there are paintings on multiple sites of a rock face or only
one side.
IPresence ofWater (on # of sides) Il side
The type of body of water next to or nearby the pictograph site must also be identified (i.e.
lake, river, stream, channel or ocean) and its name given.
IBody of water surrounding/ adjacent to site ILake - Crow Portage Bay
The archaeologist should indicate whether the water freezes or not during the win ter,
whether possibilities existed that the water level fluctuated at the site, what and/or who
caused these fluctuations, when they occurred that year, and whether major changes had
occurred in the water level and for what reason.
Freeze/Partially freeze in winter Water Level Fluctuation (yes/no) Cause of water level fluctuation
Freeze Yes Dams, lncrease ln precipitation, artificial control of water levels by Lake of the Woods Water Control Board.
Page 227
Extent of Fluctuation ~inches Eer annuml Estimate of Average of75.96 cm /6"- 2' Control ofWater Levels by whom/what? Lake of the Woods Water Control Board -
dams Season ofWater Level Fluctuation Spring/ early summer. Major Changes in Regional Water Level (in Rise ln water level ln 1888 caused by years) building of Rollerway in northern part of the
Lake of the Woods -123 years ag9
It is important to indicate the number of water levels visible on the rock and their height
relative to the current level. If there is only one visible it must be the current one.
IVisible Water Levels (number, height) Il (currentl
The location of the pictograph(s) must be recorded, i.e. whether the paintings occur on cliff
faces or in caves.
Location (cliff faces / rock outcro s/ caverns/ out of water
Cliff face
The final category in this section, "general notes," allows recording of any additional
information that the archaeologist thinks should be collected.
1 General Notes INone
Form B's first three categories are the same as Form A's, since the Borden number
of the site, the name of the archaeologist and the date when this site was examined should be
noted. l devised Form B to consider the sites and the images as a whole unit rather than as
separate units. l recorded the rock face with images as single large unit rather than as
separate units to avoid smaller groupings being arbitrarily imposed by the archaeologist on
the images. l first described and recorded each rock face in terms of its physical attributes.
Then l described the images present.
On Form B the archaeologist also indicates the units used for any measurements
taken, i.e. metric or imperial. He assesses the type of rock. Information from the
archaeologist, and subsequently from geological maps, may be distinguished by names beside
the information recorded.
Page 228
Figure 4.6: Field drawing of DgK1-2 by M. Bell in 2001.
Type of rock Granite. Ice and water eroded Archean granite (Mining, G. V. and Sharpe, D. R. 1991 Surficial geology, Rat Portage Bay, Ontario: Geological Survey of Canada Map 1770A Scale 1: 100 000)
The contextual approach reqwres that additional information about rock types
gleaned from geological maps of the region must be added later. The number of cliffs or
caves and their height must be indicated.
Number of cliff(s) 1 Height of top of cliff(s) from present day 13.73 metres water level Number of visible water lines 1 fcurrent) Height of top of cliff(s) from visible water line 13.73 metres
The height of the cliff from the present day water line must be recorded. The information
for caves is different since these paintings were not created on the surface of a cliff face. The
number of caves must be counted. The distance from the roof of the cave to the present day
water level, as weil as the height of the cave must be measured. The former measurement
may seem odd for caves, but two were found in close proximity to water. High water levels
could cause the water to enter the inside of these caves. The width of the cave must also be
measured.
Page 229
The vegetation at the top of the cliff faces, on the side and/or at the surface and base
of the cliff and the cave should be recorded. The archaeologist must determine whether
vegetation over paintings hinders their visibility from the water, and whether it contributes to
the graduaI deterioration of the cliff face and images. The type of vegetation must be noted,
because it is possible that it has been removed or changed over time.
Vegetation of tOE of cliff~s) Birch, jack pine, juniper, sedges Vegetation on side 1 surface of cliff( s) Birch, white pine, bracken fem, grass, rock
tripe, lichen Vegetation on base of cliff(s) None -water
Since the presence or absence of features such as ledges, overhangs and large cracks,
crevices, and deep holes is important for finding pictographs, they should be described if
they are present, particularly if they exist close to each other as weIl as the paintings.
Rock overhangs Yes = 1 (big one - 2 metres deep) Proximity of rock overhangs to one another NIA Proximity of rock overhangs to paintings Above painting§, sorne on overhang Large cracks Yes = 3 diagonal, 1 horizontal Proximi!f of large cracks to one another Refer to photograph Proximity of large cracks to paintings Through 1 beside Large crevasses Yes = 1 large horizontal Proximity of crevasses to one another NIA Proximity of crevasses to paintings Above paintings Rock shelfl shelves Yes = 1 very big one at base of cliff Proximity of rock shelfl shelves to one another NIA Proximity of rock shelfl shelves to paintings Below paintings Rocks present at base of cliff(s) Yes and they are very big
As mentioned, l removed several categories initially devised prior to conducting
fieldwork. lnitially, archaeologists were required to record the specific orientation of the rock
surface at different parts of the cliff face. l initially amalgamated these readings into a single
category called orientation, implying that a reading be taken giving the orientation of the rock
at the location of the paintings' site. l found applying this category very difficult in the field,
even though l had devised it because several archaeologists (e.g. Pastershank 1989; Rusak
1991) had noted that sorne sites face particular compass directions.
Number of angles of painted face + location 15 (base, middle, to )
In the case of caves, the orientation of the cave opening should be recorded.
A category exists in which the archaeologist notes whether offerings were made at
the site, both under and ab ove water. These must be recorded in the same category
regardless of where these offerings occur.
Offerings Yes = I<ids' and adults' clothes (pants, skirts, t-shirts, socks, chenille sweater, Bannel clothes - sorne old, sorne new) , dried food stuff, bandages, earring (mother of pearl with porcupine quills and silver and blue beads), earring - plastic heart (never in pairs), tobacco, empty tobacco container ("Copenhagen Tobacco"), beer bottle glass, clear glass, "Prayer Sticks" ( 26 bundles of 12, 12 cm long by 12.5 - 17 cm wide, tied together with white cotton string, half of each stick painted blue and the other half painted red, large diameter sticks always tied together and small diameter sticks always tied together - never mixed, 1 bundle of 12 in the left-hand cave, refer to field sketches and digital and manual photographs).
More offerings were left on the rock bench under overhang when we came back later to record more photographs; modern paintings also visible at site. Returned August 5th 2001 with a team of divers who discovered the rocks immediately below and the vicinity around the rock art site. They discovered that it was a mud bottom and that there was a pile of rocks but no offerings were evident, only a Mountain Dew soft drink can, a man's bicycle and another cavern below the water level to the right and 2.14 metres below the surface of the water.
The methods of recording the site must indicate the techniques used.
Page 231
r-'. !
[MethodS of Recording Digital images, photographs (colour), field notes, field drawin
Place details concerning the weathering of the rock surface, exfoliation, any white mineraI
deposits, rock tripe as weil as lichen encroachment and growth, or any oddities used in the
creation of these images in a final section, "general notes."
General Notes -- ----_.-
None of the images identified by Pastershank 1989 can be identified with exception of the two handprints because 0
white mineraI deposits over the surface of the rock. However, new images exist in modern red paint most probably created using a spray cano One of the paintings-an animal shape overlies the ochre smear under the large overhang, while two of the new motifs are beside the ochre under the overhang. The fourth new image at this site differs sin ce it is an outline of an animal shape in red spray paint, which is hoilow. It was not painted using a paintbrush but a spray can like the other new images. However, it differs in location since it not above or nearby the offerings under the rock overhang but on the side of the overhang exposed to the elements and it is difficult to see. Exfoliation resent.
AIl of the field data were entered into Excel spreadsheets to keep an electronic record
of this information, and to establish quickly and easily whether patterns emerged.
(c) Image analysis of potential sites in the field
We examined potential pictograph sites to determine whether ledges, overhangs,
large cracks, crevasses, or deep holes existed. Often it was doubtful whether a site existed or
not because of the difficulty in distinguishing pale ochre from a pink coloured granite
surface. To establish reliably whether the 'redness' seen was just the colour of the rock or
ochre, l employed photographic tools. The red/orange patches on the photograph below is
not ochre but orange coloured lichen.
Page 232
r·
Figure 4.7: Red/orange lichen on the western end ofCrowrock Island.
The travel time involved and the weather forecast must be considered to determine
the best time of day to take photographs of the sites and surrounding areas. We used a Sony
Cybershot DSC-S70 digital camera to confirm whether a pictograph site existed. Each
evening, we downloaded TIFF files and conducted preliminary analyses of potential and
known pictograph sites using Photoshop 6.0. By employing the enhancement tools available
in this software, we could establish whether what we suspected, from examining the site in
the field, might be faint ochre was in reality an image or a collection of images. By
controlling the colour ranges from a part of a weIl known site caIled DjKn-l, visible in figure
4.8, where the red ochre was very strong, we could identify images that had either faded or
become obscured by the visible white mineraI washes.
Page 233
Figure 4.8: Part of DjKn-l used as the control me.
However, these enhancement tools could not detect images existing below lichens growing
on the surfaces of the cliffs themselves.
Once we had discovered that very faint images existed, we returned to the site in
question to survey and record them. We discovered that in-depth photography using manual
cameras demonstrated how the different images related to each other and to the surface on
which they lay. The digital camera was less valuable in this aspect of the work, since the size
of the file and relatively small quantity of memory, together with the distance between sites,
meant that using manual cameras was easier. Furthermore, despite aIl the advances in digital
technologies, it was still difficult to fully control the lenses and the aperture. The camera
settings provided by the manufacturer dictated the extent that one could collect raw data in a
digital format. Digital cameras must develop to broaden the use of lenses.
Image files obtained from the digital camera were recorded as TIFF files and burned
onto CD-ROMs each evening. The images were recorded as Tagged Image File Format,
which provides the purest form of image currendy available. This format allows the use of
the new techniques developed by the Intelligence, Agents and Multimedia Group directed by
Professor Wendy Hall, at Southampton University, who advised me on this aspect of my
research. TIFF files are compressible, retrievable, and stable. JPEG image fùes, in
comparison, lose data when compressed. TIFF fùes can be used in a non-proprietary format
with the viewing programme not being controlled by a single vendor. The final advantage is
Page 234
.~.
that this format is extensible; hence it can evolve over time to meet the changing needs of
different users and does not require new software for viewing files. The Kodachrome slides
produced with the manual cameras were scanned, saved as TIFF files and burned into CD
ROMs. Two copies were made of each CD-ROM.
Section 4: The method used for describing shapes and how does one consider time
Time, the age and the shape of the images are interrelated because the physical
outline of an image changes as it gets older. An image should not be considered static with
regards to its physical shape since its physical outline changes over time. Time must always
be considered since neither the rock nor the images are static entities. However, none of
these images can be dated using absolute chronometrie techniques.
Unfortunately exfoliation, fading of the images, deposition of white mineraI deposits,
and lichen and rock tripe growth and encroachment affect the images. Each must be
discussed.
(a) Exfoliation.
Exfoliation of the rock surface occurs because the rock surfaces upon which the
images exist are affected by more than the fluctuating water level of the Lake as discussed in
chapter II. Rusak (1992a: 25) strongly asserted that pictograph sites in White Otter Lake, a
lake east of the Lake of the Woods, were "in a po or state of preservation" since in sorne
places the rock had exfoliated "to a depth of several milimeters". She concluded that this
occurred when dams to the north- and southwest artificial1y raised the water level. The
Shevlin-Clarke logging company constructed these dams in 1912 to aid the transport of logs
down Turtle River from White Otter Lake. However, extreme temperatures also contribute.
Exfoliation occurs when granite starts a hydration process in which the rock breaks, splits,
and peels as a result of expansion in salt crystals and/or water present in smal1 cracks in the
rock surface.
Mean temperatures range between 12.5°C to 14°C in summer, and -17°C to -12°C in
winter. This region experiences a maximum period of between forty and sixty frost-free days.
Snow accumulation in the Boreal Forest is higher than in the neighbouring Great Lakes-St.
Lawrence Forest. Snow coyer can last for six months.
Page 235
Taylor, Meyers and Wainwright (1975: 88) argued that freeze-thaw, often called frost
action, caused exfoliation of the rock surface at sites. Essentially, the outer layer of the rock
expands and contracts with the dramatic changes in tempe rature s, causing small cracks.
Moisture accumulates in these cracks. Over time, this moisture freezes, expands, and thaws
causing the rock surface to become weaker and break away from the main body.
However, the question exists whether pictograph sites which have paintings near the
water level would also be affected by the movement of ice during the winter and spring
breakup. Two views exist regarding this question. Sorne argue that the growth rate and
stability of ice is different depending on where the ice is situated in the lake. Bays would have
more stable ice sheets than channels because the water in a channel has a high velocity
(Layman pers comm. October 2002). Water flows faster and more efficiently in channels
than in bays; hence bays provide a "more conducive environment" for the growth of an ice
sheet (ibid.). Those pictographs existing on cliff faces in protected bays are more likely to be
affected by the movement of ice sheets than those sites that exist on rock surfaces beside
channels. This is because the ice will shift more times than in the bay and take a longer
period of time to increase in volume. The second opinion is that ice in bays and other
current-free areas can bec orne ice sheets but during break-up and freeze-up it does not shift
much unless it is broken. This is because the ice just sits there and either just builds or melts.
The rare exception to this rule could occur at break-up during a strong wind if bays were
relatively sheltered from all but one direction. Therefore, bays that had cliffs that faced the
prevailing wind, which is northwest, but were otherwise relatively sheltered would be free of
ice. In areas where a strong current exists then ice is swept through the channel, piles up, and
becomes huge pack jams that rub and grind against each other and the shoreline. Free
flowing ice floes still moves along the shore.
(b) White mineraI deposits
Groundwater leaches mineraIs identified as potassium, calcium, sodium, silicon, and
aluminium from the soil and cracks in the interior of the rock. Sorne of this water flows over
the paintings during the spring thaw or after heavy rainfall (Taylor, Meyers and Wainwright
1975: 88). Sorne of the groundwater evaporates as it runs down the rock face leaving an
"insoluble minerai deposit 'skin' on the rock surface" (ibid.). Taylor, Meyer and Wainwright
maintained that, initially, this deposit is invisible but as it repeatedly accumulates it becomes
Page 236
white. No one has established the deposition rate at which these white mineraI deposits
accumulate.
Taylor, Meyer and Wainwright argued that the creator of images couid essentially
place images on these white mineraI deposits rather than directly on the rock surface. To
establish the structure of the pigments on the sites in question, they examined several
samples using a scanning electronic microscopy. Chapter III includes a discussion of this
work. Wainwright and Taylor (1977: 31) also used scanning electron microscopy to establish
the structure of the pigment of two samples from two sites. Analysis of the cross-sections of
the two samples revealed distinct, almost parailel layers (ibid.). Wainwright and Taylor
posited that a chemical or a physical process separated the original pigment into distinct
Iayers. They concluded that the pigment must have migrated within the deposit because the
the samples had been taken from images that appeared not to have been repainted.
(c) Lichen and rock tripe growth and encroachment
Taylor, Meyers and Wainwright (1975: 89 asserted that lichen and algae probably
increased the rate of decay of rock images. Lichens do not con tain chlorophyll as most plants
do. Instead they consist of symbiotic fungal and algal cells (Blackmore and Tootill 1984:
209). They live in areas where other plants cannot survive and grow very slowly.
Several species of lichen grow on the surfaces of rock in this region, including
Xanthoria elegans, or rock orange lichen Oohnson, Kershaw, MacKinnon and Pojar 1995: 338).
It appears in places where there is a vast quantity of calcium and nitrogen. Although often
found in conjunction with cliff-dwelling birds' nests (ibid.), 1 did not find any birds' nests
upon examining the pictograph sites where this lichen grew. Xanthoria elegans grows rapidly,
but its growth rate apparently depends on its habitat. It has been used to date surfaces.
Two types of rock tripe, a type of leaf lichen, were found. Umbilicaria vellea, or
Forested rock tripe (ibid.: 351), grew at the majority of sites. Umbilicaria muhlenbergzï, or Plated
rock tripe, also grew at sorne sites. Umbilicaria vellea occurs in conjunction with sorne
moisture seepage. Both occur on acidic rocks in shaded or open places. The
Ojibwa/ Chippewa used both plants as a soup for sick people or made a syrup from them
which was eaten to expel tape worms.
The average lichen arguably grows only 1mm a year (Blackmore and Tootill 1984:
209). McCarthy (1999) critically evaluated ail of the models used in lichenometry. He defined
Page 237
.~.
lichenometry as the technique which utilises measurements of the thalus diameter to generate
the minimum estimate for the amount of time that "colonized surfaces [of rock] have been
exposed and stable" (ibid.: 379). McCarthy critically examined the biological basis for several
of the assumptions that have been used to interpret lichen-size age data. He (ibid.: 384-385)
concluded that the majority of the lichenographic ages could not be verified. They had not
been produced by methodologies that used widely accepted biological principles. He argued
the importance of considering the species, habitat, density, and space availability as well as
how growth had been affected by other events.
Section 5: Assessing age
Once these images are created and a site exists, it is highlighted as a locale of
importance in both the physical and mentallandscape of regional residents. The moment in
which these images were created is important but very difficult to determine. 'Conventional'
artefacts, once discovered during an excavation, are argued to represent fragments of time at
which specific events occurred. Images, once placed upon the surface of a rock face, become
integrated into the region's time sequence. They do not just represent one instance of time.
Since they can be revisited, reused and repainted, they become part of the mental and
physical record of the geographic region in which they exist. Images often are superimposed
or manipulated, especially if a known group of images is used as mnemonics or vocabulary.
Audiences can use and understand older images in conjunction with those created by
subsequent image creators. Moreover, audiences of these images change and can react
differently to both the images and their locations. The deposition of an item, called an
'offering', acts as evidence of such a reaction. Interaction between images and the people
who subsequently view them is difficult to mark, or delineate, in terms of specific known and
identified time events. The presence or absence of offerings at a site acts as a good indicator
of whether the images present are still being used; of whether they are still part of the mental
landscape of a current cultural group. Subsequent creators of newer images might also draw
upon meanings implied by placing new images on older images for new audiences.
These rock image sites experience seasonal climatic changes, as well as fluctuations in
water levels and temperatures. The sites and their images continually undergo changes that,
though difficult to evaluate on a daily basis, become evident over long periods of time. They
Page 238
change in size, and perhaps disappear as a consequence of rock exfoliation, or mineraI
deposition and/ or plant growth over the images.
Section 6: Description of a shape
It is important that the archaeologist should describe, not interpret, the evidence
when recording a rock image site. This operation involves thinking very carefully about how
one describes images, since any description will affect subsequent analysis. She must not
impose any interpretation based upon physical shape. Anyone concerned with the
methodological and theoretical issues relating to the study of rock image sites must describe
the shape of an image objectively, and in detail. Images can be organised and discussed by
their shape. A rock image must be considered as a shape to reduce the possibility of investing
the description with meaning during the recording stage. It is vital to understand the physical
outline, form, and structure of an image before discussing its context or meaning.
Considerable care must be employed when choosing the vocabulary to describe the shape or
form of an image in order to avoid inadvertently or prematurely ascribing meaning(s). A
shape's form should be referenced and nothing more. Meaning must be ascribed at a
different and later level of analysis.
The precise size of an image should not be considered. It is difficult to establish
whether size was an important issue from a detailed examination of the ethnographie record
of the indigenous peoples. The current size of an image may depend upon uncontrollable
elements, su ch as the deposition of mineraIs upon the rock surface, exfoliation, and plant
growth. Nevertheless, the archaeologist should note whether a shape is large or small after
describing it with respect to the other images at the site. Hence, stating that a small, long,
horizontalline or short, vertical rectangle existed is possible.
The arbitrary division of the images painted upon the surface of a rock into parts
must be avoided. Sites, regardless of whether previous archaeologists originally described
them in several parts, were described again, since the archaeologist must only describe, not
interpret, the data.
Ail the images were described from the left- to right-hand side of the site, using ail
the digital and manual photographs that l had of each site. Since describing the images from
left to right prevents the sites from being subdivided into sections, terms such as face and
panel become redundant. If a site is physically broken into sections because images spread
Page 239
across related or nearby rock surfaces (which occurs at DhKm-18, DhKm-4, DgKm-2,
DgI<l-2, DhKn-1, DjKn-1, and DhKm-1), the archaeologist must not rein force these
divisions by describing the images in groups bounded by their physicallocation. It is possible
that these subdivisions were not important to the creator of the images.
While describing the shapes, the notion of continuous time also must be considered.
It is important to note uncontrollable elements that affect the size of images over time:
exfoliation, lichen or rock tripe encroachment, and white mineraI deposits.
1 had to consider carefully the vocabulary that 1 used. 1 drew upon the work of ]aritz,
an art historian interested in the every-day life and the material culture of the general
populace, and not the aristocracy, of the Middle Ages to help me think about how to
describe the images. His research data consisted of medieval paintings of every-day life,
forcing him to undertake highly detailed descriptions of complex images. Standards for
describing an image do not exist, as ] aritz and Schuh (1992: 144-145) observed. Despite the
fact that many proposals exist, none of these standards is widely accepted by the academic
community who describe and deal with images as potential sources of information. They
offered five reasons for the lack of standards:
1. formaI language could not easily discuss the specifies of an image,
11. different interests and consequently different approaches existed for different art
historians,
111. there are considerable gaps in "certain areas of terminological research" (ibid.:
145),
IV. there exists a "fear of losing flexibility and openness by using a system of strict
standards", and
v. the different techniques used by academics who need to analyse varying amounts
of pictures for their research (ibid.).
]aritz and Schuh offer suggestions for creating standards which 1 think are reasonable, given
that the images they considered were European, while the ones 1 was describing belonged to
a different cultural tradition.
Therefore, 1 decided to use a vocabulary that implied as little meaning as possible,
and that described only the physical outline of an image. My goal attempted not to permit
any supposition concerning poorly preserved or ambiguous figures. A technique must be
Page 240
/" ..
used to de scribe these images, even ones that are terribly faint. When describing a shape, one
should consider its present form and nothing more. Smears, blobs, and spots exist and must
be treated as described.
It is impossible to describe a stick figure without reference to terms such as arms,
legs, horns, ears, head, and body. The key point is that a stick figure should not be called an
anthropomorph, or a hum an or spirit figure. These words could be misconstrued at a later
date. The image could be interpreted as the representation of a human being, when, in reality,
its painter intended it to represent something other than a human.
l examined the TIFF flle of each site using Adobe Photoshop 6.0 while the images of
each site were being described. This enabled me to examine areas of rock surface that had
white mineraI deposits perhaps coating over the red images. Occasionally, this revealed
hitherto undetected "new" images at long-known sites such as DiKp-1, as shown in the
photograph below in figure 4.9.
Figure 4.9: DiKp-1 in June 2001 annotated by Cols on in 2002 to indicate locations of previously unknown images.
This photograph has not been altered. Red images clearly exist above the area where images
described by Ashdown (1973), Lawson (1885) Dewdney (1959), Reid (1976) and Molyneaux
(1975) were traditionally known to exist.
Presupposition must be avoided concerning poorly preserved or ambiguous figures.
Finding the precise edges of each image was not easy. Zooming into the TIFF flle of each
site and using the various tools available in Adobe Photoshop 6.0 such as hue/saturation,
Page 241
~ ..
desaturate, brightnessj contrast and colour balance tools aided my search (see Appendix 2 for
examples of what each tool does to a selection of the images from DiKp-1). However, l also
empIoyed VIPS, an image processing package4. This software highlighted the problems of,
and assisted in describing poorly preserved or ambiguous figures. It also highlighted several
features that aided their written description, as l will discuss below.
Section 7: The tools used to aid the analysis of the data
Several computing tools were crucial in undertaking data analysis. KLEIO lAS, a
database management system, Adobe Photoshop 6.0, and Vasari Image Processing, both for
image manipulations, served as the main software tools. l used the database tool to link
information about rock image sites with photographs of the sites. The image manipulation
tools aided my search for patterns in images and groups of images.
(a) KLEIO lAS
KLEIO lAS enabled me to connect my written descriptions of each image with its
electronic record.5 Originaily designed for medieval specialists, the software had a Latin
command language, with documentation in German. An English-Ianguage version became
available in 1993; l used the latter version. The software allows the researcher, usuaily an
historian, to enter information in a structure that reproduces that of the original document.
This preserves valuable information about the document, while ailowing the selection of a
method of organizing the data after the data-gathering process. It is possible to minimise
coding or mark-up on the digital record of a source prior to analysis, preserving features of a
document that may sustain conflicting interpretations.6
KLEIO lAS can manage different types of documents, induding image files. The
programme stores ail records in a manner that preserves context among records, but these
can be defined as logicaily equivalent for processing (1993: xiv) (see Appendix 4). The most
abstract group of sources must be at the top of KLEIO IAS's hierarchical structure, and the
most concrete group of information at the bottom.
4 More than a simple set of routines it is an image processing framework which is used in student projects and PhD research at Southampton University (http://www.vips.ecs.soton.ac.uk/ 26th November 2004).
5 The image ftles were produced in the TIFF format, a non-proprietary ftle format compatible with most image processing software. 6 KLEIO usually runs under UNIX, not Windows.
Page 242
Three levels of information exist: documents, groups, and elements. Documents are
the highest level of information and can be organised into groups that are abstract groupings
of information in the document. The term, 'documents,' corresponds to the type of
information being described. The historian designer used this term to imply that, at this level,
information be provided as to the document type examined, such as census records, parish
records, or paintings. Documents consist of groups which correspond to rows or records in
other traditional database systems. Each group in KLEIO lAS has a specified relationship
with others in the database specified by the user. This relationship can be quite complex,
since several groups may be dependent upon a higher group, and the same group name may
be used elsewhere in the database. Groups are related to each other in a hierarchical fashion.
Each group of information contains elements pertinent to that group. Elements are the
smallest unit of information that KLEIO lAS can access. lt is possible for these groups to
contain identically named elements, allowing these related elements to be 'implicitly joined' if
the system suspects that the user might bene fit in considering both elements as of the same
type (ibid.: xv). Elements can consist of any number of specified entries, or any type of
specified information.
KLEIO lAS offers a collection of basic algorithms to deal with data. This enables
assumptions about the data to be included in the database independently of the data itself.
Thaller (in Woollard and Denley 1993: xvi) argues that these algorithms constitute a form of
expert system that makes use of context-sensitive data. This system provides a detailed set of
basic database operations, such as data retrieval and report generation. The researcher can
move easily, therefore, through highly complex structures, and obtain results to complex and
unwieldy queries. He may even join different databases.
The file that specifies the structure of my data defines eight databases, of which l
have needed to use only four (see the file in Appendix 4). Each database contains a specific
type of data, and each is subdivided into several parts. The data file con tains all the site
information collected in the field, including information gathered in forms A and B, and the
textual description of each image from each site. The image flle contains the images
themselves. This flle also references the textual descriptions in the data file. Hence the user
can query the database either from the textual records or from the images.
l wrote and stored all of the image descriptions in KLEIO lAS. l also stored the
information collected as to the contextual details about the sites and their physical features in
Page 243
.~ ...
KLIEO lAS. The database contains all the information gathered in the two forms, A and B,
and the detailed descriptions of each image that existed at known pictograph sites in the Lake
of the Woods.
(b) Image analysis using Adobe Photos hop 6.0 and VIPS
1 conducted a preliminary analysis of the image files while collecting site data in 2001.
The analysis, done with Photoshop 6.0, revealed that a limited range of motifs appeared in
the rock image sites. It also increasingly provided sorne insight into the way in which human
beings perceived a physical shape. Close examination of each image revealed that perception
of its precise physical shape could depend upon the viewer rather than upon the presence of
red ochre. 1 confirmed this by using a second image processor, Vasari Image Processing
Software [VIPS],7 which is used by a number of world-class museums and galleries. The
National Gallery in London England uses it for most of its imaging research.8 A large
number of research projects analysing large images also have employed VIPS (Martinez, et.
al. 2002, Martinez 1991). 1 loaded the image files collected during my 2001 fieldwork season
onto a server to process them using this software.
Each shape, each graphic shape in red ochre, in each TIFF file, was extracted with a
'slider' tool built by Dr. Kirk Martinez. Image processing software assumes that the image
photographed was on a flat surface perpendicular to the camera, and not upon a slanted or a
curved surface. The slider allows for sorne control of images that do not meet these
requirements. It permits me to process each graphic shape independently. In fact, none of
the images at each site occurs on flat surfaces truly perpendicular to the camera. For this
reason, the colour saturation of each image is uneven. Preliminary image analysis conducted
in the field had already indicated this. It became clear that the different angles of each rock
face had an effect on the quality of the image obtained.
On the image of a given site, graphic shapes may rest on surfaces that are at different
angles to one another. If all the graphic shapes are processed at the same time, using the
same settings, results differ for each of the visible red shapes on a given image. The slider
enables me to change and alter the hue saturation levels according to each individual ochre
shape rather than for a group of shapes. The slider also assists in writing descriptions of
7 The VIPS software was developed by Dr John Cupitt and Dr Joe Padfield (The National Gallery, London) and Dr Kirk Martinez University of Southampton (http://www.vips.ecs.soton.ac.uk). 8 VIPS runs under UNIX.
Page 244
graphic shapes. However, l find turning the graphic shapes into black and white images the
most useful technique for the description of the shapes themselves. The software reveals that
shapes often have indistinct edges. It highlights the role that the human eye and brain play in
the process of recognizing images.
Conclusion:
Although these software tools were complex and technically challenging, they
enabled me to examine systematically the textual and visual data that l had collected, and to
gather the maximum amount of information possible about these pictograph sites.
Section 8: Vocabulary used to describe each type of image found
l attempted to use a nested classification for describing these images. The images
were described from the perspective of an individual standing or sitting in a canoe
immediately in front of the site and the image in question. The angle or the slant of a line
was not obtained because it was not in reality straight once it was examined using VIPS.
Furthermore the slant would also depend on the angle of the photograph taken. All the
photographs were taken using the water level as a horizontal baseline.
The question obviously arises as to how l would describe a shape such as the following:
+ x l described this shape as two short lines, one horizontal, and one vertical, crossing at a ninety
degree angle. It is not necessarily, or even probably, a 'cross,' as it is frequendy designated in
the literature of the rock images of the Canadian Shield. Since the creator of this image was
not necessarily someone who held the J udeo-Christian concept of a cross, no good reason
exists for it to be labelled as a cross. Interpretation is embedded in that word. If it is
diagonally depicted it should be still described as two short lines crossing each other at a
ninety degree angle, or, possibly, diagonally.
Page 245
l designated this shape as a series of horizontal short lines one above the
other with two vertical lines at either end of the horizontal lines. It should not be called a
"ladder", a "Midewiwin ladder" or a "lattice" and the short horizontal lines should not be
termed "rungs".
l described this shape as a series of short horizontallines one above the other.
Neither should it be labelled as a "ladder", a "Midewiwin ladder" " or a "lattice" and the
short horizontal lines should not be termed "rungs". If the short lines do not occur in close
proximity to each other they should not be considered as a unit.
Both of these images should be called diagonal lines
and no comment should be made regarding the angle. Any comments made as to the angle
may be linked to interpretation.
Page 246
l term both of these shapes triangles even if they
occur in conjunction with other images such as stick figures where there is a temptation to
refer to them as 'hats', especially if they are on the 'heads' of stick figures. The commentator
must record whether the shape is hollow or not.
s t , n • S .3 !~
Ali these shapes are of creatures. Images that have vertical rather than horizontal bodies
should also be recorded as creatures. If the creature has short lines sticking upwards on the
head they should be recorded as ears, horns or antlers. The commentator must record
whether the shape is hollow or not.
A shape loosely categorised as 'bird' differs from that loosely categorised as 'turtle' or
'creature' because of its different physical outline. The curved, downward pointing, usually
horizontal, perhaps vertical or even diagonal lines emanating from either side of the central
body of the shape distinguish that loosely categorised as 'bird' from 'creature'. The lines
emanating from either side of the shape's central body should be termed "a wing." At one
end of the body shape, a small triangle should exist; one of the triangle's points labelled
'head,' another, 'beak'. Occasionally attached at the opposite end of the shape to which the
small triangle is attached, one may find either a larger triangle or rectangle, or two diagonal
lines. This larger attribute should be called the "tail."
Shapes loosely categorised as creatures, in contrast, have a maximum of four short
verticallines from only one side of a central part of the shape, or no lines at aIl. This central
shape will probably have a smaller round, or oval shape, referenced as a 'head,' attached at
one end. A short line, termed as a 'tail,' may be attached to the other end of this central body.
Page 247
Images such as these must be
called birds. The commentator should note whether the image is hollow and the direction in
which the beak on the head is pointed. The number of short verticallines on the underside
of the curved horizontalline must be counted integral to this shape. If the centre of the bird
is hollow, the commentator should state whether anything is represented in the hollow
interior of the shape.
Images such as these must be called turtles.
Turtles' usually have four short lines, called 'legs,' emanating in four directions from the
outside edge of the shape. Extended into the shape's body, these lines would, in theory, cross
in its centre. The head and the tail of the 'turtle' exist at opposite ends of the shape, each
between two 'legs'. The commentator should note whether the shape is hollow or solid, and
the direction the head points. If the centre of the turtle is hollow, the commentator should
note any representation within.
If a shape resembles any of the four shapes above, the
commentator should record this shape as a crescent. State whether it is vertical or horizontal,
and whether its points face left or right if it is vertical. If the crescent is horizontal, the
commentator must additionally comment whether the points face downward or upward.
Page 248
One should not describe this shape as a boat or a canoe, as it often is in the Hterature of the
Canadian Shield.
, JI J This shape must be described as a combination as a group of
vertical or horizontal Hnes, and noting the number of Hnes. A recording would read as
foliows: "group of four or five vertical or horizontal Hnes." One should not use the
frequently used term 'taliy marks.'
ci _,#$ tr
Loosely categorise this shape as 'parallel lines.' Note
whether this shape appears vertical, diagonal or horizontal. Note whether the Hnes touch
each other at ali (as in ex ample at right of above figure). Do not use the frequently used term
'taliy marks.'
m Lb
Cali this shape a line and state whether it is horizontal or
vertical.
(
Page 249
l call this shape a right hand print if the thumb is on the left, and a left if the thumb is on the
right hand side. However, the thumb may have faded or never been represented, in which
case, use 'hand print.' It must be recorded whether the shape is hollow or not. The person
using this term must realise that using the term hand print does not necessarily imply that the
image painters painted a hand with ochre and pressed it on the rock surface.
These shapes are examples of
different types of stick figures. Do not use the terms: anthropomorph, human being,
stickman or stick man, since these imply more specific meanings of some type and/or
impute gender. Difficulty in describing these shapes arises, since the words: leg, arm, head,
body, and, possibly, horns, immediately imply assumptions about the different parts of the
stick figure to the reader. The option exists to describe them as "a collection of four short
lines that are diagonal, vertical, and horizontal with a spot at one end." However, this would
make searching for this shape very difficult when and if recovering it from a large database.
Therefore, l decided to use the terms leg, arm, head, body, and, possibly, horns, or ears.
Comments must be made regarding the presence or absence of legs, arms, etc., and whether
they are depicted upward or downward. It is important to also note if the body or the head is
hollow or not. If the head has short vertical lines on the outside then these should be
recorded as either horns or ears, and a note should be taken if short lines are attached to the
end of each arm.
This shape can only be described as a star. The number of points should be
counted and note should be made of whether it is hollow or not.
Page 250
o l called this shape a square. One should not comment about the precise
angle of any of its corners. Again, one should record whether the shape is hollow or not.
This shape must be called a
zig-zag and note whether it occurs diagonally, verticaIly, or horizontally. It must not be
designated as a 'power line' or by some other word with a similar meaning. Such problematic
terminology, although frequently found in the rock image literature of the Canadian Shield,
meshes interpretation with the description of the images in question. Occasionally, a round
filled in circle is attached to one end of the zig-zag. One or two short curved lines attached
to the outside edge of an solid circle, or 'he ad,' could be called horns.
l called these shapes rectangles. No comment
should be made about the preClse angle at any corner. The commentator must record
whether the shape is hollow or not as weil as whether it is vertical or horizontal.
Page 251
This shape is a smear and could be called an ochre
wash on the surface of a rock. Smears come in various sizes and shapes, such as oval,
oblong, and round. These shapes cannot be placed in any category described above. They
could be, and are, larger than blobs.
-• .. These shapes are examples of blobs. They usually exist if the
ochre has either faded, exfoliated off the surface of the rock face, or has become
indistinguishable because of the accumulation of white mineraI deposits. They vary in size,
since they exist once a shape no longer has distinguishable edges that can be clearly
delimited, identified and described. Blobs vary in shape, appearing round, oval, oblong, small,
large, horizontal, vertical and diagonal. Dots are small blobs. The key points are that they
cannot be placed in any of the above categories, and they are always smaller than smears.
Section 9: Possible relationships between rock image sites and other archaeologïcal
sites
Previous archaeologists concerned with these pictograph sites have not considered
the possibility that relationships may exist between the pictographs and any neighbouring
archaeological sites. A large number of archaeological sites from a specific archaeological or
historic time period possibly may exist in the immediate vicinity of a pictograph site or
groups of pictograph sites. This would be an interesting issue to consider and comment
upon. However, this is clearly impossible to undertake using the map below, which
represents the current state of archaeological knowledge.
Page 252
Figure 4.10: Map indicating location of other known archaeological sites to pictograph sites.
Some of the Borden quadrants in the map above are blank, indicating that sites either
have not been found, or that the Ontario Govemment does not have this information.
Assertions as to possible relations between conventional archaeological sites and pictograph
sites can be made only after archaeologists have established and examined in detail ail of the
sites in this region.
Conclusion:
A careful examination of the twenty-seven sites of the Lake of the Woods, using
Forms A and B, and technical software, provided me with a vast quantity of detailed
information concerning the context of each site, the physical features of the rock surface, the
large quantities of offerings both below and above the water line, and detailed descriptions of
the images. This information provided the basis to start a thorough examination of the
pictograph site images in the Lake of the Woods. l could now employ first, the culture
historical approach in Chapter V, secondly, the contextual approach in Chapter VI, and,
finaily, one of the approaches concemed with meaning in Chapter VII. Through employing
each approach in sequence, l may achieve the four goals of my the sis to Ca) identify the
Page 253
possible vocabulary of images, (b) determine whether combinatory rules exist, (c)
reconstitute the life history of each site, and (d) ascertain whether the images can be related
to other images outside rock image sites ta determine if this can provide information about
the meaning(s) of the rock images.
Page 254
Chapter V: Culture-History Approach:
This chapter is concerned with the culture-historical approach, which permits
analysis of the form, date, and location of the images, and supplies archaeologists with the
basic data that they need to perform higher-Ievel operations. They must concentrate on
assembling detailed information about each individual site, and consider each image as a
separate entity at this level. Seven sections exist in this chapter.
Deterioration over time affects the shape, existence, and groupings of images. The
first section considers the clarity and the blurring of images caused by either, or a
combination of, exfoliation, lichen and rock tripe encroachment, and the deposition of white
mineraI deposits. The second section is concerned with the distribution of the images
throughout the Lake of the Woods. It establishes the minimum number of images per site,
the large st number of images per site, and the smallest number of images per site. The third
section considers the rock image style of the images of the Lake of the Woods in comparison
to how previous archaeologists have considered style in Northern Ontario. The fourth
section examines the ethnographic information regarding the components of the paint, and
how it was applied to the surface of the rock. Unfortunately, information does not exist as to
the type of implements and techniques used to create images. The fifth section examines the
ethnographic literature of this region regarding the large quantities of gifts such as tobacco
which were often given to supernatural beings to obtain supernatural bene fit, especially in a
ritual context. The sixth section of this chapter discusses the wide range of objects, called
offerings, above and below the water level. This is because several sites had a very large
number of offerings, including tobacco, clothing, liquor bottles, and coins. This was quite
unexpected based on the previous literature from this region and elsewhere in the Canadian
Shield. The seventh section of this chapter considers how it might be possible to date these
images since direct dating techniques cannot be used.
Section 1: Blurring loverlay discussion and darity of images.
A detailed examination of the rock image sites in the Lake of the Woods reveals that
many of the images suffer from exfoliation of the rock's surface, rock tripe and lichen
encroachment, or the accumulation of white mineraI deposits (please see Appendix 5).
Page 255
I~
Exfoliation is not evident at DhKm-20, DhKo-l, DiKm-49, DiKm-4, DjKr-23, and
DhKm-19. It is difficult to establish whether exfoliation has taken place at DgKm-l, DhKm-
5, and DiKq-10. Exfoliation of the rock's surface may have occurred at the microscope level
and be invisible to the naked eye. Exfoliation is present at several sites: DiKm-50, DgKI-l,
DiKp-l, DiKm-51, and DjKn-l. Severe exfoliation has caused many images to be unclear at
Figure 5.7 above, depicting part of DiKn-1, further il1ustrates how lichen and rock
tripe often obscure images at pictograph sites. Rock tripe appears with lichen at five sites
including: DiKm-3, DiKn-1, DgKl-19, DhKm-18, and DjKn-1 (xq722 and xq723).
White mineraI deposits are a common occurrence in relation to sorne of the images
at several of the pictograph sites: DgKl-1, DgKl-2, DgKm-1, DhKm-1, DhKm-3, DhKo-1,
DhKm-20, DiKm-3, DiKm-51, DiKp-1, DiKq-10, DjKr-23, and DhKm-5 (xq725 and
xq735). The accumulation of white deposits was so severe at DiKq-10, seen in the two
photographs below, that it was almost impossible to see and identify the images.
Page 260
~ ..
Figure 5.8: Severe deposition ofwhite mineraI deposits at DiKq-10 in 2001.
This locale, at Picture Rock Point, was designated as a site by Dewdney in 1960, who
cailed it Site # 95 (Dewdney and Kidd: 44). This site is on either side of the corner of the
block of granite in figure 5.8 above. It is the best ex ample in the Lake of the Woods where
white mineraI deposits have severely affected the images, since it is very hard to see the rock
images without using VIPS/ip and Adobe Photoshop 6.0. Examination of other sites that
have white mineraI deposits demonstrates that these deposits are often unequaily distributed
across cliff faces. Of interest concerning this site is that very little rock tripe growth and
lichen encroachment affects the images in comparison to the surface of the rock that lacks
these red images. Occasionaily paintings are superimposed on white mineral deposits
occurring over earlier images, as observed at DhKm-5 and DhKm-3.
Several images at DhKm-3 in the photographs, figure 5.9, below suffer from the
deposition of white mineraI deposits. Others suffer from exfoliation as weil as rock tripe and
lichen encroachment.
Page 261
Page 262
Figure 5.9: AIl of the images at DhKm-3 in 2001.
White mineraI deposits exist both under and over some of the red images ln the three
photographs of DhKm-3 presented above in figure 5.9.
Close examination of photographs reveals that the accumulation of white mineraI
deposits has affected the images at DhKm-3, DiKm-3, DhKo-l, DiKp-l, DiKn-l, and
DhKm-5. These shapes can be described by carefully using the VIPS software. This software
was useful in establishing whether images were superimposed over white mineraI deposits
which in turn obscured other red images. DhKm-5 is an example of a site where red images
are superimposed over each other and are affected by white mineraI deposits.
My examination of the photographs of this site using VIPS/ip reveals that images
have been created at DhKin-5 several times over an unknown period. The different colours
indicate that images have been painted over other images. The white images in the black
background in A22 are the same red images affected by white mineraI deposit in A1. Once
the image file is loaded into VIPS the slider turns all the red pixels selected in the Al box
into white pixels and every other colour into black pixels. Ali the red pixels, regardless of
whether they are below or superimposed on white mineraI deposits, will become white
pixels, shapes among the black pixels in the box called A22 in figure 5.10.
Page 263
Figure 5.10: Screenshot of sorne of the red images be10w the white mineraI deposits.
Figure 5.11: If this screen shot is compared with the screen shot above it is c1ear that the accumulation of white mineraI deposits at DhKm-5 is not uniform.
Page 264
It is possible to establish whether red(s) was superimposed upon another red are a, or
whether the pale red/pink areas are covered in white mineraI deposits in the are a of red
pixels in box A7, by comparing it with the red pixels in the box A2. When l moved box A7, l
isolated a shade of red coloured pixels different to the one that l had selected in the previous
screenshot, to see whether other images that are not easily visible existed on the same surface
isolated in box A2. If white images in box A22, which are in reality red, are compared
(figures 5.1 0 and 5.11) it is clear that superimposed images exist at the same place and are
affected by the white mineraI deposits. This deposition is not uniform, and it does not coyer
the entire image. It does cause images to become paler, and to "disappear" underneath it.
Image analysis of the TIFF files using software such as VIPS and Adobe PhotoS hop 6.0
indicates the fragility of the sites and their images together with their changing nature. It also
demonstrates the key role of the human eye in recognizing the images themselves.
The human eye and the brain play a major role in the identification of individual
images. This role is most clearly evident if a specifie image has been isolated from its context
upon the surface of the rock using the VIPS software, and is viewed against a black
background away from its original context. It becomes evident that these red images were
not solid blocks of red upon a grayish granite surface, sin ce the edges of a group of white
pixels that once constituted the red image were not clearly delineated and the centre of an
image was not solid white.
l used various tools available to try to mathematically remove the white pixels
deemed 'noise' and attempted to make the image in question become solid.
Page 265
Figure 5.12: Screenshot showing what occurs to the same images using different tools when 1 mathematically tried to remove the white pixels deemed 'noise".
l realised, by doing so, that my eyes and brain were 'filling in the gaps'. Essentially the
brain and the eye of the observer make decisions as to the precise edges of each image. The
part played by the human being in identifying the shape of an image increases when the
edges of superimposed images must themselves be estimated.
Conclusion:
Unfortunately, many of the images have been affected by exfoliation, white mineraI
deposits, and the accumulation of rock tripe and lichen. This means that the images that l
described no longer had their original distinct clear edges, and many had been destroyed.
Many of the images that l did describe were nearing the end of their liEe-cycle, creating
severe problems for those attempting to describe them. Image analysis using Adobe
PhotoS hop and VIPS indicated the fragile nature of these images, the sites, and their
changing natures. Use of this software highlighted the considerable role played by the human
eye and brain in the identification and subsequent description of these images.
Page 266
Section 2: Distribution of images
It is generally assumed that a vast range of images was produced at these sites.
However, relatively few images exist. A total of three hundred and eighteen images were
recorded with an average of eleven images per site and a minimum number of one. The map
below in Figure 5.13 indicates the number of images at each site.
1, DiKm-3, DiKm-4, DiKm-51, DiKn-l, DiKp-l, DiKq-l0, DjKn-l, and DjKr-23) have
ninety-three images that cannot be classified as having solid or hollow centres because these
shapes are either individuallines or are created from connected lines (xq757).
Lines that are connected to each other can constitute another shape. Therefore lines
(xq325 and xq324), horizontal lines (xq337 and xq338), vertical lines (xq339 and xq337a),
diagonallines (xq747 and xq747a), triangles (xq370a), rectangles (xq748), squares (xq746) and
circles rarely exist as individual shapes since they are mostly parts of other shapes (xq751 and
xq751a). The image below was described as: rectangular box shaped image with round circle
on one end of the lowest of the four v shaped lines that cross the middle of the rectangle.
The top end of the rectangle is closed by the first of the vs (xq748) in figure 5. 17 below.
Figure 5.17: Close up of a rectangle with the circle and the diagonallines from DhKm-3 in 2001.
Shapes often exist inside other shapes. This makes it difficult to determine whether a
shape is hollow, solid, or unclear. Sorne shapes just contain lines, while others incorporate
other shapes. Three shapes have lines in their centre at DhKm-3 (xq753), while images other
than lines exist in one image from DhKm-18 and in five images from DhKm-3 (xq759).
Completely recognizable images such as stick figures were found to exist inside other shapes.
At DhKo-l there were two vertical stick figures inside other shapes (xq 515). Here a vertical
Page 276
stick figure exists inside a rectangle that is narrow and has open gaps at the right and left
ends. This hollow rectangle, visible in figure 5.18 below, contains a vertical stick figure with
two arms down and two legs and verticalline that crosses inside of the right hand end of the
rectangle (xq748, xq358b, and xq358a).
Figure 5.18: The rectangle with the stick figure at DhKo-l in 2001.
Conclusion:
Every attempt was made to objectively describe the images at these sites. Many of the
images in the Lake of the Woods, as described in depth, are unique. Hence, a large number
of different types of image also exist at only one site. Images are either hollow or solid, and
may be constituted of several shapes, such as the example in the photograph above.
Furthermore, if images are described more generally as 'creature' or 'zig-zag', it is easy to
establish that the range of images is small and there are large numbers of the same type of
shape. A nested hierarchy of types clearly leaves the most options open with regard to
classification.
Section 4: The techniques used to make paintings
Ethnographie information concerning the type of implements and techniques used to
create images does not exist. Pohorecky (1968: 106) posited that paint was applied to the
surface of the rock either with fingers or simple brushes. It is likely that fingers were used to
apply paint to the surface of the rock.
Fortunately, some ethnographie evidence exists as to paint components. At least
some paint was made from ochre and isinglass. Isinglass comes from the air bladder of the
sturgeon, a fish commonly found in this region of North America. Jones (1980: 48) reported
that, in 1972, he was told bya Swampy Cree informant from Norway House, Manitoba that
an old man had told mm "that ochre was mixed with sometmng from a sturgeon to make a
painting or rock paintings." Jones (ibid.) argued that this was the "only direct evidence of the
Page 277
connection between ochre pigment and the use of isinglass (or sorne other substance) from
sturgeon as a rock painting binder." Skinner (1911:23) observed in 1908 that the East Cree
men put a mixture of red ochre and isinglass on their legs which would not wash off and
must be le ft to "wear away." He also described a Northern Saulteaux legend from Martens
Falls Post on the Albany River, Ontario, where isinglass was described as coming from
sturgeon (Skinner 1911: 171). Stevens (1971:89) related a legend collected from the Ojibwa
from Sandy Lake, Ontario, where an old shaman gave a young man "a powerful glue that
had been made from a marrow of a sturgeon backbone." He used this glue on his feet and
hands and it allowed him to climb up to the top of sorne steep cliffs where the Thunderbirds
lived. James Isham in his "Observations and Notes on Hudsons Bay, 1743" reported that:
the Glue the Natives saves out of the Sturgeon is Very strong and good, they use itt {sic] in mixing with their painting, which fixes the Colours so they never Rub out
(Rich 1949: 169).
Dewdney (Dewdney and Kidd: 1967: 21) stated that B. Amos Ratt, an informant
from La Ronge in Northern Saskatchewan, had told him that "he had taken earth from a
deposit on the Churchill River" and had "reddened" it by "bringing it to a white heat in a
frying pan. Amos said that he subsequently mixed this "earth" with oil rendered from white
fish gut and used the substance "to waterproof his log cabin and paddles."
Densmore (1974:370) discovered that the reddish substance that "rose to the surface
of certain springs" in one of the regions where she conducted fieldwork, was collected, dried,
and baked in a fire. According to Densmore it "became hard like a stone". This substance
was subsequently powdered and the fine red powder was kept. Tt was mixed with grease to
produce paint "that was reddish but not vermillion," and which, Densmore, stated was used
on arrows, faces, and bodies. She argued that this "scum" contained iron oxide and the
powder was referred to as ochre. This is interesting information but it is unclear where it
came from, since Densmore conducted the fieldwork for this publication at several
reservations in Minnesota, including White Earth, Red Lake, Cass Lake, Leech Lake or Mille
Lac, the Lac Court Oreilles Reservation in Wisconsin, and the Manitou Rapids Reserve in
Ontario (ibid.: 281).
Page 278
.~ ...•
Conclusion:
Very little ethnographie evidence exists as to how the paint was made for the images
and how it was prepared and applied. However, a few reports do exist from different parts of
the Canadian Shield, from the Cree and the Chippewa but not from the Lake of the Woods,
eoncerning the paint's eonstituents. It is likely that it was made from oehre and isinglass.
Unfortunately, ethnographie information do es not exist eoncerning the type of implements
used to ereate images on the surface of the rock face.
Section 5: Discussion of the offerings left at sites
Ethnographie literature of this region indieates that gifts sueh as tobaeeo eould be
given to supernatural beings to obtain supernatural bene fit, espeeiaily in a ritual eontext.
Both Hilger (1936: 20) and Densmore (1974 [1928]: 325) observed that the Midé always
offered tobaeeo when gathering plants for medieinal uses.
Warren (1984 [1885]: 79) observed that giving something, like tobaeeo, to an eIder
when genuine information was sought from them, was "an invariable eustom" of their
religious beliefs. He noted (ibid.: 64) that if Indians wanted to address Kitehemanido this
address was always aeeompanied "by a sacrifice of tobaeeo or other article deemed preeious
by the Indian." Warren (ibid.: 287 and 337) observed several instances where tobaeeo was
also used as a gift item to prevent war or aggression. He deseribed several instances where
gift giving was part of negotiation as a means of preventing warfare (ibid.: 284, 365, 369, 371,
and 375). In one instance, Warren wrote that Monsieur Cadotte, a fur trader, gave presents to
the Dakota of tobaeeo and clothing as a means of preventing warfare.
Coleman (1947: 83) wrote that tobaeco had great power and eonsequently was a
eommon offering to ail the manitous. She noted that putting tobaeco on the ground or upon
the surface of water was eommonplace and that it "was often placed beside the dead for their
use on their long journey." Coleman, in an earlier published work (1929: 54) stated that those
who were siek and wanted to be eured by a specifie rock, owned by a pers on, and known to
cure the siek, brought it blankets, tobaeeo and other gifts to obtain favour. Her informant,
Mrs. Fairbanks, added that the person who owned the specifie rock took the gifts.
Over a century later, Pufahl learned from his informants at the Eagle Lake Reserve,
near the pietograph sites he was studying, that these sites were saered to the local indigenous
eommunity and "routinely receive[d] offerings of tobaeeo" (1990: 19). Smith (1995:33)
Page 279
argued that the use of tobacco was a sign of respect between humans and the manitous, and
asserted that it was common practice to offer either loose tobacco or tobacco in the form of
cigarettes to the EIders. However, if tobacco was buried it protected the individual buried
from the Thunderers (ibid.: 72). Tobacco was also either strewn into the water, smoked
before a trip or offered to Mishibeshu (ibid.: 119-120). She noted that Hilger had observed
that this was common practice between 1930 and 1940.
Offerings above water have always been sought by archaeologists who have
examined pictographs. Fox (1974), Dewdney and Kidd (1962 and 1967), Lambert (n.d.),
Pastershank (1989) and Reid (1975 and 1976) examined pictograph sites in the Lake of the
Woods for previous offerings. Therefore, it was important for us coilect this type of
information as weil. Neither of the two caves held anything that could be cailed an offering
(xq 124b). Fifteen sites in total, including the two cave sites, did not have any offerings:
1979b) and Reid (1974, 1976, 1977, 1979, 1988) did not examine this pictograph site.
Lambert examined this site the same year, July 1985, as he did DkKn-7, also in Blindfold
Lake. He argued that the detail of the shapes he saw were hard to distinguish as a
consequence of weathering on the surface of the rock. Furthermore, the rock was covered
with white mineraI deposit which was in turn "overlain by extensive but patchy growths of
lichen" (ibid.: 58). Lambert argued that, although lichen encroachment had started to occur,
it did not affect three of the five shapes that Lambert had identified. l examined this site as
part of my fieldwork in 2001. The five shapes identified by Lambert in 1985 were only
identifiable as four oblong blobs in figure 5.125 below. One of the shapes identified by
Lambert had disappeared.
Figure 5.125: DkKn-6 in 2001.
The disappearance of the fifth shape was probably caused by a combination of
factors. The photograph above illustra tes that lichen growth and encroachment is a huge
problem. Rock tripe has also begun to grow amongst the lichen. lt is also possible that
exfoliation has aided the destruction of the four shapes that are currencly barely visible.
DhKn-2:
This image, cal1ed Site # 107, existed on the south end of the Channel between
Obabikon Lake and Obabikon Narrows and was called the Obabikon Narrows
Page 358
Lichenoglyph (figure 5.126). Dewdney located it at the same time that he found DhKn-1
thanks to information S. C. O. Linklater had supplied to him. Bob Archie from Morson,
Ontario, informed Linklater of its existence and noted its physical location as well as the
following brief description as "a figure of a sort of head. The moss grows all around the
figure. The pictograph is on a boulder about 5 or 6 feet above the water line." (New World
Archives, Royal Ontario Museum)
, .
. ' 7'''''-WdU.
"7usfIJ ,/j60
SITe #107 -adc,ign scropad outof/tch.rn on boa/der O,fockrt"d with offènil'l>
in Obabli:a NarroWf .North ofOba",l<a /Ja''{,
soorn of AUIIlf?aU P€ni'l,l/Ia,
Lak.< of' tne U'RJOq<.
Figure 5.126: Dewdney's field drawing of Site # 107 in Obabika Narrows undertaken on August 19 1960.
Molyneaux and his team from the Trent University examined the lichenoglyph and
concluded that "deterioration by means of lichen encroachment was inevitable" (1974: 10).
Indeed, photographie analysis of this shape indicated that lichen growth had "obliterated"
many of its specific details (ibid.).
During my 2001 doctoral field work we attempted to find the image drawn by
Dewdney. We could not find it despite spending several hours searching. It can only be
presumed that the lichen had grown so much that the image no longer exists.
DjKn-1:
Dewdney recorded this site, called "the Blindfold Lake Site", as Site # 29 in 1958.
Figure 3.18 indicates the precise location of the offerings that Dewdney found in 1959 as
well as few of the images. However, other photographs of this large site were not found in
the Dewdney collection in the Royal Ontario Museum. The photograph in figure 3.19 is of
Page 359
part of this site, and was originally published in Dewdney and Kidd's publication (Dewdney
and !<idd 1962: 27 and 1967: 29). It was not found in the New World Archives held by the
Royal Ontario Museum. Despite only having one photograph in figure 3.19 it is evident that
the site suffered from exfoliation and that lichen grew on the rock's surface before 1962. It is
impossible to determine whether the whitish colour visible in the photograph was caused by
the accumulation of white mineraI deposits.
In 1974, Molyneaux and the Trent University team examined this site and argued that
the rock shelf, evident in Dewdney's photograph in figure 3.18, was used for leaving
offerings. This data, Molyneaux argued, indicated an "active, contemporary belief in the
sacred qualities of the site" (1974: 14). A deposit that varied in colour from "pale reddish
brown to white" was described as existing over many of the shapes. Molyneaux observed
that "deposit flaking" occurred and apparently caused the "pigment loss" (ibid.); perhaps
Molyneaux is stating that the shapes were exfoliating off the surface of the rock face.
However, he had observed that, although several small areas of exfoliation existed, there was
evidence that this had occurred before the shapes were created. Stress fractures also existed
in the rock itself but Molyneaux argued that they did not cause erosion of the rock's surface.
Lichen colonies were found "in spring seepage areas" (ibid. 15) and had started to encroach
on sorne shapes, but Molyneaux concluded that lichen encroachment was not an extensive
problem for this site. Vandalism was evidently a problem for Molyneaux since he found
sorne names and initiaIs painted on the rock face "100 feet to the right of the site" (ibid.).
Molyneaux published a photograph (figure 5.127) of part of this site but it 1S unclear in which
year he took it. He implied in his article that he took it after the dry winter of 1973.
Figure 5.127: Part of DjKn-1 taken by Molyneaux.
Page 360
Reid examined this site in 1975 and assessed it as having "high representability and
high interpretability" (ibid.: 14). Unfortunately he did not provide a photograph of this site;
hence it is difficult to establish the physical status of the shapes, the amount of lichen and
rock tripe encroachment, the deposition of white mineraI deposits, as weil as the possible
rate of exfoliation. In 1978, Pelshea took photographs of this site (figures 5.128 and 5.129).
The red is very bright in both pictures. Lichen, rock tripe and white mineraI deposits were
present on the images when Pelshea photographed this site.
Figure 5.128: Pelshea's photograph of the left hand side of DjKn-l taken in 1978.
If the ab ove photograph is compared to Dewdney's photographs taken in 1959
(figure 3.18 and 3.19), it is impossible to undertake a detailed comparison since Dewdney's
photograph only covers a smail portion of the images. The water level was higher in 1978
than in 1959, since the lowest image on the left hand side in Pelshea's photograph (figure
5.128), is doser to the water level than it was in Dewdney's 1958 photograph (figure 3.18).
Page 361
Figure 5.129: Pe1shea's photograph of the right hand side of DjKn-1 taken in 1978.
Unfortunately, it is difficult to de termine precisely how much higher the water level
was when Pelshea recorded the site in comparison to when Dewdney recorded it in 1969.
Exfoliation is evidently still an important issue, as is the deposition of white mineraI deposits.
The quantity of lichen and rock tripe has evidently increased since 1959 when these images
were first photographed.
Lambert argued in 1985 that, prior to conducting field work at this site, he had
observed that Dewdney's drawings and photographs were not identical and showed
differences from the site as Lambert recorded it in 1985. Whether this i5 a valid statement
remains impossible to establish, since Lambert's photographs of the entire site were
impossible to obtain. He (n.d.) concluded that microscopie erosion was inevitable "because it
was a natural process" (ibid: 26) but that no evidence existed of rock spalling (ibid: 27). He
asserted that two types of white mineraI deposit existed. One type on1y existed in the central
portion of the cliff face of Face 1, and it was so thick that it had obscured portions of three
shapes (ibid.). The second type was thinly overlaid over all of the shapes. Broad areas of
lichen existed on the top half of what Lambert termed Face 1 and "obscured sorne painted
areas" (ibid) and probably sorne shapes.
This site was photographed in 2001 and has three parts (figures 5.130, 5.131, and
5.132).
Page 362
Figure 5.130: Part of DjKn-l in June 2001.
Lichen and rock tripe encroachment is evident at the top of this part of DjKn-1 in
figure 5.128. The images are severely affected and in sorne instances impossible to identify
clearly. The site is also evidently affected by the accumulation of white mineraI deposits in
the centre and this has caused sorne of the images to become faint. Exfoliation is also a
problem at this site. The high water mark is clearly evident at this site.
Figure 5.131: Part of DjKn-l in June 2001.
The shapes in figure 5.131 are not as affected by lichen or rock tripe encroachment
as those in figure 5.129. White mineral deposits clearly superimpose sorne of the shapes.
Page 363
Exfoliation is evident underneath the shapes, as Lambert had noticed earlier. The two large
cracks are evident as Lambert (n.d.) observed, but it appears that further exfoliation is not
evident.
Lichen and rock tripe encroachment is evident on the red hand print in figure 5.132
below. This image was found and identified as part of the site although it does not exist
beside the other parts of this site but higher up the same cliff. It is likely that exfoliation is
present but it is difficult to determine the quantity at this part of the site.
Figure 5.132: Part of DjKn-1 in June 2001.
DiKn-1:
Dewdney recorded this site as Site # 91A and B on 15 July 1960 (figure 3.16). It is
found on Hay ter Penin sula, on the eastern portion of the Aulneau Peninsula. Unfortunately,
a close-up photograph of the site and rus drawing of the entire site could not be found and l
only found the photograph below (figure 5.133) in Dewdney's papers held by the ROM.
Page 364
Figure 5.133: Setting from a Distance of Site # 91A and B taken by Dewdney.
Dewdney divided trus site into two parts, which he called A and B. Part A was the Iower part of the site wrule Part B was the upper part of the site. He was particuiarly interested in a shape called "a checker-board," which he argued couid be a "weaving texture" (Dewdney and Kidd 1962: 50; 1967: 52). Figure 5.134 shows this image.
Figure 5.134: "Face lB figure c + d" by Dewdney in 1960.
Page 365
Figure 5.135: "Face lA fig a + b" by Dewdney in 1960.
It is clear from the photograph ab ove (figure 5.133) that exfoliation is quite severe
and that evidence exists of rock tripe and lichen encroachment. White mineraI deposits are
not as clearly evident here as at sorne sites.
Figure 5.136: DiKn-l in June 2001.
This site was photographed and examined in 2001 (figure 5.136). It is evident that it
suffers from severe lichen and rock tripe encroachment and that many of the images are
severely exfoliated. Exfoliation rather than the accumulation of white mineraI deposits is
more of a problem at this site. It also means that the images are very difficult to identify.
Page 366
Indeed it is difficult to see some of the shapes that Dewdney photographed in 1960 (figures
5.135 and 5.136, since they have been so severely exfoliated (figures 5.137 and 5.138).
Figure 5.137: Blob beside exfoliated area at DiKn-1 in 2001.
The 'checker-board' that Dewdney identified in 1960 beside the blob in the photograph
above has disappeared due to severe exfoliation of the rock's surface.
Figure 5.138: Dewdney's "Face lA fig a + b" at DiKn-1 in 2001.
It is evident if the photograph above (figure 5.138) is compared with the photograph in
figure 5.135 that the images viewed by Dewdney in 1960 are still visible but they are affected
by lichen encroachment and exfoliation of the rock's surface. It is clear from the photograph
below (figure 5.139), that lichen and rock tripe encroachment has severely affected the
shapes Dewdney identified in 1960.
Page 367
Figure 5.139: Red images are visible although partially hidden by the lichen and rock tripe in the upper half of DiKn-1 in 2001.
If the photograph below (figure 5.140) which is a close-up of this part of the site is
examined, it is clear that the red shapes in the upper part of DiKn-l, which Dewdney called
Part B, are impossible to identify.
Figure 5.140: Shapes ab ove the main collection of shapes at DiKn-1.
Lichen and rock tripe are clearly so severe that it is impossible to identify any images.
It is only possible to state that red images are present.
Page 368
DjKn-2:
Dewdney did not examine this site but, Lambert argued (n. d.: 52) that he had
established its existence from Dewdney and Kidd's (1975) publication of their original (1963)
book. Lambert noted that Dewdney had mentioned this site while he described the existence
of Site # 29 or DjKn-1. Reid (1976: 25) observed that the site was in good condition but
some danger existed that it might be vandalised since it was so weil known to the general
public.
Lambert reported (n.d.: 52) that rumours existed as late as the 1970s that this site was
destroyed since someone had fired at it using a shot gun. Lambert examined the site to
ascertain whether it had been used for target practice. He ascertained that weathering and
rock spalling had affected the shapes at this site rather than gun fire. Lambert also concluded
that white mineraI precipitate and lichen was not present at this site. In 2001, l heard the
same rumour about gun fire having affected this site. However, l was not able to find this
site when l searched for it.
DhKm-20:
A Borden number has been obtained since this site was discovered in 2001 during
my fieldwork. Unlike most of the other sites in the Lake of the Woods this site is possible to
date since it was dated by the author. The shapes at this site (figure 5.141) are quite unusual
and were created using modern paints in red, blue, and white.
Figure 5.141: DhKm-20 in June 2001.
Page 369
These shapes were painted by R. Bird and the images are dated 1982. The date is
painted in red paint below the lower front part of the fish shape. Hence these images were
nineteen years old. The red shape, above the fish shape, is affected by lichen encroachment
on its Ieft hand side while the white and blue fish shape is affected by white mineraI deposits.
These images are not affected by rock tripe encroachment. It is difficult ta establish whether
the images are affected by exfoliation since it is possible that it has started to occur on the
edges of the images but evidence of it occurring is hard to determine with the naked eye.
DhKm-19:
This site was also discovered during 2001 and a Borden number obtained for it. It
exists on the south-east facing shore of Turtle Lake a few metres from red channel marker
S44. It, too, was created with modern green and black paints but, unlike the shapes at
DhKm-20, these shapes are not dated or signed (figure 5.142). Neither the style nor the form
suggests the same author created the images.
Figure 5.142: DhKm-19 in 2001.
This image is not affected by the accumulation of white mineraI deposits or exfoliation. It is
also not affected by lichen or rock tripe encroachment.
Page 370
DgKm-Z:
This new site was found near to DgKm-1 during 2001. A Borden number was
obtained for it. It (figure 5.143) lies southwest of Sand Bay in a small, unnamed bay.
Figure 5.143: DgKm-2 in July Z001.
Six shapes exist here. They suffer from exfoliation, and probably from spalling,
which may explain why these images are smears. White minerai deposits cause problems for
these images but fortunately the lichen and rock tripe encroachment exists above the images.
DiKm-49:
This site, found in 2001, is inside a granite cave immediately behind DiKm-3. The
site is on an island southwest of Return Point, north-northeast of Three Sisters islands, in
Whitefish Bay. Two lines are present on the ceiling of this cave, both are visible in figure
5.144 below.
Page 371
Figure 5.144: DiKm-49 in June 2001.
These two lines are near the entrance of the cave, hence the sunlight visible on the
right hand side of the opening. The shapes on the ceiling are affected by lichen but not by
rock tripe encroachment. Accumulation of white mineraI deposits and exfoliation are not a
problem.
DiKm-50:
This site, in a cave, lies a few metres south of DiKm-51 (initiaIly caIled Site # 270 by
Dewdney) in Whitefish Bay. It is located on the southwest facing shore of the lower arm of
the second, large, horseshoe-shaped island just south of the Three Sisters Islands. This cave
site was discovered in 2001 and a Borden number was obtained for it. The photograph
below (figure 5.145), indicates that the shape here is a large smear which exists on the ceiling
of the cave. Unfortunately, exfoliation is a problem causing the image to faIl off the ceiling.
This could be accentuated by water coming into the cave through an opening near the
water's edge.
Page 372
Figure 5.145: DiKm-50 in June 2001.
Neither the accumulation of white mineraI deposits, nor rock tripe and lichen encroachment
threaten the images in this cave.
Conclusion:
Direct dating techniques cannot be used as stated in the beginning of this section,
rendering it difficult to date ail the images at these sites. However, the deposition of white
mineraIs, lichen and rock tripe encroachment, and exfoliation of the rock surface take place
over time, destroying or causing the images to fade (see Chapter V, section 1). The offerings,
especiaily the artefacts left at sites, can be dated but they only indicate that the site was
visited by people. It is possible the rock images were created there both before and after
artefacts were left there. It is also possible to date by association sorne of the sites since
independent evidence is available. Clearly, this question of the age of the images is important
and therefore three questions were asked at the ons et of this section:
(a) Is there independent evidence of when these paintings were created?
(b) Is it possible to calibrate the rate at which white mineraI deposits accumulate,
rock tripe and lichen grow and spalling occurs?
(c) Have the images present at pictograph sites changed stylisticaily over time?
The discussion of the images and the drawings of these sites and their images
illustrates that it is relatively easy to evaluate the age of sorne of these sites and to answer
Page 373
question (a). lndeed, DhKm-20 was identified as a pictograph site in 2001 but it was created
by an individual who left his signature, "R. Bird," and he or she provided a date of 1982
below the right hand corner of the front of the fish (figure 5.146).
Figure 5.146: DhKm-20 created by R. Bird and signed in 1982.
The four modern red spray paint images present at DgKI-2 were painted at DgKl-2 after
Pastershank recorded the ochre images in 1989 and before 2001 when land my crew
recorded the site. The three solid images, superimposed over older ochre images that
Pastershank had recorded in 1989, are visible in Figures 5: 147 and 5.148.
Figure 5.147: Two of the modern solid spray painted images at DgK1-2.
Page 374
Figure 5.148: The third modern image which is solid at DgK1-2.
The fourth modern image created from spray paint was a hollow shape existing to the Ieft of
the older ochre image in figure 5.149 below.
Figure 5.149: The modem hollow spray painted image at DgK1-2.
A vast quantity of offerings, wruch kept on growing in quantity throughout the
summer of 2001, were found at DgKl-2. l established that some of the new images here were
connected to a court case regarding charges of incest in one of the reserves in the Lake of
the Woods. However, l have been asked by W. A. Ross, who was the regional archaeologist
in 2001, not to provide any more details regarding the images and why they were created
(Ross. pers. Comm. July and August 2001). Unfortunately, none of the offerings Ieft at trus
site had anytrung that could be securely dated.
Page 375
The image present at DhKm-19, discovered in 2001 (figure 5.150), was painted using
modern spray paints but unfortunately it cannot be dated using independent evidence. Prior
to its being recorded in 2001 no-one had either recorded it or reported its existence to
Ontario MÎnistry of Culture and Communications. However, more can be concluded about
this image with respect to its style in the answer regarding question (c).
Figure 5.150: Modern spray paint image at DhKm-20.
Independent evidence must include objects that can be dated to specific time
periods, such as offerings found either below or above the surface of the water at pictograph
sites. The presence of the offerings underwater enables one to posit that these sites, and not
just the images, have had considerable importance for a long period of time, possibly, in the
case of DhKm-1 back to at least 800 AD. This statement can be made because objects, called
o fferings , were found below the surface of the water immediately beside the cliff on which
the paintings exist (See Appendix 5 of this chapter). Offerings were also found above water
level in 2001 but it is the offerings under the surface of the water that provide the strongest
indicators of the length of time this place has been considered as important to one or more
groups of people. A small ceramic pot (see figures 5.37 to 5.44) was found at the base of
DhKm-1 at 15.85 metres below the surface of the water. It was dated on the basis of its
physical similarity to those found amongst the grave goods during Kenyon's excavation of
the five Armstrong Mounds (1986: 12-16), which he ascribed to the Late Laurel period and
argued were constructed about 1,000 years ago. Ross also dated this pot, based upon a
careful examination of the photographs above, from the period between 1400 and 800 A.D.
(Ross pers. comm. August 2001). A small hand-blown glass bottle dated to the nineteenth
century was found about 1.52 to 1.83 metres away from the ceramic pot, also below the
surface of the water (see figures 5.32 to 5.37). Meyer disagrees with this interpretation
because he identified as belonging to the Duck Bay complex of the Rainy River composite
Page 376
dated by Leruus and Olinyk (1990: 84) as being from the period between 1100 A. D. and
1350 A. D. (Meyers pers. comm. September 2006). The divers also found a large pile of
clotrung leaning against the cliff face at the same depth as the pot and glass bottle. A wide
range of objects was found, including moccasins, large quantities of fabric and leather clothes
of different styles and different periods. They also found a wrute cruna plate with a brown
design, which had the maker's mark of Alfred Meaken Royal Ironstone China, England,
wruch had Queen Victoria's crown, lion, and unicorn on the base (see figures 5.30 and 5.31).
Evidently, DhKm-l has a vast range of evidence that provides a strong indication of the
considerable length of time it has been used. Dewdney's observation of the offerings in 1960
and Fox and Molyneaux's descriptions of the offerings in 1974 also confirm that trus site has
been continually used for a long period of time (see Appendix 5).
A chipped bowl was found under the surface of the water near the pictograph site
DiKm-50, on an island locally called Horseshoe Island, located north of trus site but also in
Wrutefish Bay. It was found at an underwater fireplace that was 2.44 cm (8 feet) away from
DiKm-50 (see Appendix 5). DiKm-50, the cave site on Horseshoe Island is also a few metres
south of DiKm-51 wruch is a pictograph site on a cliff face first recorded by Dewdney in
1970. It is possible that trus fireplace could have been used prior to the raising of the water
level since the crupped bowl was found below the freeze level. However, a broken plate was
also found with the same design and the same maker's mark at the base of DgKl-17, in
Sabaskong Bay. Given that trus plate was found at the base of DgKl-17 and another at
DhKm-1 (see figures 5.30 and 5.31), it is possible to argue that the physical places where the
paintings now exist have been considered important since the nineteenth century. Indeed,
the presence of the china plate at DgKl-17 enables the archaeologist to provide a date for the
possible use of trus site as a place of significance in the landscape. Dewdney cleady knew of
this site but did not record it, although he had an undated black and wrute photograph of it
(figure 5.106) as well as a sketch map as to how to find it in rus notes. It was first recorded by
Pastershank in 1989 but she found it because she was informed of its existence by Ron Kelly
and Ron Seymour, from the nearby reserve.
Coins are not as useful indicators in comparison to plates, bowls, and pots as to the
length of time that a physical locale used for a long period although many were found at
DiKn-1, DjKr-23, DiKm-4, DhKm-5, and DhKm-1. Indeed, a quarter dating from the reign
of George V as well as an American Penny was found at DhKm-1. Otherwise all the coins
Page 377
were Canadian. It is difficult to rely on coins as datable objects although they have dates on
them because these coins could have been in circulation for a considerable length of time
before they were left as offerings.
Another method of determining when a physicallocale was first used as a pictograph
site is to examine the records made by previous archaeologists. All of them have recorded
the presence of objects, which are offerings. These were found by Pastershank in 1960 by
DgKl-1, DgKI-2 but not at DgKI-17 and DgI<l-19 (Appendices 1 and 4). However, by 2001
offerings were found at DgKl-1, DgI<l-2, DgI<l-17 but not DgI<l-19. It could be argued that
DgKl-1 and DgI<l-2 were still being used but it is unclear when they started to be used. It is
clear from the offerings at DgKl-17 left below the water level indicate that it has been used
for some time since the offerings found contained a broken plate that dated from the
nineteenth century. Dewdney found and recorded offerings at DhKm-1, DhKm-5, DhKn-1,
DiKn-1, and DjIv-23 in 1960, and offerings were found at these sites in 2001. Dewdney also
recorded offerings at DhKm-18, DiKq-10, DjKn-1, DiKm-3, DiKm-3, and DjKn-1, but
offerings were not present in 2001. He did not record the presence of offerings at DhIZrn-3
and DhKo-1 when he recorded both sites but offerings were found there in 2001. Although
a considerable body of information was collected by Dewdney and subsequent archaeologists
regarding offerings, they did not provide any information that enables these items to be
identified within a specifie time period. Very litcle evidence exists as to when these red
images were first created. Indeed, Dewdney, who first recorded many of the sites, relied on
reports of these sites from informants. But none of these informants gave any indication of
the length of time these sites had existed prior ta report. Lawson, a geologist, was the first
pers on to record and describe a pictograph site in the Lake of the Woods, DiKp-1, in 1885
(see figure 3.3). Dewdney, although he was not the first person to photograph DhKo-1, did
establish that this site was known about as early as 1910. He found this site using a sketch
map (see figure 5.67) and a sketch of the images provided in 1936 by a guide, H. T.
Thompson, who maintained that the site was first found in 1910 but was known by all the
boat captains who worked in the Lake of the Woods prior to this date.
If it were possible to take a sample of pigment from a site it might be possible to date
the superimposition of paintings over painting and white mineraI deposits. However, none of
the sites can be examined using this type of analysis. But it does lead us to consider the
answer ta question (b). It was very difficult to answer this question. Critical comparison of
Page 378
field drawings, photographs and the associated reports and slides indicate that care should be
taken regarding any assertions concerning the accumulation of white mineraI deposits as weil
as growth of the lichen and rock tripe over images if they are considered crucial in calculating
the possible age of a site. Information from previous studies is useful, since it assists in
establishing the rate at which lichen and rock tripe encroach, exfoliation occurs and white
mineraI deposits accumulate at rock image sites in the Lake of the Woods. But it must be
remembered that measuring lichen and rock tripe encroachment is particularly problematic
since both can be removed relatively easily from sites as the photographs of the images of
DhKo-1 demonstrate.
Several researchers studied the micro-stratigraphy of the surface of the rock face in
which the pictographs exist. Although these studies were discussed in Chapter III it is
important to highlight the key points. Taylor, Myers and Wainwright (1974) examined seven
samples of paint from the Agawa pictograph site in Ontario, and from a site on Lac
Wapizigonke, Québec to determine its formation and composition. They intended to
establish the causes of the natural deterioration of pictograph sites in the Canadian Shield,
prior to devising any conservation treatments. However, they (ibid.: 40) concluded that the
pigment was "firmly attached to and intimately mixed with the [white] mineraI deposit in a
sandwich structure" instead of adhering to the surface of the rock. The white deposit found
on these cliffs coats the images and provides a "protective coating from the leaching action
of the rain and erosion by wind" (ibid.) as weil as preventing the pigment from dissolving
away. Taylor and his coileagues argued that the graduaI accumulation of this white mineraI
deposit on pictographs led to "the 'faded' appearance" of many of the images. Essentiaily
paintings were becoming obscured over time and the paintings eventuaily "disappear[ed]".
They (ibid.: 41) proposed that it might be possible to utilise the thickness of the
white mineraI deposit upon the surface of the pigment to estimate a date. However, they
abandoned this ide a because too many variables affected the rate at which this deposit
accumulated. It was unknown whether the rate of accumulation of white mineraI deposits
was continuous or had changed over time and whether it was influenced by other factors
affecting the surface of a cliff face, such as crack formation. Other factors that should be
discovered are the exfoliation of the rock's surface, lichen and rock tripe growth and
encroachment.
Page 379
Myers and Wainwright's (1977) study of samples from over 30 sites m the
Similkameen Valley and the Canadian Shield, confirmed Taylor and his colleagues' (1974)
conclusion. lndeed, Myers and Wainwright (1977: 31) asserted that the pigment was
"embedded in the deposit before and after the paintings were made". Wainwright and
Taylor's (1977) study using scanning electron microscopy determined the structure of the
pigment of two paint samples. One was taken from a painting on Lac Wapizagonke, Quebec,
and the other was removed from a painting at the Bon Echo site on Mazinaw Lake in
Ontario. Their results were important, although only two samples were examined, since they
demonstrated severallayers of pigment may exist at pictograph sites which appeared to have
been painted oruy once.
It is possible to condude that dating the images usmg the stratigraphy of the
pigments on the surface of the rocks is not very useful to my discussion but it darifies the
precise relationship of the pigment to the white mineraI deposit on the surface of the rock.
Another avenue of research for dating these images would be to consider the rate at
which lichen and rock tripe encroachment occurs, and whether either plant can be dated. It
is a commonly established fact among those who have examined pictograph sites in Québec
and Ontario that lichen and rock tripe occurs on pictograph sites (E.g. Arsenault 1998: 26-
27, Arsenault et. al. 1995: 46-47; Dewdney and Kidd 1962: 5-6 and 1967: 5-6; Molyneaux
1974 and 1974). Molyneaux (1974, 1975) reports that lichen and rock tripe affected the
pictograph sites of the Lake of the Woods but he did not draw any conclusions as to whether
different quantities of lichen and rock tripe exist on rock faces that face particular
orientations.
Neither did Dewdney specifically state whether the direction of a site may have
influenced the growth of lichen. However, he wrote that "fuzzy green lichen" grew on sites
"on which the sun never shines" (Dewdney and Kidd 1962, 1967: 6). However, Vaillancourt
(2003: 11) used this reference to support the view that north facing rock faces with images
are "habitually covered in lichen." She also based her conclusion on the following references:
Rajnovich (1994: 11), Arsenault (1998: 26-27) and Arsenault and his colleagues (1995: 46-47),
to substantiate her daim. However, none of these archaeologists made such a daim.
Arsenault and his colleagues (1995: 46-47) commented at sorne length on the orientation of
rock faces that had pictographs but they did not state that north facing faces had lichen.
Arsenault, later, (1998: 26-27) stated that sites predominately faced south, southeast and
Page 380
south west, although sorne faced north and west. Once again he did not specifically state that
lichen was found on north-facing rock faces with pictographs.
Dewdney (Dewdney and Kidd 1962: 6 and 1967: 6) claimed that it was north facing
cliff faces that did not experience sun that had "green fuzzy lichen" (ibid.) but he did not
clearly establish that north facing sites had lichen. Furthermore, the text that Vaillancourt
(2003: 11) quoted from Rajnovich did not support this opinion. It is impossible to draw the
conclusion that sites that face north have the largest quantity of lichen and rock tripe since
only one site, DiKo-l, faced north and it is clear from the photographs examined that the
lichen and rock tripe has been removed repeatedly.
Establishing the orientation of five pictograph sites in the field was difficult sin ce the
compass would not function at DgKm-2, DhKm-3, DhKm-l, DhKm-S, and DhKm-18 due
to the iron content of the rock upon which the paintings exist. However, it is possible to
establish their orientation using a compass and a map. Both cave sites must be excluded
from this examination since the paintings are inside the cave and the physical structure of a
cave site is different to that of the pictograph sites on cliff faces. Furthermore, each cave
examined can be entered and exited from two directions. Both sites, when compared are also
radically different. If appendix 5 is examined, it is clear that six sites are oriented towards the
southeast: DgKl-2, DjKn-l, DhKm-S, DiKp-l, DhKm-19, and DiKq-l0. DkKn-7, DhKm-
18, DiKm-3, and DgKm-2 face southwest while DgKm-l, DiKm-4, DgKl-17, and DgKl-l
face east. DhKm-20 faces northeast and DhKn-6 face northwest. DhKo-l faces north and
two sites, DgKl-19 and DhKm-l, face in a direction between east-southeast and east. DiKn-
1 faces a direction that lies between south-southeast and southeast and the face of DhKm-S
is orientated in a direction between southeast and east. One site, DhKn-l, faces west
southwest. DjKr-23 faces south-southeast while DhKm-4 faces in a direction between south
and south-southwest. DiKm-Sl faces in a direction between south and south-southwest.
These results, with the exception of the three sites that face north and northwest, confirm
Dewdney's (Dewdney and Kidd 1962: 6) general statements regarding the orientation of the
surface of paintings discussed in Chapter III. The physicallocation and the orientation of the
pictograph sites on cliffs in the Lake of the Woods cannot support Reid's (1980) observation
discussed in Chapter III. Therefore Rajnovich (1980b: 34)'s claim that south-to-east facing
rock walls were selected as places to paint pictograph, discussed in Chapter III, is equally
difficult to consider. A connection may exist between the orientation of the site and the
Page 381
~ ..
presence of exfoliation. Unfortunately, it is difficult to establish whether exfoliation exists for
DhKo-1 and DhKm-20. Exfoliation has occurred at DkKn-6 which faces northwest (see
Appendix 5). Appendix 5 demonstrates that only five of the ten sites that face directions
between south and east are affected by exfoliation. It is difficult to establish whether
exfoliation has occurred at DhKm-5, DjKr-23, and DgKI-1 while exfoliation does not appear
to occur at DiKm-4 and DhKm-19. Yet ali seven sites that face between south and west
southwest suffer from exfoliation
Appendix 5 illustrates that the three sites facing north, northeast and northwest
suffer from lichen growth. Fifty percent of the sites that directly face east or southeast suffer
from lichen growth and encroachment. Ali sites oriented between south and east have
images that suffer from lichen growth and encroachment. Ali four sites that face southwest
also suffer from lichen growth, but the images at DhKn-1 and DhKm-4, which face south
southwest and west-southwest, do not suffer from lichen growth. Why both sites are similar
in this respect is unclear since the physical location of the images at each site is different.
However, both of the se sites suffer from severe exfoliation. Neither suffer from rock tripe
encroachment. This is bizarre, especialiy since DiKm-51, which faces a direction between
south and south-southwest, suffers from exfoliation as weli as rock tripe encroachment.
Rock tripe encroachment should be considered next since it often occurs ln
conjunction with lichen. It occurs at two of the three sites that face in a northerly direction.
Rock tripe does not grow at DhKm-20, a new site, but it occurs at DhKn-6 and DhKo-1,
two older previously known sites. However, lichen growth occurs at ali three sites. The lack
of rock tripe at DhKm-20 may be because rock tripe takes a longer period of time to
establish itself in comparison to lichen. Rock tripe does not occur on any of the sites that
face east, or on four of the six sites that face southeast. The two southeast facing sites where
it does occur are DiKp-1 and DjKn-1. Both sites were among the first to be recorded and
discovered in the Lake of the Woods region. (Lawson recorded the images of DiKp-1 in
1885 and Dewdney recorded the images of DjKn-1 in 1958, whereas he recorded the
majority of the sites he examined in this region in 1960 or later.) The other sites that face
southeast were recorded from 1960 onwards until 2001. The five sites that face directions
between south and east, excluding those that directly face southeast, ali have rock tripe
affecting the images. It is clear that the images at DjKr-23, facing south-southeast, have not
yet been encroached on by rock tripe. Appendix 5 shows that the seven pictograph sites face
Page 382
directions between the west and the south. However, only two of these sites, DiKm-3 and
DiKm-51, are affected by rock tripe.
Appendix 5 makes it clear that all three sites that face in a northerly direction suffer
From the deposition of white mineraI deposits. It is clear that only two of thirteen sites facing
From an easterly to a southeasterly direction do not suffer From the deposition of white
mineraI deposits. Neither DhKm-3 nor DhKm-19 appear to suffer From exfoliation.
Appendix 5 also indicates that the majority of sites facing between south-southeast and west
southwest do not suffer From an accumulation of white mineraI deposits. Only two sites in
this category suffer From an accumulation of these deposits: DiKm-3 and DgKm-2.
Correlations exist, since white mineraI deposits, lichen and rock tripe always occur on
sites that face a direction between east and southeast, such as DhKm-5, DhKm-1 and DgKl-
19. Rock tripe sometimes occurs here. Sites that face southeast will most probably suffer
From exfoliation, the deposition of white minerai deposits, and, possibly, From rock tripe and
lichen. The relatively large number of sites in this group provides me with the ability to draw
some simple conclusions regarding the possible sequence of processes that occur at a site
after images have been created there. It is impossible to establish the precise date of image
creation at each site but sorne observations can be made if the recording date of each site is
used as a base line.
DiKp-1 was first recorded in 1885. By 2001, it suffered From exfoliation, the
deposition of white mineraI deposits, and lichen and rock tripe encroachment. Dewdney first
recorded DjKn-1 in 1959. In 2001, it suffered From lichen and rock tripe encroachment,
exfoliation and the deposition of white mineraI deposits. DhKm-3, recorded in 1958 by
Dewdney suffered, in 2001, From all of the same problems as DjKn-1 except for the growth
of rock tripe. Dewdney recorded DiKq-10 in 1960. It suffered From the deposition of white
mineraI deposits, but not From rock tripe and lichen encroachment. Unfortunately, it is
difficult to establish whether exfoliation was a problem when Dewdney recorded it.
Although DgKI-2 was first recorded in 1989, it suffered From the deposition of white
mineraI deposits and exfoliation by 2001. Lichen and rock tripe were not evident at this site.
DhKm-19, one of the newly discovered sites with new images, belongs to this group, and in
2001 it did not suffer From anything. Ali of the sites except for DhKm-19 suffered From the
accumulation of white mineraI deposits. It is difficult to establish whether exfoliation is the
second process that occurs at site since it is hard to evaluate its presence at sites if has just
Page 383
commenced. However, it is clear that lichen grows at sites before rock tripe. An additional
problem is whether rock tripe and lichen encroachment occurs over white mineraI deposits.
Clearly a sequence exists for sites that face southeast but unfortunately it is impossible to
calculate the rate of each process and when each one commences.
An interesting possibility is a correlation between the presence of blobs and
exfoliation, existence of white mineraI deposits, lichen and rock tripe encroachment and the
orientation of the rock face. The most common shape is a 'blob,' and seventy-seven different
types were identified to exist at the pictograph sites. However, a correlation does not exist
between the level of exfoliation, existence of white mineraI deposits, lichen and rock tripe
encroachment and a large proportion of blobs at these pictograph sites. Examination of
Appendix 5 indicates a relationship exists between the orientation of the site with the
number of blobs.
It is clear from Appendix 5 that the strongest relationship between orientation and
blobs exists for DiKn-1, DhKm-5, DgKI-2, DjKn-1, DhKm-3, DiKp-1, DhKm-19, and
DiKq-10. These sites face between east-southeast and south-southeast. A weaker correlation
is evident in Appendix 5 for those sites on cliffs that face southwest. Two of the four sites in
this group severely suffer from a large number of blobs. Unfortunately, it is difficult to
determine why only two of the four sites have blobs. It is likely this could be related to the
age of the images at each site. It is clear that not only does the rate of these processes differ
from one site to another (see Appendix 5) but Dewdney (an amateur researcher) and
subsequent archaeologists may have been interfering with this process by removing the
lichen and the rock tripe to establish the range and nature of the images at each site.
Dewdney indicated that he removed the lichen with a wet sponge from surface of the images
at several sites so that he could better see and record the images (for example Dewdney and
K.idd 1962: 6).
Another problem that complicates establishing the age of a site is that different
archaeologists have disagreed as to what was an image (please see Chapter III for a
discussion of why images such as smears were not considered to be images and the effect of
this decision.).
The accumulation of white mineraI deposits is influenced by the quantity of soil at
the top of the cliff, the number of times it rains during the year and the physical structure of
the rock itself. However, sin ce ground water sporadically seeps down the rock face, mainly
Page 384
,-r-...
during the spring and fall period, it is possible that faint images are older than the bright
ones.
An interesting alternative to resolving the possible age might be to examine DhKm-
20, a pictograph site whose surface suffers from exfoliation, lichen encroachment and white
mineraI deposits but not rock tripe. The red shape at the top of this site is affected by lichen
encroachment and the fish shape below is affected by white mineraI deposits. However,
none of the images is affected by exfoliation or by rock tripe encroachment although it is
present nearby. It also has a date of "1982", faces northeast and has no blobs. The site was
only 20 years old when l examined it in 2001. It is possible that these images have not started
to noticeably decay because they are not created from ochre paints.
If the orientation and the new sites are examined in conjunction with one or two of
the four problems that affect sites, interesting observations may be made. It seems likely that
lichen growth or the accumulation of white mineraI deposits are probably the first problems
to occur. The other issues just exacerbate the problem. Since lichen was removed on the
older sites by previous archaeologists, it is impossible to posit the precise rate at which
encroachment might occur.
So, what can be concluded? It can be argued that many pictographs are probably
older than 50 years. Wainwright and his colleagues during the 1970s pushed the date
backward with their observation that the pigments migra te , a process that would take sorne
time. However, image analysis of the rock surface aided me to establish that, although lichen
encroachment had occurred at the earliest documented site in Lake of the Woods (DiKp-1),
paler images, which Lawson (1885), Dewdney (1960), Molyneaux (1974) and Reid (1974) clid
not record, exist at this site. None of these adclitional, paler, and hence possibly older, images
were observed or cliscussed although the principal group of images had been examined in
considerable depth (see pages 330-335). Today, it is still possible to identify the images first
recorded in 1885 by Lawson, later by subsequent researchers, and only one of the images is
paler, since it has been affected by the accumulation of white mineraI deposits. Fourteen
images were identified at this predominately southeast facing site, of which two were faint
unidentifiable shapes and five were blobs. The blobs were not identified by previous
researchers and probably were not cleaned of lichen, if the surface of the main group was
cleaned. Hence, it is possible to posit the argument that lichen grows slowly on southeast
facing surfaces.
Page 385
It is difficult ta assess the quantity of mineraI deposits left on the rock surface, but it
is likely influenced by the depth of soil on the top of the cliff face, rather than in the rock
itself since relatively little or no white mineraIs were found on the paintings that occur on the
roof of both caves. Perhaps because cliff faces are vertical, rather than horizontal, water drips
straight down from the top of the cliff carrying with it mineraIs which get caught in
microscopie irregularities on the rock face. DiKm-50 was affected only by exfoliation while
DiKm-49 was affected only by lichen. None of the images at either site was new. The images
at both sites were mostly likely affected by wave action and changing water levels since they
both have sorne water close to their base. It is very hard to account for the rate of exfoliation
because of other influential factors such as the changing water levels, and ice formation in
different channels.
It is difficult to answer question Cc) regarding stylistic developments because changes
in how archaeologists have considered images, such as smears, has led to sites like DgKl-2,
DhKm-4 and DhKm-l becoming clearly defined, others recognised as being larger, and, in
the case of DhI<m-4 and DhI<m-l, as being separate sites. However, whether this has
affected the recognition of additional images at sites such as DiKp-l is difficult to determine.
It is clear that the use of new computing software has enabled images that were difficult to
see on sites such as DhI<m-3 and DiKp-1 to be more visible. Other than commenting on the
difference in the centres of the ochre images, very little can be concluded regarding the styles
of the ochre images. This chapter has already established that the ochre images in the Lake
of the Woods resemble those found throughout this part of Northern Ontario. However, the
images painted in modern paint, discovered and recorded in 2001, differ both From images in
ochre and between themselves as regarding their style and colours.
The images in the Lake of the Woods are different. Sorne were created in ochre,
sorne were painted in modern spray paints with a date, sorne were painted in modern paint in
several colours, and sorne were created using modern paint over ochre images after 1989.
The new modern images at DhKm-20 were painted in red, white, and blue paint. The fish
shape was created in blue and white with red spots either side of a horizontal white line
which runs the length of the fish shape. These shapes bear a strong resemblance to those
created by the artists who are considered members of the Woodland School of Art
spearheaded by Norval Morrisseau.
Page 386
(\
Time affects both the physical size and condition of an image and can totaily alter the
style in which the image was originaily created. Ultimately, the image cailed a blob exists
regardless of the original style of the images. If an image was hoilow or constructed of lines,
its shape might become unrecognizable faster than if it were solid. But ail these images with
different centres are found co-existing in the same geographical region (see Rajnovich
1981b).
Ideaily specific styles of images could be dated, enabling stylistic sequences to be
established, and providing the archaeologist with the tools to determine the order in which
the sites were painted. Such information would en able the archaeologist to consider and
confirm speculations regarding whether images were placed in groups, or one at a time
ultimately creating the effect of a group of related images (see Rajnovich 1980b). Indeed,
secure dates for specific styles enables the archaeologist to establish the time sequence in
which images may have been superimposed on one another if images were repeatedly over
painted at sites. Secure dates for specific styles of different images would sanction the
archaeologist to contrast and establish any offerings left at a site that could be dated with the
style of the images at the same sites or groups of sites.
A detailed examination of the previous field recordings of each site reveals that
considerable developments and changes have occurred in the techniques used to record
these sites (see pages 136 to 161 of Chapter III). Any changes in the size of the same images
could reflect these developments rather than anything else. Furthermore previous
archaeologists have differed in their opinion as to the importance and recognition of
different images (as evident in Chapter III). However, a detailed critical comparison of the
field drawings, photographs, and reports was difficult to conduct, since it was impossible to
obtain a complete set of information.
Conclusion for the chapter:
This chapter has presented and discussed the information that should be coilected in
the culture-historical approach. Ail the information discussed in this chapter is summarised
and presented in Appendix 5. Many of the sites are affected by the deposition of white
mineraI deposits, exfoliation of the rock's surface, lichen and rock tripe growth and
encroachment. Hence, large numbers of images are blurred, obscured, and almost destroyed.
Ail the images probably once had clear, distinct edges but they are ail ultimately becoming
Page 387
blobs before they are completely destroyed. These images have, like the rock surface on
which they exist, a life-cycle. All of these processes cause difficulties for anyone describing
these images. Analysis of the images using Adobe PhotoShop and VIPS indicated the fragile
nature of the images, the sites, and their changing nature. It also demonstrated the major role
played by the human eye and the brain in the identification of the individual images.
Relatively few sites exist in the southern and the northern portion of the Lake of the
Woods region. Two sites exist in the northwestern portion of the region but sites do not
exist in the southwestern region. This is possibly a consequence of the fact that the geology
is different, the land is lower than elsewhere in Lake of the Woods, and hence this part of the
lake was probably the most severely affected as the water levels were artificially raised from
1888 onwards. It is therefore difficult to determine whether trus distribution is connected to
the fact, raised in Chapter II, that in the early seventeenth century the Algonquian-speaking
peoples seasonally resided in the eastern portion of the Lake of the Woods and only
gradually took control of the western portion as the Assiniboine srufted westwards in the
middle of the eighteenth century (Heidenreich in Harris 1987: Plate 37).
Section 2 of trus chapter established that a total of three hundred and twenty-four
images were recorded. The average number of images per site is twelve and the minimum
number of images per site was one. DhKm-19, DiKm-SO, and DhKn-6 have only one image
while DhKm-S has the largest number of recorded images, trurty-seven. Deposits of white
mineraIs, lichen and rock tripe encroachment, and exfoliation of the rock surface takes place
over time and results in the destruction or fading of images. These images, or shapes, are
neither static nor timeless. A relatively small range of image types is found among preserved
images. However, given that these images are on rocks exposed to the elements and start to
deteriorate and often begin changing shape soon after they were created, it is impossible to
establish the full range of combinations of images that once may have been present. The few
types of images found to exist in the Lake of the Woods are a, contrasts sharply with the
popular belief that a relatively large vocabulary of images was in use.
Section 3 of trus chapter demonstrates the great difficulty 111 classifying and
describing any image and that the level of description given to an image affects its analysis. It
is possible to classify the images differently, and the vocabulary used heavily influences the
possible outcome of any discussion of perceived style. It is important to take the middle road
between the "splitter" and the "lumper" approach when describing images. A nested
Page 388
hierarchy of types leaves more options open with respect to classification. The presence of a
large range of different shapes that occur only once causes me to speculate that a
considerable number of people probably moved through this region and used these images
to convey information. It suggests that these images were part of the larger vocabulary of
signs used by everyone and not solely the product of specialised religious practitioners, such
as the Midé. This makes it more likely that each image was created independently and was
understood separately from the other images at a pictograph site. Tt is possible that the
organization of images and the combination of images does not reflect any particular set of
rules, other than the fact that individuals used the same place over a period of an unknown
time. Perhaps the variability in the form of images reflects a lack of standardization. It is also
possible that this variability also reflects the fact that the images changed over time.
It is impossible to precisely determine either the implements used to apply the paint
or its ingredients. Tt is possible to posit that fingers were probably used to apply the paint.
Ethnographic evidence supports the ide a that iron ore, ochre, was used and possibly isinglass
from the air bladder of the sturgeon, a fish that lives in this region of North America.
Not ail the pictograph sites had offerings when they were examined in 2001. The vast
range and quantity of offerings Ieft at the sites suggest that the physicai locations and the
paintings were and are important. Most of the offerings that have survived were Ieft in cracks
or crevices or dropped into the water in front of sites. Ethnographic evidence provided
additionai support regarding the importance of these offerings (Densmore (1974 [1928]),
Hilger (1936), Coleman (1947), Warren (1984 [1885]), Smith (1995), and Pufahl (1990).
Dating these images remains a problem, since direct dating techniques cannot be
used. The presence of datable offerings at a few of the sites provided an indicator of the
length of time, or periods in time, in which these sites have been used. But they did not
provide a strong indicator of the age of the images. Indeed, it remains impossible to establish
the precise date at which ail but the most recent of these images were created. Dewdney,
who first recorded many of the sites, relied upon reports of the sites from informants who
often did not give any indication of the length of time these sites were believed to have
existed prior to them being "discovered."
It was very difficult to use lichen growth as a measure of age since the rate of growth
appears to differ from site to site and it is impossible to establish how many visitors,
researchers, and archaeologists foilowed Dewdney's practice of removing the lichen that
Page 389
coyer many of these paintings. This prevents establishing a baseline for each site that would
ailow archaeologists to determine which images existed first and which ones were
subsequently added.
Whether the images present upon the pictograph sites have changed over time is
impossible to determine. The great variability of images suggests that doing this stylisticaily
will be difficult if at ail possible. This method is problematic given the incomplete
documentation regarding these sites in combination with the changing methods used to
record them.
Although it is impossible to establish the exact age of each site, it is possible to
conclude from the presence of blobs at sites throughout the Lake of the Woods that some of
the images were created some time ago. VlPS software and the Adobe Photoshop 6.0
provided me with the ability to establish whether additional paler images existed beside the
readily visible images. Both highlighted the fact that it was important to look beyond the
areas of the surface of the rock that had strong, clearly defined, red images since paler images
could exist elsewhere on the rock surface. The presence of both the pale and the dark red
images in ochre as weil as the modern spray painted images indicates that these places were
and are being used.
Observations regarding the distribution of different shapes, different combinations
of shapes, the association between the five required physical traits with images and any other
possible patterns wiil be examined in the next Chapter.
Page 390
Chapter VI: Contextual Approach
This approach connects the culture-historical information discussed in the previous
chapter with the distribution of the images found at each site and the natural features of each
site. The previous chapter applied the culture-historical approach to the data collected in
2001 and established the range of the images, their possible age, and their location. A nested
hierarchy of types was believed to leave the most options open. Individual images were not
static but changed over time, since they were subject to the accumulation of white mineraI
deposits, lichen and rock tripe encroachment, and exfoliation of the rock surface.
This chapter applies the contextual approach to consider a broader set of
associations and relationships regarding the generallocation of the site, between the images
and rock surfaces, and among the images themselves. At this level l offer low-Ievel
generalisations from my data and do not search for explanations of what these data might
mean. This chapter also considers previous research regarding these same issues. The first
section discusses the physical location and physical features of the sites in the Lake of the
Woods, the second section considers the relationship between the images and the rock
surfaces, and the final section examines the images themselves and considers whether
relationships occur between each shape. (The same data was also examined for clusters using
SPSS and the results were presented as dendrograms in Appendix 16.) Shapes occur on cliff
faces and inside caves; however, this difference in location is only important for section two
which deals with the issues concerning the images and the rock surfaces.
Section 1:
(a) Possible correlation between the geology, the geomorphology, and the type of
land mass on which the sites exist.
Fieldworkers accept that pictograph sites occur on granitic cliff faces throughout the
Canadian Shield. This belief about their physicallocation has fundamentally affected the way
in which archaeologists have surveyed them. This supposition presumes that the geology and
the geomorphology of each locale, of each site, is the same. It also presumes that the
landscape is the same.
Chapter V established that the Lake of the Woods sites were on Pre-Cambrian
granite on the northern and eastern portion of the Lake of the Woods, but the geology and
Page 391
geomorphology of the sites are quite complex (see Appendix 8). If the geology of ail of these
sites is examined, it is evident that twenty-five of the sites are on rocks identified as
hornblende, gabbro, diorite, periodite, or serpentine (see Appendix 8). DiKp-1 and DhI(m-
19, in contrast, were found on schistose volcanic rock with smail amounts of iron bearing
sediments. Considerable quantities of volcanic rocks and younger sediments, according to
Tanton (1939), also occur at both sites.
Most of the sites described in Appendix 7 share the same general geomorphology.
Twenty-four are in areas where more than eighty percent of the rock is outcrop composed of
ice eroded Archean granite with patches of thin drift and scattered boulders. The
geomorphology of DjKn-1, DkI<n-6, and DkI<n-7 is not the same as that of the other
twenty-four sites, although they have the same geology as the majority of the sites (Appendix
8).
A strong correlation cannot be established between the geology and geomorphology
of these sites. Nether is there a strong correlation between the geomorphology, geology, and
the type of landmass of each site (see Appendices 8).
(b) Possible correlation between the main travel routes and the physicallocation of
the sites
It is important to consider whether the physicallocation of each site is connected to
mam tracks, or travel routes, through the region; and the significance of any such
relationship. These tracks, the recommended routes indicated by either red continuous or
discontinuous lines on the charts produced by Fisheries and Oceans Canada, indicate the
safest route though the islands, reefs, and shailows from the twentieth century onwards.
These thoroughfares are used by sailors and boatmen, as discussed in Chapter IV, to travel
safely through this lake. These routes ailow relatively inexperienced people with little
knowledge or no knowledge of the region to move from one part of the lake to another.
These routes use the deepest part of every region of the lake and avoid lee shores (shores
facing the prevailing winds).
Although these routes may weil be the best ones for motorised boats, they are not
necessarily as suitable for canoes or sail boats. Travelling by canoe or by sail requires a
different range of techniques and methods. Both types of vessels are affected differently by
the wind, the water, small local rapids, strong local currents, and changes in the weather.
Page 392
People travelling by canoe would have preferred travelling in areas protected by islands and
large land masses, since the water and wind action would have been reduced. Historically this
region was explored by canoe and not by motor boat, although people travel through this
region by motor boat today.
Canoeists crossed the Lake of the Woods using routes protected from strong winds
and open water where wave action would have been strong enough to easily swamp and
capsize canoes. This could explain why two routes existed through this region from the fur
trading posts in the Rainy Lake region southeast of Lake of the Woods to Rat Portage at its
north end. One route went through Whitefish Bay, east of the Aulneau Peninsula to Rat
Portage. The second, longer route ran west of Aulneau Peninsula via the Big Traverse, which
is the large body of open water in the southwest portion of the Lake of the Woods. Those
who visited and created these rock image sites knew how to manoeuvre and travel in and
around as weIl as through the Lake of the Woods. Chapter V established that the majority of
the sites are located in the northern and eastern portion of the lake and it is clear from
marine maps of the region that they are not always near the modern major thoroughfares.
Chapter V established that two sites were known prior to Dewdney's recordings of
the 1960s: DhKo-1 and DiKp-1. Lawson first recorded DiKp-1 in 1885 (1885), and,
although DhKo-1 was first recorded in 1936, aIl the boat captains who worked on Lake of
the Woods in 1910 knew of it. Neither site is beside a main track used by motorised boats
(see Appendix 9). But both sites were visited, otherwise no one would have reported their
presence to individuals such as Dewdney. Campbell's observation, discussed in Chapter V,
that she found religious slogans over the images at DhKo-1 when she revisited the site in
1951 confirms that previous knowledge.
The introduction of motorised boats in the 1870s changed the method of travel
through this region; now it could be travelled and explored without total reliance on
manpower and dependence on marine weather conditions. In 1872, the first motorised boat,
a small steam launch or tug with a "3 or 3 % feet draft," (1.07 metres) left from Hungry Hall,
the Hudson's Bay Company post on the mouth of Rainy River, and crossed the southern
part of the lake to Northwest Angle Inlet on the southwest si de of the Lake of the Woods
(International Joint Commission 1917: 195). Two more boats built by the Canadian
Government especially for this route (ibid.) repeated this journey in 1873. The government,
mining and lumber companies, and private individuals built and launched boats of different
Page 393
/- ...
sizes and types from 1876 onward for transporting freight and passengers, for towing logs,
and for the fisheries. How the raised water levels affected the co mm on travel routes in the
region is impossible to determine, but they were affected.
Some sites lie beside or near the main modern thoroughfares through the lake. lt is
possible that the physical location of new sites such as DhKm-19 and DhKm-20 may be
connected to several changes in the routes, the type of boat, the physical location of the
previously known sites, and possibly even the changes in the water levels discussed in
Chapter IV. However, these links cannot be established.
(c) Possible correlation between the local reserves, the offerings left, and the sites
The presence of offerings could be linked to the proximity of Indian reserves to sites
in the Lake of the Woods. Trurteen, or 48%, of the pictograph sites are close to an Indian
reserve (xq091). Using maps of 1:250 000, 1:80 000 and 1:40 000, and assuming that
Algonquian-speaking peoples left the offerings that archaeologists found, l established wruch
reserves were closest by boat to each site. The reserves in trus area were set up after the
Algonquian-speaking peoples signed Treaty #3, the North West Angle Treaty, in October
1873. If pictograph sites were located near reserves, then offerings should be present. The
word 'near' implies that site is within 6 miles, or 5.21385744 nautical miles, or 9.6506404
kilometres by boat from the nearest reserve. The journey would only take between 15 to 20
minutes by boat. The data in Appendix 9 indicated that DgKI-17, DgKl-2, DgKl-1, DjKr-23,
DhKn-1, DiKm-4, and DhKo-1 are located near an Indian reserve and had offerings in
2001. But DgKm-2, DhKm-19, DgKl-19, DgK.m-1, and DiKq-10, are also near a reserve,
but they had no offerings in 2001. Although, DhKm-20, DiKn-1, DhKm-1, DhKm-3,
DhKm-5, and DhKn-1 were not near a reserve, offerings were discovered at these sites in
Table 6. 1: Hierarchical Cluster Analysis - dendrogram using Average Linkage (Between Groups) using Euc1idean distance method to establish whether a re1ationship exists between the sites and the different features.
CAS E LaIle1 DiRq-1O
DqRrn-Z
DqRl-19
DgRrn-l
DhRm-19
DgRl~17
DjR"-23
DgRl-1
DgRl-2
DkRn-6
DkRn-7
DbKJn-4
DbKJn-18
DiRrn-3
DiRrn-51
DiRp-l
DbKJn-1
DbKJn-5
DiRn-l
DiRrn-4
DhRo-l
DiRrn-50 DiRrn-49
DhRn-l
DbKJn-3
DbKJn-20
DjRn-l
Rescaled Distance Cluste:t: Combirle o 5 10 15 20 25
Table 6. 2: Hierarchical Cluster Analysis - dendrogram using Average Linkage (Between Groups) using Pearson correlation method to establish whether a re1ationship exists between the sites and the different features.
Page 401
(f) Possible correlations between the physical features of the rock surface of
pictograph sites
Previous archaeologists, basing their daims on ethnohistorical and anthropological
sources, have asserted that key features of cliff faces are prerequisites for paintings in the
Canadian Shield (Rajnovich 1994: 11; Rusak 1992a: 23; Vastokas and Vastokas 1973: 47-54).
Caves will not be induded in this examination since their physical structure is radically
different. The key features of rock image sites on cliffs indude: deep holes, large cracks, large
crevices, rock overhangs, and rock benches or ledges. l examined the pictograph sites on
cliffs to ascertain the presence of these physical features at each site presented in Appendix
12. Not all sites possessed every physical feature deemed "crucial", or "required" by previous
archaeologists, as data in Appendices 12 and 13 reveal. Sites often had a minimum of two of
the five physical features, sorne of which occasionally affected the paintings themselves.
Each physical attribute is examined in sequence.
(i) Deep holes:
DgKI-2 and DkKn-6 had the physical feature called a "deep hole" in 2001. It remains
undear from the literature reviewed in Chapter III whether the size of the hole is important.
The key idea is that it is deep and part of the site. The hole at DkKn-6 is smaller than the
holes at DgKl-2. DgKl-2 contained two holes above the water level, with an additional,
smaller hole at the back of the larger hole indicated in the photograph below (figure 6.2).
Divers who assisted me found a third hole to the right and 2.13 metres underneath the
surface of the water. The annotated photograph below indicates that these holes acted as
openings to two caves.
Page 402
Figure 6.2: Annotated photograph by Cols on of the holes which are called caves at DgKl-2 in the rock above the water level.
Figure 6.3 depicts the interior of the large cave at DgKl-2 indicated above in figure 6.2. An
adult could move about inside this cave.
Figure 6.3: Inside the cave at DgKl-2
Page 403
Figure 6.4: Map of the pictograph sites that have deep holes in 2001
The map (figure 6.4) indicates the physicallocation of sites with this feature using the letter
"H" beside the Borden number. Both sites are in the eastern portion of the Lake of the
Woods.
(ii)Large cracks:
The second prerequisite is the "crack". Twenty-three of the twenty five sites had
large cracks (xq132). The distribution of these sites in the Lake of the Woods is indicated in
figure 6.5 by the letter "C" beside the Borden number. DgKl-l and DhKm-19 have small
cracks (xq132). Large cracks affected paintings at sixteen of the pictograph sites: DgKl-2,
Table 6.3: Hierarchical Cluster Analysis - Dendrogram using Average Linkage (Between Groups) using Euc1idean distance method to de termine the nature of the relationship between smears and sites
Table 6.4: Hierarchical Cluster Analysis - Dendrogram using Average Linkage (Between Groups) using the Pearson correlation to de termine the nature of the relationship between smears and sites
Page 420
(b) Blobs
Blobs are caused by taphonomic processes and reflect the passage of cime. The tine
in figure 6.13 connects ail the sites that have blobs. This map also indicates that the majority
of the sites had blobs and that there were relacively few sites that did not have blobs. It is
important to remember that blobs are the decayed remains of shapes that once had clearly
defined edges.
Figure 6.13: Sites with blobs
Twenty of the twenty-seven sites had blobs: DgKl-2, DgKl-l, DgI<l-19, DgKm-l,
and xq753). Each 'creature' bearing site had approximately two such shapes. The map below
indicates that these images existed at sites in the southern, eastern, and northwestern parts of
the lake (figure 6.19). The line connects these sites.
Figure 6.19: Sites with creatures
Eight, or one third, of these shapes faced right, while thirteen, or half, of the
'creatures' faced le ft (xq382). The creatures that faced left were at DgKI-l, DgKl-17,
Page 427
DhKm-S, DhKo-1, DiKp-1, DjKll-1 and DhKll-1, while those that faced right were at
DiKm-3, DgKl-1, DgI<1-2, DhKm-3, DhKo-1, DjKn-1, and DhKn-1. Three shapes of
creatures faced neither 1eft nor right at DhKm-3, DiKm-3, and DhKo-1. The sites that have
creatures facing right together with creatures facing left (DgKl-1, DhKo-1, DjKll-1, and
DhIZn-1) are located east, northeast and south of the Aulneau Peninsula. Seven had hollow
centres, at DhI<m-3, DhKo-1, DjK-1, and DgK1-17. Three of the four 'creatures' at DhKo-1
were hollow; two faced left and two faced right. DhI<m-3 had two 'creatures.' One was
vertical, hollow, and faced neither 1eft nor right, while the other had a solid centre and faced
right. The creature shapes at DjIZn-1 were hollow but two of them faced right while the
other faced left. The sole hollow creature at DgI<l-17 faced left. Creatures often had horns or
ears. Nine sites located throughout the lake contained such images: DgI<l-l, DgI<l-17,
DhI<m-3, DhI<m-S, DhIZn-1, DhKo-1, DiI<m-3, DiKp-1, and DjIZn-1 (xq382), DjKr-23
and DiKq-10 in the northwestern portion of the lake were devoid of such examples. The
dendrograms in Appendix 16 indicate that a weak correlation exists between the shape called
'creatures' and the pictograph sites. 'Creature' shapes do not always exist with stick figures
(xq382), with hand print shapes (xq328, xq328b, and xq328a), or with crescents (xq348).
(i) 'Crescents'
These shapes possibly have been called 'canoes' by previous archaeologists. Thirteen
different shapes were categorised as 'crescent' at DgI<l-l, DgI<l-2, DhKm-S, DiKm-3,
DiI<m-S1, DiIZn-1, and DjIZn-l (xq348). These sites are in the eastern part of the Lake of
the Woods. These shapes had solid centres, but several different types of crescents were
identified (xq 348) (see Appendix 16 and figure 1.1). Occasionally, sorne of the shapes
loosely categorised as cres cents occurred on their own. Sometimes they occurred together.
Bach combination of images was unique. Six 'horizontal crescent with points facing upwards
with verticallines attached to topside' were found (xq332, xq348, and xq 350). Three of these
crescent shapes had the same shape; two existed at DgI<l-l and one existed at DiIZn-1 (see
Appendix 16 and figure 1.1). Ail three could be described as a horizontal crescent with points
facing upwards with three short verticallines attached to the topside of the crescent (xq348).
DgI<l-2 was in Sabaskong Bay and DiIZn-l in the northern part of the Whitefish Bay. Both
were located in the eastern and southern portion of the region. DhI<m-S and DiI<m-S1 each
had a horizontal crescent with points that faced downwards (xq348a and xq348). Both of
Page 428
these sites were ln Whitefish Bay. There are two 'horizontal crescent, points facing
downwards with verticallines on bottom side' (xq 350 and xq348) at DjKn-1. The images
were not identical, since one of the shapes had three short verticallines, the other, four short
vertical lines attached to the underside of the horizontal crescent. The dendrograms in
Appendix 16 indicate that a weak correlation exists between the sites and the shapes calied
'crescents'.
The 'crescent' does not appear with 'stick figure' (xq348, xq382b and xq382a), or
with the hand print shapes (xq328, xq328b, and xq328a).
(j) 'Turtles'
Three shapes can be described as turtles at DhKm-3, DhKm-19, and DhKn-1
(xq382 & xq362). (A discussion of what differentiates the shape of a turtle from the shape of
a 'bird' or a 'creature' is in Chapter IV.) The turtles at DhKn-1 and DhKm-19 were solid and
painted using ochre while the turtle shape at DhKm-3 was holiow. While ali three shapes
were termed as "turtles" ali of them were different (see Appendices 7 and 16). Both turtles at
DhKm-3 and DhKn-l were painted using ochre. The painter employed modern spray paints
at DhKm-19 (xq327 & xq362). Turtle shapes made with ochre always occurred with ochre
'creatures', while turtle shapes created in modern paints existed on their own (xq327 &
xq362). The creatures at DhKm-3 were a solid creature facing right, and a holiow one facing
neither right nor left (xq382). The creatures at DhKn-l were solid, and two of them faced
left. These 'creatures' at DhKm-3 and DhIZn-l had either horns, ears, or antlers. Both of the
turtle shapes painted in ochre occurred with a shape of a 'stick figure' at DhKm-3 and
DhIZn-l (xq382b). Each of these vertical stick figures had two arms and two legs (xq382b).
Each site is in a different part of the lake. DhIZn-l was in Sabaskong Bay, DhKm-3
was in Whitefish Bay, and DhKm-19 was in Turtle Lake, located between these two parts of
Lake of the Woods. Ali three sites were in the southeastern portion of the Lake. The turtle in
modern paint at DhKm-19 did not occur with any other shape, supporting Rajnovich's
(1980b: 34) speculation that shapes could exist on their own as weli as with other shapes at
sites. However, the ochre turtle shape at DhKn-l occurred in conjunction with a vertical
stick figure, various 'creatures,' and several smears. This is in comparison to the red ochre
turtle shape at DhKm-3, which occurs with a larger range of shapes.
Page 429
(k) 'Birds'
Two shapes loosely categorised as 'birds' are found at DhKm-3 and DhKm-18 in
Whitefish Bay in the eastern portion of the Lake of the Woods (xq33S). Both shapes of birds
appear in conjunction with vertical stick figures (xq382b). Despite both shapes called 'birds'
occurring at DhKm-3 and DhKm-18 in conjunction with other shapes, neither shape had
the same physical description (see Appendix 16).
(1) 'Group of three diagonal paraUe1 short tines'
Shapes described as 'a group of three diagonal parallel short lines' (xq346) are found
only at DgKl-17, in Sabaskong Bay in the southeastern portion of the Lake. Nine types of
shape exist. They occur with only one 'creature' (xqS02).
(m) 'One vertical and one horizontal short tine that cross each other at ninety
degrees'
Eleven types of the shape called 'one vertical and one horizontal short line that cross
each other at ninety degrees' occur at five sites (xq336 & xq336a), but they can be subdivided
into two general groups (see Appendix 7). The shape in figure 6.20 called 'one vertical and
one horizontal short line that cross each other at ninety degrees' occurred nine times at
DhKo-1, DjKn-1, and DjKr-23 (xq336).
Figure 6.20: An examp1e of a shape caUed 'one vertical and one horizontal short line that cross each other at ninety degrees'.
It is clear from the map in figure 6.21 below that this shape is found at sites in the
northeastern, northwestern, and southern portion of the Lake of the Woods. The line
connects these three sites together.
Page 430
Figure 6.21: Sites with 'one vertical and one horizontal short line that cross each other at ninety degrees'
This shape is not found with 'stick figures' (xq382b), 'crescents' (xq348), 'creatures'
(xq382), 'zig-zags' (xq 321), or hand print shapes (xq328, xq328b, and xq328a).
The second type of this shape, which was slightly different, was described as 'one
vertical and one horizontal short line cross each other at ninety degree angle with short lines
at the end of each of the vertical and horizontallines'.
Figure 6.22: An example of a shape called 'one vertical and one horizontal short line that cross each other at ninety degrees with short lines at the end of each of the vertical and horizontallines'
Two examples of this shape were observed, at DiKp-1 and at DhKin-3 (xq366a) in
the northern and the eastern portion of the region. A comparison of figure 6.21 with figure
6.23 demonstrates the radical difference in the geographicallocation of each type.
Page 431
~ .. Figure 6.23: Sites with the shape called 'one vertical and one horizontal short line cross each other at ninety degree angle with short tines at the end of each of the vertical and horizontallines'
This particular shape is found with 'zig-zags' and 'stick figures' at both sites. The
stick figures were vertical and had two arms down and two legs (xq382a). But neither of the
zig-zags were the same, since the zig-zag at DiKp-1 was horizontal while the zig-zag at
DhKm-3 was either diagonal or vertical (xq321 and xq365). The shape called 'one vertical
and one horizontal short line cross each other at a ninety degree angle, with short lines at the
end of each of the vertical and horizontal lines' is found with 'creatures' at DiKp-1 and
DhKm-3. Both shapes were solid, had curved tails, either ears or horns on their heads, a
back and a front leg, but the shape at DiKp-1 faced left and the creature shape at DhKm-3
faced right. The shape called 'one vertical and one horizontal short line cross each other at
ninety degree angle with short lines at the end of each of the vertical and horizontal lines'
existed in conjunction with 'crescents' (xq348) or with the shape called hand prints (xq328,
xq328b, and xq328a) (see Appendix 16).
(n) 'Group of four diagonal short lines'
Two types of shape called a 'group of four diagonal short lines' (xq342) co-existed at
DhKm-5 and DhKm-3 with 'creatures' (xq382). Both sites are in the southern part of
Whitefish Bay in the eastern portion of the Lake of the Woods.
Page 432
Figure 6.24: Sites with groups of four diagonallines
The shapes of the creatures were different (see Appendix 16). The shape 'group of four
diagonal short lines' did not occur in conjunction with shapes loosely categorised as
'crescents' (xq348), 'stick figures' (xq283b), or 'zig-zags' (xq321). It occurred twice with right
hand print shapes, at DhKm-3 and DhI<.in-S (xq328, xq328b, and xq328a), but not at the
other sites that also had right hand print shapes; therefore, it cannot be argued that right
hand print shapes occurred with the shape called 'a group of four diagonal short lines'.
(0) 'Group of six short horizontallines'
Two shapes called a 'group of six short horizontal lines' (xq343) were observed at
DhKo-1 and DhI<.in-3 in the southern and the eastern portion of the Lake of the Woods.
They oruy occurred with 'creatures' (xq382), and vertical 'stick figures' (xq 382a). One of the
vertical stick figures at DhI<.in-3 (xqS11) was exactly the same as one at DhKo-1, except that
it was not inside a hollow square (xqS1S).
(P) 'Two paralle1lines'
Four examples of 'two parallel lines' (xq 347 & xq749) which were alternatively
diagonal, vertical, or horizontal were identified at DhI<.in-4, DgKl-1, DiI<.in-3, and DjKn-1
(xq347). These sites are in the eastern, southeastern, and northeastern portions of the Lake
of the Woods. None of these shapes was identical because one of the lines in the vertical
version of this shape just touched the middle other vertical line at DjKn-1 (xq347) (see
Appendix 16).
Page 433
(q) 'Three Hnes joined together at one end'
One shape called 'three lines joined together at one end' was painted at DiKp-l and
DiKm-3 (xq324). DiKm-3 is on an island east of the Aulneau Penin sula in Whitefish Bay and
DiKp-1 is on an island north of this peninsula. Both sites therefore were in the eastern and
northern portion of the region (figure 6. 25).
Figure 6.25: Sites with shapes called 'three Hnes joined at one end'
This shape occurred with 'stick figures' at both sites (xq382b). The stick figures were not
identical since the stick figure at DiKm-3 was diagonal while the stick figure at DiKp-1 was
vertical (xq382b). It also occurred with 'zig-zags' at DiKp-1 and DiKm-3 (xq321), but
examination of the detailed description of each zig-zag shows that the zig-zags were different
(xq321) (Appendix 16). The zig-zag at DiKp-1 was horizontal with a round head that had
two horns while the zig-zags at DiKm-3 were long, diagonal, with a pointed end and a head,
or vertical with a round head with horns and a forked tail (xq321). The shape 'three lines
joined at one end' did not occur with 'crescents' (xq348), or with hand print shapes (xq328,
xq328b, and xq328a).
(r) 'Group of four diagonal Hnes'
This shape is at DhKm-3 and DhKm-S (xq342) in the eastern portion of the lake.
Both groups of four diagonallines occurred with the shape loosely categorised as a 'creature'
(xq382). Only one of the creatures at DhKm-3 vaguely resembled those of DhKm-5. The
shape at DhKm-3 had long horns or ears, faced right, had one back leg and one front leg, a
curved tail and was solid (xqS11). AIl the other shapes of creatures at this site were hollow.
Page 434
Ali the creatures at DhKm-5 (xq513) were solid and faced le ft; otherwise, they differed. This
shape did not occur with 'crescents' (xq348), 'stick figures' (xq323), 'zig-zags' (xq 321), or
hand print shapes (xq328, xq328a, xq328b).
(s) 'Group of six horizontallines'
The shape called 'a group of six horizontallines' is found at DhKm-3 and DhKo-l
(xq 343), in the eastern and southern parts of the lake. This shape occurred with vertical
'stick figures'. If the precise description of each stick figure is examined at DhKm-3 and
DhKo-1, neither shape was exactly the same (xq382b) (see Appendix 7). This shape was also
found with 'creatures' (xq382) but neither the 'creature' nor the vertical stick figure was
similar. This shape did not occur with 'crescents' (xq348), 'zig-zags' (xq321), or hand print
shapes of either type (xq328, xq328a, and xq328b).
(t) 'Group of nine short lines'
The shape 'a group of nine short lines' appears at DhKm-5, DiKm-3, and DiKn-1
(xq344). These sites are in Whitefish Bay in the eastern portion of the Lake of the Woods.
None is exactly the same shape (see Appendix 7). The shapes at DiKn-1 were vertical while
at DiKm-3 and DhKm-5 they were horizontaL Furthermore, the ninth horizontal short line
in the shape at DiKm-3 was very faint.
(u) 'Square, created by two vertical and two horizontallines of equal shape'
Five examples of this shape are found only at DhKo-l (xq746) in the southern
portion of the lake (see Appendix 7). Four of these squares were exactly the same, while the
fifth had a short vertical stick figure with two legs and two arms in the bottom left corner of
the square.
Summary for Section 3:
The large number of blobs (xq349) and five types of 'unidentifiable shape' (xq744)
reflect that the pictograph sites are undergoing change from use, reuse, and natural attrition.
It is impossible to establish the full range of combinations of images. Smears are larger than
the blobs. Their physical size suggests, as discussed in Chapter IV, that they resulted as a
consequence of someone placing a wash of ochre on the rock surface, since they are often
Page 435
larger than other shapes such as creatures, zig-zags, and crescents. They could also reflect the
passage of time as a large image may have faded, exfoliated from the surface of the rock face,
or become indistinguishable because of the accumulation of white deposits.
Shapes such as smears, blobs, and unidentifiable shapes cannot simply be ignored as
they once were. Dewdney (1958-1964) and Lambert (n.d.) were perhaps mistaken in ignoring
them and not perceiving them to be indicators of pictograph sites. Perhaps these images
were not perceived to be interesting or informative, by either archaeologist, from the
interpretive perspective but these are not valid reasons to exclude them. By 1991 Rusak
(1992b) had recognised them as being significant shapes at pictograph sites. The smears,
blobs, and the unidentifiable shapes reflect that people used these rock surfaces to express
something and the images they created have just altered with the passage of time. This
conclusion is supported by the presence of a large number of shapes that occur only once
(Appendix 7). Examples of several of these images are visible in figures 5.17 and 5.18. Sites
such as DjKn-l, DhKm-3, DhKo-l, DiKm-3, and DiKn-l had a large number of these
uniquely occurring shapes. It must also be remembered that many of the varieties of shapes
called 'creatures', 'zig-zags', 'crescents', 'hand prints', 'stick figures', 'birds', and 'turdes' also
only occur once or twice once they are examined and described in depth (Appendices 6, 7,
and 16).
It is possible to consider combinations of various shapes. Right hand print shapes
exist if a left hand print shape is present (xq328, xq328a, xq326b). Shapes that are hands that
are left and right co-exist with 'creatures' at DiKm-3, DjKn-l, DhKm-5, and DgKl-19
(xq382). Vertical 'stick figures' only existed with hand print shapes at DjKn-l and DhKm-3
(xq382a, xq382b, and xq323). The four sites with creatures facing right and left are: DgKl-l,
DhKo-l, DjKn-l, and DhKn-l (xq382). While the sites DiKm-3, DhKo-l, DhKm-3, DiKp-
1, and DjKn-l have 'creatures' and stick figures, only DhKo-l has 'creatures' with horns,
ears, or anders (xq382b) with 'stick figures' that had horns, ears, or antlers (xq382). Two
crescent shapes whose points face downwards exist at DhKm-5 and DiKm-51.
Ochre turtles are at DhKm-3 and DhKn-l 'creatures' as well as the same type of
stick figure which if it is described in detail is a 'vertical stick figure with two arms and two
legs' (xq382b). The shape 'one vertical and one horizontal short line cross each other at a
ninety degree angle with short lines at the end of each of the vertical and horizontal lines'
existed with vertical 'stick figures' and 'creatures' at DiKp-l and DhKm-3. Two types of
Page 436
shape cailed a 'group of four diagonal short lines' (xq342) exist with 'creatures' at DhKm-S
and DhKm-3.
Two types of shape cailed a 'group of six short horizontal lines' (xq343) exist at
DhKo-l and DhKm-3 with 'creatures' and vertical 'stick figures' (xq382a). The shape cailed a
group of four diagonallines occurs at DhKm-3 and DhKm-S (xq382) with 'creatures' that
have either horns, ears or antlers. The shape cailed a group of four diagonallines at DhKm-3
and DhKm-5 is associated with right hand prints and a 'creature' (xq382). Both of these
'creatures' have either horns, ears, or antlers.
Two shapes were painted by themselves at a site. A turtle shape existed on its own at
DhKm-19 and a smear on its own at DiKm-SO. The existence of these individual images
supports Rajnovich's (1980b: 34) speculation that images could occur on their own as weil as
with other images at sites. Rajnovich (1980b: 35) also observed that paired images sometimes
occurred but only one pair of images was discovered at one site in the Lake of the Woods at
DgI(l-l. Here two shapes existed side by side which were described as a horizontal crescent
where three short vertical lines attached to the topside of the crescent (xq348). It is very
difficult to uncover combinatory 'rules' because of the smail dataset that 1 have at hand and
since many of the shapes have decayed beyond recognition becoming blobs. Indeed, the
most frequent shape identified is a blob (see Appendix 16).
Were the images created by professionals, specialists in these images amongst the
Algonquian-speaking community, or by amateurs? The creators of these images were
professionals in the sense that it is likely that they were trained in the creation and the use of
the images. This is a difficult question to answer with regard to the images created in ochre.
Yet, the shapes at DhKm-20 and DhKm-19 in modern paints so strongly resemble those
created by the members of the Woodland School of Art, founded by Norval Morrisseau
during the mid twentieth century, that the painters of these images were professionals or
inspired by professionals. This assumption is supported by the presence of the signature of
the painter and the date at the base of the images at DhKm-20. Unfortunately, it is
impossible to make any assertions regarding the ochre images since evidence does not exist
regarding their authors.
Sorne images occur in sorne geographical regions of the Lake and not others. Blobs
clearly occurred throughout the entire lake but smears were mainly found at sites in the
southern, eastern, and southeastern portion of the lake. The largest number of different
Page 437
shapes were found in the eastern portions of the lake. Here seven different shapes existed: 'a
group of nine short lines,' 'a group of four diagonal short lines,' 'a group of four diagonal
lines,' 'birds', 'crescents', and 'zig-zags, but diagonal zig-zags were only on DiKm-3 in the
same region of the lake. Two shapes are evident throughout the eastern, southeastern, and
the northeastern portion of the Lake of the Woods. These shapes are 'two parailellines' and
the 'hand print'. Two shapes existed in the northern and eastern portion of the lake, the
'three lines joined at one end' and 'one vertical and one horizontal short line cross each other
at ninety degree angle with short lines at the end of each of the vertical and horiz0!l-tallines.'
Two shapes existed in the southern and the eastern portion of the lake, 'a group of six short
horizontal lines,' and 'a group of six horizontal lines.' Two shapes occurred only in the
southeastern region of the Lake of the Woods: 'a group of three horizontal short lines', and
'turtles.' 'Unidentifiable shapes' existed at sites in the northwest and the southeast region of
the lake. 'Creatures' occurred in the southern, eastern and northeastern portions of the lake.
Hand print shapes occurred throughout the southeastern, eastern and the northeastern
portions of the region. The shape 'one vertical and one horizontal short line that cross each
other at ninety degrees' is found throughout the northeastern, northwestern, and southern
portion of the lake. The 'upper torso of vertical stick figure' occurred twice at sites in the
eastern and the southeastern portion of the Lake. The shape cailed 'a square, created by two
vertical and two horizontallines of equal shape' occurred in the southern portion of the lake.
As discussed in Chapter V there were many shapes that only occurred once and they were
predominately at rock image sites in the southern, southeastern, and eastern portion of the
lake. Only a few of the shapes occurred elsewhere in the region and they were either at
DjKr-23 in the northwest or DjKn-l in the northeastern parts of the lake. Only two
pictograph sites, DjKr-23 and DiKq-l0, existed in the western and the northwestern potion
of the lake. Petroglyph sites tend to be more numerous in the northwestern and northern
region of the lake. When the northeastern region was examined it is important to remember
that DjKn-l was the only pictograph site where the images were not ail described as blobs.
Sorne shapes appear in sorne parts of the lake and not in others. Perhaps sorne of the images
in the northeastern and northwestern pictograph sites occur in the petroglyph sites of the
northeastern and the northern part of the Lake of the Woods. This speculation can only
confirmed once these sites have been examined and the images described in the same
manner which the pictograph sites were examined.
Page 438
Conclusion:
This chapter sought to discover any relationships between the physical existence of
the sites in the larger landscape, the physical features of the rock's surface and the images
themselves. It was important ta establish any correlations and patterns ta provide a solid
basis for considering the meaning of these images.
Pictograph sites in caves and on cliffs should be perceived as different physical types,
Slnce each has their own characteristics. For other purposes, they must be examined
together. The lack of detailed correlation between the geology, the geomorphology in the
immediate vicinity of the sites and the landmass implies that another reason existed for using
images. The results provided do indicate that granite cliffs did not need to have large or small
crevices, cracks and holes for painting to take place. These results are supported by the
dendrograms presented in Appendix 15. The creators of these images perhaps had another
important reason in mind, such as some type of supernatural connection. In section II, 1
established that a link did not exist between the images and the rock surfaces, between the
number of images at sites and the type of land mass, the number of the physical features
present at cliff faces, whether offerings were found in 2001, and the number of cliffs used at
each site. The lack of positive correlations between the number of images and these other
features, especially o fferings , is interesting, since it suggests that other factors and reasons
explain why offerings were found at some of the sites in 2001.
Section III examined the images to determine any relationsrups between them. The
widespread distribution of the blobs (xq326) and of the five types of 'unidentifiable shapes'
(xq744) throughout the geographical area containing pictograph sites indicates that the sites
have been used for a long period of time because these shapes are produced by decay. The
large number of individu al images occurring once, such as those in figures 5.16 and 5.17, the
large variety of 'creatures,' 'stick figures,' 'zig-zags,' 'crescents,' 'hand prints,' 'turtle,' and
'bird,' and the large number of blobs, suggested that the people who used these images had a
large repertoire of forms from which to draw. These shapes reflect that people have used
these rock surfaces to express sometrung, and that time has passed since they were initially
created. The presence of a large number of unique images supports this idea.
The number of similar images within the large number of shape types is relatively
small and makes it difficult to identify combinations of specific shapes. The number of
shapes at a site ranges from one to trurty-seven. DhKm-19, DiIZm-50, and DkKn-6 had only
Page 439
one shape, while the largest site, DhKm-5, encompassed thirty-seven images. Blobs were the
most common type of shape. Combinations appeared if shapes were loosely described but, if
the shapes in question were described in detail, relatively few image combinations existed at,
and between, sites. This information is confirmed by the analysis and the dendrograms
presented in Appendix 16. A very weak correlation occurs between the physical location of
the sites, its physical attributes, and the shapes that occur in large numbers: 'creatures', 'zig
zags', 'cres cents', 'blobs', 'smears', 'parallellines', 'birds', 'turtles', 'hand print', and the shape
called 'three horizontallines'.
Sorne images occur in sorne parts of the lake and other types in other parts of the
lake (see Appendix 16). Blobs are decayed images, and their widespread prevalence
throughout the lake demonstrates that blobs rather than smears may be more important
indicators of the passage of time. Smears were found at pictograph sites in the southern,
eastern and southeastern portions of the lake. This difference in geographical distribution
between the two shapes suggests that perhaps sorne smears were placed intentionally at sites
and do not always reflect the passage of time. The largest number of shape types occurred in
the eastern portion of the lake, where seven different shapes are found: 'a group of nine
short lines,' 'a group of four diagonal short lines,' 'a group of four diagonal lines,' 'birds,'
'crescents,' and 'zig-zags.' Diagonal zig-zags were found only on DiKm-3 in the same region
of the lake. 'Creatures,' which constitutes the large st number of any type, were painted in the
southern, eastern, and northeastern portions of the lake.
l will consider the meaning of these patterns and observations in the next Chapter,
where l use the homological approach to consider how to establish the meaning of these
patterns, and their relationship to other images outside the Lake of the Woods rock image
sites.
Page 440
Chapter VII: Approaches Concerned with Meaning.
This chapter discusses the meaning of the sites, and of the images themselves, in
relation to their physical location. l explored the ide a of considering the images individually
or as a group forming a narrative, with an aim to determine whether the images at pictograph
sites should be viewed as iconic or as narrative. l undertook a detailed examination of four
birch bark scrolls, using the same techniques l applied to the pictograph sites, to establish
whether images were understood as groups or individual elements. The birch bark scrolls
served as an illustration to counterpoint the rock image sites because the Algonquian
speaking peoples made and used both.
The first section briefly discusses the two approaches considered use fuI for
establishing meaning, and establishes which is the most appropriate for analysing the images
on the pictograph sites. Section two provides a detailed examination of four birch bark
scrolls whereby l applied the same methods and techniques of analysis using the culture
historical, contextual, and homological approaches. The goal of this small study was to
establish whether the images of the birch bark scrolls are similar to those of the rock image
sites, and if they provide dues as to how the images of the pictograph sites can be discussed.
Section three discusses efforts to consider rock image combinations as stories.
Present day indigenous Algonquian-speaking peoples are interested in these images.
Some of the images at several of the sites have been re-interpreted. This indicates that those
who currently live in this region have integrated these archaeologically and historically
important places into their mental and physical landscape. The fourth and final section of
this chapter discusses the notion that new images have been added to the graphical
vocabulary of the pictograph sites. Seven shapes were painted using modern paint but four
of the new images in modern paint were new vocabulary in the Lake of the Woods rock
image sites.
Section 1: The two useful approaches concerned with meaning
The literature reviewed in Chapter III indicated that archaeologists are more
interested in assigning meanings or explanations to the images themselves or to groups of
images than in studying rock image sites and their relationships. The majority of
archaeologists working on rock image sites of the Canadian Shield, in the vicinity of the Lake
Page 441
of the Woods, have resorted to the homological approach. However, as argued before, there
are two use fui approaches for establishing the meanings of these images: the homological
and the analogical approaches. They can be employed individuaily or in combination.
Therefore, it is important first to briefly outline again and reiterate what each approach
implies for the user.
(a) Analogical approach:
The wide spread use of the analogical approach, associated with New Archaeology or
the processual approach, was a reaction by positivists against the culture-history approach
during the 1960s. Those who foilowed this method of analysis argued that behaviour was
predictable from mate rial culture because of the many uniform connections between the
various components of socio-cultural systems, material culture, and human behaviour. As
Lyman and O'Brian (2001: 303) observed, the use of general comparative analogies required
application of the theory of orthogenesis. Yet general comparative analogies were not new in
the 1960s since they as weil as homologies, (specifie historical analogies), had been used to
examine the archaeological record of North America during the late nineteenth and the first
part of the twentieth century (ibid.: 304).
Biologists defined analogies as similar features of different species that do not have
close evolutionary relationships. Similarities have resulted from natural selection operating to
adapt different species to a similar environment (Abercrombie, Hickman & Johnson 1985:
20). An archaeological analogy in the narrow sense is a likeness or partiallikeness assumed to
exist as a consequence of convergent development under comparable conditions.
Interpretation using analogies ailows scholars to use strong cross-cultural regularities
between behaviour and material culture in systemic contexts to attribute behavioural
correlates to mate rial remains recovered from archaeological contexts. Binford maintained
that ethnographie analogies are useful sources of middle-range generalisations (Binford 1977:
288). An analogical interpretation is based on the existence of strong functional correlations
between specifie behavioral aspects and specifie aspects of material culture. Given such a
correlation, the presence of similar material culture in the archaeological record ailows an
archaeologist to presume that similar or associated behaviours existed in the pasto Watson,
LeBlanc and Redman (1971: 51) argued that analogies provided archaeologists with the
"richest sources of hypothesis" available.
Page 442
/~.,
What types of bridging arguments do scholars use for justifying an analogical
interpretation? Those who adopt an analogical approach conclude that it is only worth
employing univers al regularities in human behaviour. They presume that correlations exist
between past and present day cognitive and behavioural capabilities of human beings. If
similar behavioural characteristics can be established between specifie aspects of material
culture and behaviour in the modern (present day) world, they can be extended to cover the
same or similar aspects of material culture in the archaeological record (Binford 1981).
Scholars adopting analogical approaches use universal generalisations rather than concepts
specifie to individual cultures or historically related cultures. It is impossible to include the
idiosyncratic facets of an historical tradition if one considers only universal correlations
relevant. This approach utilizes the results of different developments that, because of
functional correlations, are similar regardless of their cultural origins or the environmental
setting within which they developed.
Does the analogical approach have anything to contribute to the Lake of the Woods
data? Related to a positivist and universalising view, this approach does not provide an in
depth understanding of culturally specifie meanings, including any changes in meanings.
Perhaps sorne universals exist with respect to meaning because human beings have had the
same neural network since the Aurignacian period, but the development of human
physiology, ideas, and objects is subject to numerous, unpredictable changes through time.
Univers ais seem more prevalent in human behaviour than in belief systems. This makes
using an analogical approach and discovering culturally idiosyncratic belief systems in the
archaeological record difficult. Meanings and the reasons attributed to artefacts clearly
depend on the ethnie, social, cultural, religious, and geographical identities of the people
involved. Scholars also must be aware of the dangers of extrapolating general analogies from
specifie bodies of ethnographie evidence without considering the specifie contexts within
which that evidence was created or applied. This approach could be useful if drawing on a
similar body of data produced by the Algonquian speaking peoples proved impossible.
Fortunately, a large body of data is available for analysis from which correlations can be
drawn between the images used elsewhere on objects made by the Algonquian-speaking
peoples, the images on the birch bark scroIls, and those on pictograph sites.
Page 443
(b) Homological approach:
An archaeological homology is a similarity in two or more cultures that occurs as the
result of a shared historical origin unobscured by cultural drift or adaptation to different
cultural environments. Archaeological homologies differ from those in biology, because they
can result from both diffusion and common descent. In biology, reproductive isolations
mean that homologies can occur only among species that are descended from a common
ancestor. Consequendy, homological similarities imply evolutionary relations among
organisms (Abercrombie, Hickman, & Johnson 1985: 145). Scholars establish interpretations
using homologies by demonstrating and tracing cultural continuities through time in a single
cultural tradition or a series of historically related cultural traditions. Interpretations
employing homologies frequendy also resort to the direct historical approach. This approach
identifies parallels between culturally specific beliefs and their material expressions du ring the
early historical period and employs material culture to trace these beliefs back into prehistoric
times. The bridging arguments for establishing homologies between beliefs or behaviour and
material culture, and between the present and the past are culturally specifie. Homologies,
asserted Watson, LeBlanc, and Redman (1971: 50), were advantageous in regions with strong
cultural continuity, and where the same techniques and implements were utilised for a long
period of time. Written records, oral heritage, and ethnographic materials, always combined
with archaeological data from successive periods, are probably the strongest types of
evidence to be used when devising bridging arguments (frigger 1995: 452). Scholars exploit
these materials to establish which beliefs were present in specific cultures.
The direct historical approach connects archaeology with other disciplines, providing
information concerning the practices and beliefs of either specific societies or groups of
historically related societies. By relying on a wide variety of evidence, archaeologists can
attempt to establish continuity in the practices and beliefs in question. The homological
approach allows in-depth analysis of a specifie group and its material culture. It permits
archaeologists to consider what human beings might have thought about in the past and the
meaning of specifie symbols and objects. It may be possible to determine that similar images
used in a variety of different mediums held similar meanings. Yet, continuity of form does
not necessarily imply continuity of meaning. Meaning, too, can change, especially since icons
tend to be polysemie.
Page 444
Archaeologists using homological interpretations must have a detailed understanding
of the skills required to use archaeological, historical, and ethnographie materials from the
region in question. These materials are multidisciplinary in nature. Those taking this
approach must verify how entrenched beliefs are established in cultures and how susceptible
they are to evaluation in terms of that culture. Any archaeologist using the homological
approach must be aware of different cultural perspectives and acknowledge that using the
direct historical approach is neither straightforward nor simple. Material symbols may
continue to be used but their meaning is not necessarily inert. Material symbols can acquire
new meanings and become an example of what Davis (1992: 25) called an "iconographie
disjunction." Continuity or discontinuity in beliefs over time is not correlated in a simple
manner with material culture. An example of such a symbol is the swastika which has had
several radically different meanings to different peoples.
It is important to choose which approach to use. Since 1 am more interested in this
group of images' symbolism and meaning than in inferring technological processes,
subsistence patterns and numerous aspects of social and economic behaviour, the
homological approach is more appropria te. This approach is also more advantageous in
regions such as the Lake of the Woods, where cultural continuity is strong.
This approach places a heavy reliance on written materials to create bridging
arguments. Unfortunately, this study suffers from a fundamentallack of detailed historical or
ethnographie information regarding the meanings of the images that occur on cliff faces and
inside caves.
Many analysts maintain that specifie places in the landscape were used as pictograph
sites. Rajnovich (1994: 34) argued that according to the Ojibwa these images often occur on
specifie rocks in the landscape in which the medicine manitous, or spiritual grandfathers are
thought to live. Vastokas and Vastokas (1973: 47) observed that "aIl remarkable spots in the
country" were believed to be the "favourite haunt of the spirits." Physical features, such as
"boulders, rock hills, and outcroppings with unusual dimensions or character, such as clefts,
holes, or crevices," served as dwellings for manitous and mythological creatures (ibid.: 48).
Hallowell (1936: 47), Jones (1981a: 72), and Wheeler (1977b: 710) argued that the Ojibwa
believe that, after consulting manitous that lived in the rock, shamans created the images,
either to record a vision que st or as lessons to others. The rocks marked the places where
holes or openings penetrate the surface of the Earth, which consisted of several layers,
Page 445
through which supernatural beings could pass from one world to another. These holes
served as meeting places connecting the four distinct, horizontally-Iayered worlds of the
gave examples of images likely used in the profane world and which, hence, underwent sorne
level of physical change and shifts in meaning as discussed in Chapter II. Images and
indentations used to mark the totem of the individu al buried were placed on grave markers,
or on treaties between indigenous peoples and Europeans, on objects to measure the number
of generations past, on headboards of bark canoes, on copper plates, as a "signature" which
was recognised over a geographical area, and on personal objects. Hunters, trappers, and
travellers used images to indicate information about boundaries, trails, and notices of
different sorts. Everyone over a wide geographical area recognised and understood this
group of images, therefore the origins and meanings of this group of images were not
shamanic.
It is hard to de termine whether images at pictograph sites were made at the same or
at different times. Images made at the same time may tell a story, or the images may have a
Page 450
meaning which is only understood by individuals. If images were made at different rimes,
then the implications for the meaning of these images is impossible to establish.
Very few image combinations occur repeatedly at any pictograph sites. The large
number of blobs indicates that many images have decayed, and that sufficient time has
passed for them to have deteriorated beyond recognition.
The variability of the images at the different sites depends on how professionaily
these images, and the sites, were produced. If a shaman produced these images, there would
be litde variety in the specialized vocabulary used. Yet, it is impossible to establish the length
of time over wruch professionals may have created these images. If the patterning of these
images at the individual sites is deliberate, then perhaps these images are designed to teil a
story.
A few sites, DiKm-4, DhKm-3, DhKo-1, and DhKn-1 have been re-interpreted
since 1990, using information provided by elder members of the Ojibwa Oackson 2000), who
also have interpreted images on birch bark scroils. The images are used to teil a story, but it
is impossible to teil whether the story is ancient, or whether it reflects the manner in which
the site and images have become part and parcel of the modern Ojibwa's cultural and mental
landscape.
The images on pictograph sites cannot be dated using absolute dating techniques.
Their style does not provide any indication of whether they were made either at the same or
at different rimes. The presence of the blobs, discussed at length in Chapter V, in
combination with the large number of blurred and exfoliated images suggests that the rock
surfaces were used repeatedly. Therefore, interpreters should treat the images as having
individu al meanings, and avoid considering a coilection of them as a group.
However, the question of treating a group of images at a site as telling a story
remains. Images considered as a story should have the same funcrion as those on birch bark
scroils. Therefore, l decided to conduct an examination of several birch bark scroils and their
images to establish the range of images, and to uncover any image combination on each or
between scroils. l compared these results to the rock image sites to determine whether birch
bark scroil images were used in the same manner, and, hence, have the same type of meaning
as those on rock image sites.
Page 451
Section 2: An examination of four bireh bark serolls.
1 obtained two song seroils, one of medicinal information, and one about an
historieal event. Eaeh seroil was unique. 1 tried to obtain similar seroil types, but aehieving
this goal proved impossible. What 1 eould suecessfuily ob tain with minimal bureaucratie fus s,
teehnieal issues and problems dietated the seroil type studied. The diffieulties of digitizing
black and white photographs or plates of varying qualities limited the number of seroils used
in this analysis. It was important to examine these images in original bireh bark form, and not
as transcriptions or redrawings.
The first seroil, seroil number 1 (figures 7.1 and 7.2), is a digitized file of a black and
white photograph taken by Jacqueline Rusak and 1 in September 1993 of the Massacre Seroil,
held by Lake of the Woods Museum, in Kenora, Ontario. The figure below shows half of the
seroil. The seroil had been bent in half and eould not be opened without risk of breakage, so
two photographs were taken.
Figure 7.1: The left hand part of seroll number 1.
Page 452
Figure 7.2: The right hand part of seroll number 1.
l successfuIly digitized the black and white print ln Densmore's (1910: 100)
publication on song scralis (see figure 7.3 and 7.4). Both scrolis were produced fram a black
and white print, not a photograph. l will cali figure 7.3 scrali number 2, and figure 7.4 scrali
number 3.
Figure 7.3: One of the serolls that Densmore photographed.
Figure 7.4: One of the serolls that Densmore photographed.
Page 453
Both seroils (figures 7.3 and 7.4) are part of a large study with good information
regarding their eontext and the meaning of sorne of the images. Both are held by the
Ameriean Bureau of Ethnology at the Smithsonian Museum.
Dewdney photographed and published the fourth and final seroil (figure 7.5) ln
1970 (1970c: 27& 28).
Figure 7.5: The seroU published by Dewdney in 1970.
Nlistakes ean oeeur when these images are traced, despite every effort on the part of
the reproducer. A detailed eomparison of Dewdney's (1975) drawing of the seroil on medieal
plants, from the Ethnology Department of the Royal Ontario Museum (figure 7.6), with the
original photograph (figure 7.5) published in 1970 (1970c) reveals that fundamental
differences exist in the shapes of the images. 7
7 Careful comparison of each figure in the photograph with Dewdney's tracing ab ove shows that Dewdney missed features of several of the images. These mistakes are inevitable when images on a scroIl or document are copied by hand or traced. This demonstrates the importance of obtaining a photograph of the actual scroIl or a digital version, a tiff flle. Fulford (1990), in an article examined earlier (see Chapter II) realised that problems often occur in the transcription of images on birch bark scroIls. It was impossible, he maintained, to provide definitive transcriptions and translations of the Hoffman material (ibid.: 128). He re-transcribed and retranslated the chants that Hoffman (1891) had compiled before reworking the entire corpus to develop glosses to improve the flow, or to supply better connections between the word and the pictographs. Fulford decided to do this beeause he saw inherent problems of Hoffman's informants' poetie use of language and Hoffman's own inadequaeies. In the end, Fulford (ibid.: 127) concIuded that the reproductions of the scroIls were generally accurate.
Despite great efforts, the diffieulty in improving the quality of digitized images of published bireh bark scroIl photographs beeame evident. The impossibility of using an original bireh bark seroIl also limited my research.
Page 454
. /" ..
Figure 7.6: Dewdney's tracing of the seroil in figure 7.5.
The annotated photograph below (figure 7.7) indicates rune differences between
Dewdney's drawing and the original photographed. A numbered blue box outlines each area
where lines are missing between the original and Dewdney's drawing. Dewdney did not draw
the verticalline, which could be a crease, running through the middle of the scroil. A narrow
vertical blue rectangle highlights this region of Dewdney's reproduction .
Figure 7.7: The nine differenees between the Dewdney's drawing and the original seroil.
(a) How were the seroils examined?
l used the same method and techniques to describe and analyse the scroil images as
employed for the pictograph sites. The same sequence of theoretical approaches used to
describe and analyse the pictograph images was applied to the images on the birch bark
scroils. Therefore, l first described images on the birch bark scroils using the methods
discussed at length in Chapter N. l examined the images using the techniques and methods
used for the rock image sites discussed in Chapter V. l sought scroil image combinations as
discussed in Chapter VI for the rock images. It was important to apply the same procedures,
Page 455
methods, and techniques to another group of images ereated and used by the Algonquian
speaking peoples. As the images of the pietograph sites in the tiff files had been deseribed
from left to right, ail of the images on the seroils were deseribed from left to right. This
approaeh was taken sin ce ail image files examined using KLEIO lAS must be deseribed from
left to right (see Chapter IV). This also foilows Hoffman's (1891: 267) observation as to how
the images eailed pietographs were read: the seroils eould be read from left to right, or right
to left and sometimes some seroils used both styles. Coleman (1947: 79) noticed that the
seroils belonging to the Midé in northern Minnesota were read from left to right. Vennum
(1978: 761) argued that migration seroils, used by the Midewiwin, were read from left to
right, since the Otter traveiled from the east to the west (ibid.). The eontextual information
for bireh bark seroil numbers 1, 2, and 3 was purposely ignored when deseribing the images;
l used it only in applying the eontextual approaeh. Very little eontextual information existed
for seroil number 4.
(b) Culture-Historie al Approaeh:
The images of the four seroils are etehed on the surface of the bireh bark. None is
eoloured in. The physieal state of the seroils eould not be eonsidered beeause photographs
were used for this study and the originals could not be eonsulted.
There are one hundred and nine images in total (xq414), averaging 27.25 per seroil.
Twenty-two of the one hundred and nine shapes have hoilow interiors (xq430). Five of these
shapes are on seroil number 1, eight on seroil number 2, while the remaining nine shapes are
on seroil number 3. Seroll number 4 has no hoilow shapes. Twenty-three shapes, found on
ail four seroils, have hoilow interiors with lines. There are eight shapes on seroil number 1,
seven on seroil 2, five on seroil 3, and three shapes on seroil number 4. Only three shapes, ail
on the seroil number 2, had centres that deseribed as partiaily hoilow (xq431). The remainder
of the shapes deseribed were made up of lines, and, therefore, eould be deseribed as having
neither hoilow, hoilow with lines, nor partiaily hoilow centres.
The images present are as foilows:
1) Five types of shape eailed a 'creature,' found only on seroil 1, exist in this dataset (xq419).
2) There are four types of 'bird' (xq422), ail on seroil number 1.
3) There is only one 'turtle' (xq423), and it is found on seroil number 4.
Page 456
4) 'Two parallellines' do not exist (xq420). However, the shape called 'two parallel vertical
lines' occurs four times, found only on scroll number 1.
5) 'A line of dots' occurs four times on scrall number 4. It does not occur alone, only in
conjunction with other images such as cireles, lines, and the turtle.
6) 'Cirele' appears forty-one times (xq426). Only six of these cireles exist without other
cireles, lines, squares, or rectangles, four were found on scrall number 2, and two on scroll
number 4. The remainder of the cireles are attached to other cireles, rectangles, squares
verticallines, horizontallines and diagonallines.
7) The 'zig-zag' does not exist on its own. It does occur in conjunction with other shapes
such as rectangles, horizontal, and diagonallines on scroll number 1 and 4. The zig-zag shape
is vertical on scrall number 1 and in the shape on scroll number 4.
8) There are ten shapes that have a 'rectangle' as one of their components. There is no such
shape that consists only of a rectangle. Rectangles occur only with other shapes such as
vertical, horizontal, diagonal lines, and cireles (xq418). They occur as components three
times on scrall number 1 and 4 and twice on scrolls number 2 and 3.
9) The 'square' type of shape was found twice (xq427): on scrall number 1 containing six
verticallines, and with a 'creature' on it; and on scrall number 3 with a cirele, horizontal, and
verticallines.
10) Two 'stick figures' were found, each as part of a larger shape (xq428) together with
cireles. One stick figure occurs on each of scrall numbers 2 and 3. However, only the top
half of the stick figure is evident on scrall number 3.
11) There are two shapes called "long diagonal line with short diagonal lines on each side"
on scrall 2 (xq433).
12) Six shapes called 'short diagonalline,' alliocated on scroll number 2 (xq432).
13) Three shapes called a 'very short diagonalline' are etched on scrall number 2 (xq432). A
similar type of shape, shorter than the other lines of the same type, 'a short diagonal line,'
appears on the same serail.
15) Only one 'one long verticalline' shape exists, found on scroll number 4 (xq434).
16) There are four 'triangle' shapes (xq435). One triangle is on each of scralls 2 and 3, while
the remaining two triangles are on scrall number 4. None appears alone and ail have lines of
different types.
Page 457
Ali other shapes only occur once, and are composed of a combination of vertical,
horizontal, and diagonallines (xq432), rectangles, and circles.
It is very difficult to comment upon the style of the scroUs. However, it is possible
that the two scroUs (numbers 2 and 3) coUected, discussed, and published by Densmore
(1910) show more similarities between each other, while scroU (number 4) published by
Dewdney (1970b) and scroU number 1, photographed in 1993 by Rusak, are similar in style.
However, the physical arrangement of the images on scroUs number 1 and 4 are quite
different. The shapes on scroU number 1 are distributed around the edge while those of
scroU number 4 do not share this stylistic characteristic.
Comparison of the images found on bireh bark serolls and on the roek image sites
Comparison of the above list of shapes with the list in Chapter V (page 268) reveals
that shapes caUed blobs are absent from the birch bark scroUs because of different
taphonomic conditions. Like pictograph sites, birch bark scroUs had 'creatures,' 'birds,' 'stick
figures,' and 'turtle' occurring on their own. The scroUs also had shapes caUed 'rectangles,'
'zig-zag,' 'squares,' and 'triangles,' but none of these occurred separately like those on the
rock image sites. The largest number of shapes for the four birch bark scrolls is the group
called 'circles,' but only six of them occur alone. The large st number of images in a group
occurring on the pictograph sites were the blobs.
The next largest group of shapes on birch bark scroUs were 'creatures.' Five types of
'creature' occur on the birch bark scroUs while twenty-three shapes loosely categorised as
'creatures' were identified on the pictograph sites. The 'creatures' on the birch bark scroUs
are considerably more complex than those of the rock image sites (see Appendix 17).
Furthermore, aU the 'creatures' occur on a single scroU, number 1, while the twenty-three
types of 'creature' occur on approximately a third of the twenty-seven rock image sites.
'Birds' occurred both on birch bark scroUs and at rock image sites. The birch bark
scrolls had four types of 'bird' shape while the rock image sites had only two types of 'bird'
shapes. One of the bird shapes, on birch bark scroU number 1, is very similar to the bird
shape on DhKm-3. However, the bird shape on the birch bark scroU is immediately beneath
another image which does not occur above the bird shape at DhKm-3.
Page 458
~ ..
Although both the bireh bark serails and the roek image sites had the 'turtle' shape,
none of these images had the same physieal shape.
Although the bireh bark serails had a 'zig-zag', it did not oeeur on its own but with
other shapes such as rectangles, horizontal, and diagonallines on scrail number 1 and 4.
The shape loosely categorised as 'stick figure' also occurred on two of the birch bark
scroils this shape was always part of another shape, a circle (xq 428).
Otherwise, twenty-three of the one hundred and nine shapes on the scroils were hoilow
while only forty-one of the three hundred and eight shapes of the pictographs were hoilow
(xq752). Hoilow shapes were found on ail four of the birch bark scroils (see Appendix 17),
but hoilow images did not occur on ail of the pictograph sites. The hoilow shapes on the
birch bark scroils were drawn as hoilow while those of the roek image sites may not have
always been hoilow. Sorne of these shapes may have beeome hoilow because of exfoliation,
the deposition of white mineraI deposits, lichen and rock tripe encroachment. On the other
hand, none of the images on the birch bark scroils were solid like those of the rock image
sites. Images on birch bark seroils could not be rendered solid in the same manner as those
of the sites without etching a hole in the bark. Although these images could not be described
as solid, they either had lines or other shapes in their centres.
There were twenty-five types of shapes on the rock image sites and sixteen types of
images occurred on the birch bark scroils. The birch bark scroils had a number of images not
found on the rock image sites including: 'short diagonalline,' 'very short diagonalline,' 'one
long verticalline,' 'circles,' 'long diagonalline with short diagonallines on each side,' and the
'two parailel verticallines' shape.
The average number of images per pictograph site was eonsiderably lower than the
number of images on the birch bark scroils: 11 to 27.25. Colour use illustrates another major
difference: colour is an important feature of the rock image sites, but it is entirely absent
from the images on the birch bark scroils. Indeed, the colour and type of paint prove to be
key indicators that sites are being reused. No such clear indication of reuse of the birch bark
scroils is indicated. The range and type of images found on the birch bark scroils and the
rock image sites are quite different. This eould indieate that each group of images belongs to
a different set used to communicate different types of ideas. Birch bark seroil images are
more eomplex than those of rock image sites. The person 'reading' and using the images on
the bireh bark relied on a specialized knowledge: one knew and understood the sequence in
Page 459
which they were used. l discuss the information for the two scroIls coIlected by Densmore,
which indicates that readers required special knowledge to understand birch bark scroIl
images in the homological approach later in this chapter.
(e) Contextual Approaeh:
The smaIl number of common (see Appendix 17) and the variety of unique shapes in this
dataset make it difficult to establish whether any combination of shapes occurs in aIl of the
different scroIls.
The eontext of the information on the seroIls:
SeroIll:
Very little contextual information exists regarding this scroIl. The dark verticalline in
the centre of the scroIl is a crease. Dewdney did not examine this scroIl although he spent a
considerable quantity of time researching and examining the birch barks found in the Lake of
the Woods region for his (1975) book on birch bark scroIls. Neither does he mention it in his
(1970b) article that published the photograph of birch bark scroIl number 4, nor in the book
he co-authored with Kidd (1962 and 1967).
Serolls 2 and 3:
Both of these black and white photographs were published in Densmore's (1910)
publication on the song scroIls, coIlected during 1907, 1908 and 1909 from the Chippewa
lndians of White Earth, Leech Lake, Red Lake reservation in Minnesota, and from a
Chippewa who lived in the Bois Fort reservation in Minnesota. Her study was discussed in
Chapter II. Densmore (ibid.: 96) argued that the word "medicine" referred to any substance
by which results "are supposed to be mysteriously attained." The scroIls have images which
she (ibid.: 96-106) connected with rare medicines. Densmore transcribed the title and words
of each of the songs on the scroIls as carefuIly as possible using an interpreter and the image,
the mnemonic was obtained and published in conjunction with the musical score of the song
sung, the words, and the harmonic analysis of the song. Densmore (ibid.: 96) stated that aIl
of the songs were sung by the shaman caIled 0' deni' gûn. She asserted that the songs were
examples of songs sung at the dance that either foIlowed an initiation or at lodges during the
evenings that preceded the ceremony. Densmore noted that only those who had purchased
Page 460
the right to sing them could sing the songs. 0' déni' gûn, the shaman who sang these songs,
was, Densmore diseovered, "one of the most powerful medicine men on the White Earth
reservation" (ibid.). Unfortunately, it remains unclear who exaetly provided her with this
information on the seroils. Dewdney did not examine or mention either of these seroils in
his (1975) book on bireh bark seroils.
Scroll4:
Very little is known of the eontext in whieh the images on this seroil were ereated.
Although the photograph of this seroil was published by Dewdney (1970b) in eonjunetion
with a detailed discussion of several pictograph sites in the Canadian Shield, Dewdney made
little comment about the birch bark scroil, the photographs of whieh were also published.
Dewdney (1975: 141) asserted that this scroil was a "pharmaceutieal song seroil." How he
arrived at this conclusion is unclear. Dewdney stated that this seroil was one of several seroils
given to the Royal Ontario Museum by Francis Fisher, who was one of the last Midé from
the English River region of the Canadian Shield (Dewdney and K..idd 1967: 13). Dewdney
(1975: 147) found it while examining the large woven medicine bag whieh had been given to
him, possibly during the summer of 1960, "for safe storage in the Royal Ontario Museum"
after the death of its owner. Dewdney did not examine this seroil either in his (1975) book
on the bireh bark seroils or in his book with I<.idd published in 1962 and 1967. It is
surprising that he did not comment on this seroil. It would have been interesting to know
how Dewdney would have considered it, given the six categories of seroils, in his (1975: 21-
22) publication, based on variations in themes he had identified: (a) origin seroils, (b)
migration eharts, (e) master ri tuaI and seroils, (d) ghost lodge and sky degree, (e) deviant
seroils, and (f) en..igmatie seroils; and given his further assertion that the principal funetion of
the bireh bark seroil was mnemon..ie.
(cl) Meaning:
What do these images mean? Ideaily one would use the homologie al, and the direct
historieal approaeh, and construet bridging arguments between written data regarding the
physieal shape of images, their mean..ing, and the images themselves. This eould work
effeetively for the two seroils, number 2 and 3, since Densmore (1910) eoileeted a vast
quantity of useful information on the images on these seroils. Yet, since very little
Page 461
information exists regarding the images' mearungs on serolls 1 and 4, stating anything
eonerete about these images, or verifying any recent interpretations and readings given to the
images of both these serolls remains diffieult.
Vennum (1978) stated that any interpretation and understanding of a group of
images on bireh bark serolls ean only oeeur if one is capable of drawing upon a large body of
evidence related to eaeh partieular seroll and its specifie images. Detailed information must
exist; otherwise any conclusions beeome tentative. Vennum drew extensively upon work
eondueted by Blessing (1963) and his informants, who analysed the migration type of serolls
of the Mille Lae reservation. Vennum (1978: 788) asserted, based on Blessing (1963: 93-94),
Hoffman (1891: 290) and Densmore (1910: 26), that the images, whieh were mnemonies,
"were less generalized and their meanings seeretly guarded."
Landes (1968: 172, 224) had established that, if images were used on bireh bark
seroIls, only the shaman who ereated the seroll knew the specifie meaning of the images. The
"partieular message" of a seroIl, or, essentially, what the pietographs meant, was probably
impossible to establish unless the owner, who was also probably its maker, labelled or
translated eaeh figure. Hoffman (1891: 191-193) stated that a candidate eould learn the
meanings of these images only after paying his fee and preparing himself through fasting and
tobacco offerings. The Midé used serolls as mnemonic devices to remember the words of
their chants, for the instruction of new members, to record oral traditions, and to perform
correct ceremonial procedures.
Hence, each seroll and its associated images should be examined in conjunction with
a body of data that is seeurely connected with the seroll in question. It therefore makes sense
to eonsider the data at hand about the images on each scroll.
SeroU 1:
It is diffieult to know what the images on the scroll ealled the Massaere Scroil may
be. It would be really useful to know when the seroll was made, who made it, to whom it
may have belonged before it was obtained, and why it was obtained. The text on the baek of
the black and white photograph, in Jacqueline Rusak's handwriting, states that an Ojibwa
elder, whose name is unknown, interpreted the seroll as depicting the murder of La
Vérendreye's son, his eompanions, and Jesuit missionaries while they were holding couneil.
Their deeapitated bodies were found in a circle. However, various useful pieces of
Page 462
.r'.
information are unavailable regarding this explanation: the name of the elder who gave this
interpretation, the elder's status within the Ojibwa eommunity, whether the elder was Midé,
and the name of the elder's reserve.
This seroIl and its images were utilised by Reid (1979) in an attempt to date several
images at a number of petroglyph sites at the Lake of the Woods. Reid (ibid.: 250) relied on
an undoeumented interpretation by James Redsky, a Midé from Shoal Lake on the western
side of the Lake of the Woods, whieh was perhaps the same reserve or different to that of
the unnamed elder. It is impossible to establish from Reid's (1979) article how and whether
eaeh image was interpreted by Redsky beeause Reid did not clearly indieate whether he
eoIleeted this information. However, he (ibid.) did provide sorne information that Redsky
provided to him about sorne of the images on this seroIl
Redsky interpreted the bird-like shapes as a 'paisq'. He argued that this word eould
be roughly translated as "bird of omen" and that this image oeeurred three times on the
seroIl. Reid eonsulted Baraga (1992 [1878]: 354), a Roman Catholie missionary who ereated
the first Cree/Ojibwa and English dietionary, and learned that the word was 'peskwe' whieh
meant 'a kind of owl'. Reid (ibid.) observed that the paisq symbol for Redsky identified
"tragedies or undesirable events," If the "wings" are upward to form a "Y," the event was a
future one. If the wings were down the event had happened, and it eould not be avoided. If
the wings were to the side, as in the seroIl then the event was taking place. Redsky, Reid
(1979: 250) observed, interpreted the seroIl as "purporting to show the massacre of Father
Aulneau, La Vérendreye's son, and their eompanions in the Lake of the Woods."
Serolls 2 and 3:
The information gathered by Densmore (1910) on seroIls 2 and 3, provides more
signifieant clues to these images. She published sorne of the mnemonies, eaIled the song
pietures, for sorne of the shapes on these seroIls. She (ibid.: 97-106) provided the song
pieture that triggered the singing of eaeh song, the Ojibwa words and their translation, the
score of eaeh song, its tide, and her analysis of eaeh. Her information enabled eaeh image,
eaeh mnemonie, visible in both photographs to be identified as the ones whieh triggered the
songs to be sung. Both seroIls are presented, in figures 7.8 and 7.12 below, annotated with
white numbers and aeeompanying information so that sorne of the songs ean be identified.
Page 463
Figure 7.8: Seroll number 2 annotated by Cols on from left to right in 2005.
The shapes numbered 12, 13, 14, 15, 16, 17 are part of one song picture, in figure 7.9 below,
identified by Densmore (ibid.: 100) as the song called the "Song of Good Medicine".
SONG-:rJCT'Clt.L~,.'~' •. ~ dO·
reat.hürs are Sf,~(··n Hf:;\r
the 11lde'wlnl'11!.
Figure 7.9: Mnemonie for the Song of Good Medicine.
The shapes numbered 18, 19, 20, 21, 22, 23, 24, 25, 26 are another song picture, in figure
7.10, called the "Song of the Crab Medicine Bag" (ibid.: 102).
SONG l'TCT1Jl\E NO. 85.
The drawing Tepresents a Mîde' bag witlJ t'NO mi'gls besido il.
Figure 7.10: Mnemonic for the Song of the Crab Medicine Bag.
Page 464
The shapes cailed 28, 29, 30, 31, 32, 34 are part of another song picture, in figure 7.11 below,
discussed by Densmore (ibid.: 103) as constituting the "Song of the Fire-Charm".
fo'ONG PIGTURE NO. S6. The fiames are seen ascending froID a cl rcle of ftre.
Figure 7.11: Mnemonic for the Song of the Fire-Charm. SeroU 3:
Figure 7.12: SeroU number 3 annotated by Cols on in 2005.
It is possible to undertake the same procedure for scroil number 3 using Densmore's
information; it is evident that shapes 7, 8, and 9 are part of the song picture for the "Song of
the Owl Medicine" in figure 7.13 below.
SONG PICT"URE NO. 88. The Mïde'winl'nl', tIle man and l1is \Vife, are see-n in the wigvram,from which the owl is flying.
Figure 7.13: Mnemonic for the Song of the Owl Medicine.
Page 465
,------"
The shape on seroil number 3 known as number 10 ean be identified as the song picture for
the "Song of Starvation" (ibid.: 1 04) in figure 7.14 below.
SONG PICTURE NO. X7.
Figure 7.14: Mnemonic for the Song of Starvation.
The shapes numbered 11, 12, 13, and 14 on bireh bark seroil number 3 are identifiable as
integral eomponents of the song picture eailed "Song of the Man who Sueceeded" (ibid.: 98)
in figure 7.15 below.
SONG PICTURE NO. 82.
Figure 7.15: Mnemonic for the Song of the Man who Succeeded.
Scro1l4:
Unfortunately, it is impossible to state anything except that Francis Fisher, who had
been a Midé in the English River region of the Canadian Shield prior to his death sometime
after the summer of 1960, used this seroil. It is impossible to comment on any of the shapes
deseribed (see figure 7.16).
Page 466
Figure 7.16: ScroU number 4 annotated by Cols on in 2005.
(e) Conc1uding comments:
Differences in description between those who knew the song pictures and those with
little or no knowledge of how they should be used together must exist. Without the privilege
of knowing what these images mean, images are subdivided and dealt with in a manner that
could ultimately obscure their significance. It is very difficult to assign meaning to individual
birch bark scrolls. This may not matter since Vennum (1978) believes the researcher can
examine these images and scrolls in great depth only if he can draw on a large body of
ethnographically relevant information. If the body of relevant evidence for each scroll does
not exist, then considering these birch bark scrolls as documents, as Vastokas (1996: 55)
advocates, becomes problematic. It becomes impossible for the historian to ask the pertinent
questions of them: who made the scroIl, who created the images, why were they created, for
what purposes, and what was the date that the scroll was made. However, should this body
of evidence exist, it is still impossible to successfully provide a date for the document, the
scroll.
Section 3: Efforts to treat rock image combinations as stories.
It is very difficult, if not impossible, to determine whether the images on birch bark
scrolls have shifted in meaning. Examination of the images on the scrolls number 2 and 3,
collected by Densmore, indicates that images sometimes occurred together, although this
was unclear without examining the drawings of several of the songs. Additional information
on the images indicated the songs enabled one to identify which images occurred together.
However, at some of the pictograph sites in the Lake of the Woods where collections
of images exist, some of the images have been given new meaning, and they are considered
as connected, as stories. This contradicts the results discussed in Chapter VI, where it was
Page 467
concluded that definite patterns did not exist between the different images, despite using a
nested classification. The images on the pictograph sites in the Lake of the Woods do not
have a commemorative strain, but new meaning may have been attached to them at five sites:
DhKm-3, DiKm-4, DgKm-l, DhI<n-1, and DhKo-1. Each site is examined in sequence to
demonstrate how new meaning has been applied to the images at site.
(i) DiKm-4:
This site was interpreted by Ann Seymour, the Director of the Lake of the Woods
Ojibway Cultural Center in Kenora in 1990 (Anonymous 1990a: 10). The annotated
photograph, in figure 7.17, isolates each of the images to clarify which image Seymour
discussed. She stated that the images at this site indicated that "someone took a long and
difficult joumey that lasted a month. The snaky line denotes the journey and the crescent
beside the encireled he ad denotes the moon - a month." It was coneluded by the writer of
this short article, who spoke to Ann Seymour, that the journey taken was "a happy story"
and that the "Cirele of Heaven on the pictograph tells us that the journey was successful"
(ibid.).
Figure 7.17: The shapes at DiKm-4 according to Ann Seymour.
Page 468
(ii) DhKm-3:
This site has been named the Grand Medicine Society (Anon. 1990b: 3). Seymour
also interpreted several of the images at this site. She asserted that the turtle was "Miknik"
the messenger or translator between the spirits (manitous or manido) and the people (ibid.).
A similar statement from Seymour exists in Rajnovich's (1994) publications as weil as the fact
that the turtle was the translator because they spoke many languages (Seymour in Rajnovich
1994:36). Seymour argued that the spiritual aspect of the turtle was indicated by the two
horizontallines across its back. She interpreted the shape between the lattice and the turtle as
a snake and "offered the suggestion that the snake was somebody's name" (Anon 1990b:3).
The lattice for Seymour depicted the degree to which the painter of the rock image site
belonged sin ce the lattice was the sign of the Midewiwin Society (Seymour in Rajnovich
1994: 32). The shapes that Seymour described are indicated in figure 7.18 below.
Figure 7.18: The turtle, snake, and the first lattice discussed by Seymour at DhKm-3.
Redsky, from Shoal Lake, on the western side of the Lake of the Woods, was apparently
asked for his opinion of these images. He interpreted the turtle shape as Miknik "the
messenger" and argued that the lattice figure was the sign of the Midewiwin. The snake
shape was interpreted as possibly the "Great Spirit and Lightening" (ibid.). The photograph,
in figure 7.19, illustrates the rest of the images interpreted at this site by Seymour.
Page 469
Figure 7.19: The thunderhird, Hne, handprint, lattice, and the human figure with hands raised upwards at DhKm-3 identified hy Seymour.
The writer of trus text stated that another lattice figure occurs with a figure that Seymour
described as "a human figure with hands raised" (ibid.). Seymour interpreted the hand print
to the right of trus image "as signs for the death of a warrior." The bird visible in figure 7.19
she interpreted as a thunderbird, and the zig-zag shape connected to it as a communication
line or a "power line to the heavens" (ibid.).
(iii) DgKm-1:
This site, in figure 5.99, has been used in a land daims proposaI for an area of land in
the Sabaskong Bay area (Ross pers. Comm. August 2001). No additional statements have
been obtained as to how any of the images at trus site have been interpreted.
(iv) DhKn-1: An eIder at Manitou Rapids, a reserve to the southeast of the Lake of the Woods,
interpreted the images here (Reid pers. comm. 2005). Their interpretation was subsequently
published in a guidebook to the Lake of the Woods although it was not acknowledged in the
publication Qackson 2000: 112). The images, in figure 7.20, were described as follows: a
turtle wruch is symmetrically faced on either side by two serpentine figures. One of these
serpentine figures is described as having antlers while the other has horns.
Page 470
Figure 7.20: An annotated photograph of part of DhKn-1 demonstrating the images discussed.
To the left of both of these figures is another creature identified as a serpent which has an
open mouth, ears, and three long flippers.
(v) DhKo-1:
An altemate interpretation of the images was given to some of the shapes at DhKo-1
on Picture Rock Island (figure 7.21) in the same guide book as the interpretation for DhKn-1
(ibid.).
Figure 7.21: An annotated photograph of DhKo-1.
The creature shape in figure 7.21 has been interpreted as a sacred bear, possibly representing
a "powerful Midé leader" (ibid.). This image is "connected by a series of rectangular forms
with paths between some" but, unfortunately, it is difficult to determine what this means.
Several of the shapes were described as lodges, while one of the figures nearby was identified
as a human figure. Since a drawing demonstrating these speculations was not provided, it is
Page 471
impossible to ascertain the precise location of the lodges and the human figure. A horned
figure exists in the centre of the site, and to the lower le ft right hand of it, another image, a
"horned serpent-sturgeon" that has projecting spines along its back.
Conc1uding remarks:
The above discussion clearly demonstrates that people ascribe meaning regardless of
whether they created the images or whether or not they can rely on written or dated oral
information. It was impossible to establish the meanings of the new images at DhKm-19 and
DhKm-20, and the new images placed on the older images at DgKI-2. One person's signing
of the new images at DhKm-20 suggests that the rock image sites were given meaning by
one individual, rather than a group of people, and do not have meaning beyond the
individual. However, the new meanings given to other images at the sites discussed in this
section indicates that these images have been integrated into the lives of sorne of the people
who currently live in the region.
Section 4: The continuing importance of the images and sites.
Similarly, specifie places are still important to the native peoples who live and work in
the Lake of the Woods. Perhaps new shapes have been added to the graphie vocabulary
because newer and very professionallooking images were found in 2001 at DgKl-2, DhKm-
19, and DhKm-20. There were seven shapes painted using modern paint, but only four of
these are new to the vocabulary evident in the dataset of the Lake of the Woods. Both of the
shapes at DhKm-20 were new but the shape at DhKm-19 was one of the three turtles
located in this region. Otherwise, ail of the newer images are radicaily different from the
images elsewhere. At DgKI-2 four new images are in modern red, painted over and beside
oIder images created using ochre:
1) a creature with head, with horns or ears or antlers and a pair of front and
back legs, facing right
2) star
3) unidentifiable shape, hoilow body outlined by lines
4) vertical crescent with points facing right
Ail of these images except for the third image listed above were superimposed on an older
ochre smear which is visible in figure 4.4. Dewdney first recorded this site in 1964, at which
Page 472
time these images were not present. These new images were also absent when Pastershank
recorded the site in 1989. Although these are new images, theyare similar to the older images
at other sites. These new modern images, however, differ from the other new images at
DhKm-19 and DhKm-20, sites unknown to archaeologists prior to 2001. Ali the images at
both of these sites were made without ochre, using modern paint.
An individu al named R. Bird signed and dated as "1982" the red, blue, and white
images at DhKm-20 (figure 5.141). These images strongly resemble those images in paintings
and art objects created by members of the New Woodland School of Art. This School was
founded by Norval Morrisseau who drew on Ojibwa beliefs and traditions. This movement
developed in the late 1960s in Northwestern Ontario and, according to Trevelyan (1989: 188),
"came of age" in 1983, when the Art Gailery of Toronto mounted an exhibition of the works
made by members of this group.!
Trevelyan explored several ideas but one of them in particular should be considered.
She (ibid.: 198) argued that art was a vehicle for the transmission of new ide as and she also
posited that the development of this school was a response to a cultural crisis among the
native community.
The image of the turtle at DhKm-19, in figure 5.160, is readily identifiable as a turtle
painted in black and green paint. It does not bear as strong a resemblance as those images in
DhKm-19 to the works of New Woodland School of Art. It is possible to argue that these
new images are probably professional; they bear a strong resemblance to those found in the
artworks of this school of art.
Conclusion:
The results of the analysis of the four birch bark scroils indicate that it is impossible
to ascertain the identity of particular songs, mnemonics, without a body of data that is
reliably connected to the birch bark scroils in question. However, this study did suggest,
based on stylistic differences in the types of images, that the images on the pictograph scroils
probably had different functions and uses. 1 have already disagreed with Vastokas' (1996: 55)
1 Morrisseau's colleagues included Daphne Odjig,Jackson Beardy,Joseph Sanchez, Eddy Cobiness, Alex Janvier, and Carl Ray. These individuals have influenced two generations so that the works of Arnoo Angeconeb, Shirley Chee Choo, Blake Debassige, Don Ningewance, Leland Bell, and Carl Bearn are known internationally beyond the region north of Thunder Bay, Ontario.
Page 473
opinion that birch bark scroils should be considered as documents unless there is detailed
ethnographic description. Her suggestion that these scroils should be considered as "cultural
expressions" is valid.
However, as with the images of the pictograph sites, describing these images must be
undertaken with considerable eare and thought because it is dearly very diffieult to deseribe
an image. It is very easy to be subjective, rather than objective about them and to gloss over
minute details.
It is evident from the comparison of Dewdney's drawing of the scroil (1975: 141) and
its photograph (1970b) that he made nine transcription errors. Mistakes in tracing inevitably
occur, and the cultural perspective of the individual describing the shapes must be
acknowledged.
Vastokas makes two interesting points, both of which are difficult to sustain. She
posited that, as records of Aboriginal history and an historical source of "writing without
words," scroils acted as "representations" of events that took place over time and space
she did not provide any dues as to how to undertake this task.
It would have been useful if she had articulated precisely how she thought this goal
could be aehieved. l am forced to argue, based upon my small sample of the four birch bark
scroils, that very little can be achieved regarding the meanings of these images without a
substantial body of ethnographic information historicaily connected to the scroils. Vastokas'
(1996) paper is important, and her ideas are theoreticaily good but they represent an ideal
researeh situation that is rarely aehievable.
The red images on the pietograph sites indieate that the Algonquian-speaking peoples
used, and stiil use, different shapes for expression, but establishing the fuil range of image
combinations is impossible. Meanings of the images on the rocks are not readily obtainable,
and common sense is no guide; it depends on intuition. Their meaning cannot be ascertained
without verbal intuition. Since the number and type of images is so smail, and the people few
in number, the groups of Algonquian-speaking peoples probably were too small to have had
several specialists using different types of images. The wide range of shapes, the differences
in the range of shape type used, in eonjunetion with the physical differences between rock
image sites and those of bireh bark seroils, suggests that graphie voeabulary was not only a
product of speeialized image makers but was widely shared. It is possible to eondude that
Page 474
the images on the pictograph sites had meaning attributed to them at the individual level
rather than the group level.
Although it is impossible to ascertain the meaning of the images and the pictograph
sites, the continued practice of leaving offerings at sites such as DhKm-l proves that the
physical location of these sites held, and holds, considerable importance for sorne of the
people of this region. What les son do we learn from the examination of birch bark scroils for
those of the pictograph sites? First, these images are combined in a manner obscure to
someone uninitiated in the graphic vocabulary involved. The images of the scroils are more
complex than those of the pictograph sites. Second, only drawing on a detailed body of
ethnographie, eontextual literature direetly eonneeted to the images and the seroils in
question made establishing the eombination and meanings of the images possible. The body
of information required to identify and even start to grasp the meaning of the pictograph
images has been lost and forgotten. Perhaps those who may have used it never coilected it,
since the coilection of information regarding images the Midé made possibly was perceived
as more important. The third les son is that the role of the creator or creators with regard to
the images of the pictograph sites has been forgotten. The lite rature appears to lay too much
emphasis on subsequent twentieth century observers, espeeially on the viewpoints of
archaeologists. Finaily, care must be taken with regard to using ethnographic, and
archaeological sources for establishing the meaning of these images and the sites.
Sorne level of choice probably existed for the selection of a location and the images
used for a pictograph site, explaining why so few of the pietograph sites that had images had
ail of the five "required physieal features" (see section 1, Chapter VI). The "required physical
features" were likely unimportant in the creation of images (see Appendix 15). The images
probably were created over a long period of time by individuals and not by ritual specialists.
Hence, the importance of these features has either changed, shifted, or been mistaken.
Regardless, of their current significance it is the fact that they exist that is important. These
images are weil known by the archaeologists, the anthropologists, and the art historians and
by the local indigenous peoples. Essentiaily, everyone is talking, reading or conversing with
each other. Therefore, the information that images should occur with five specific physical
features has become an established fact, and so offerings are left where a large proportion of
the five features are present in conjunction with images. The present day indigenous
Algonquian-speaking peoples are interested in these images. Sorne of the images at several of
Page 475
the sites have been re-interpreted. Those who currently live in this region have integrated
these archaeologically and historically important places into their mental and physical
landscape. The presence of the images in modern paint on older ochre images, and new
images at new places in the landscape suggests that images are still used as a means of
communication, and that they are integral to gaining an understanding of the archaeological
and rustorical record, and of the present day.
Page 476
('
Chapter VIII: There is no Holy Gran: An Obsession with Meaning.
Introduction:
Wright (1981) described the Subarctic as a forbidding landscape challenging to its
pre-contact human occupants. As l argued in Chapter II, successful occupation of this region
requires a comprehensive understanding of its seasonally dynamic ecology, water
transportation routes, seasonal climatic extremes, and a host of hidden dangers, risks, and
opportunities. To interpret the placement of archaeological sites of different functions across
the landscape, the investigator must understand the complexity of past land use, the regional
geography' s physical nature, and the everyday challenges past occupants faced. The
archaeologist must become personally immersed in the site's physical geography and context
since, in the now more heavily populated Lake of the Woods, many of these considerations
are integral components of the process of conducting archaeological field investigations.
To travel, live, and work in the boreal forest requires a high level of technical skill,
intra-group co-operation, self-reliance, and understanding of the physical environment in all
its complexity. Even during the height of the summer, sudden weather changes, and relative
isolation require 'bush skills,' awareness of one's surroundings, and teamwork. Careful and
thorough preparation must precede any expedition to a site, even in the first years of the
twenty-first century. As a outsider from the twenty-first century with the advantage of
technology, one still must make compensations and have the challenge of maintaining a keen
awareness of how it might have been in the pasto
An archaeological field investigation, even one serviced by a large motor boat,
requires careful preparation and a wary respect for the dangers associated with travel on
waters such as Lake of the Woods. Three people might be required for such research treks:
one to operate the motor and drive the boat, another to navigate the safest course through
intricate channels, and the third to serve as a general lookout. The water surface must be
constantly scanned for deadheads [floating or semi-submerged logs], the next navigation or
hazard buoy, for tell-tale signs of submerged reefs, and especially, swimming bears, moose
and other hazards. Even short journeys require a careful review of the charts before setting
out, preparation of equipment, and consideration of weather conditions expected over the
course of the day. Such considerations would be even more critical for travellers in small
canoes.
Page 477
r--..
As we travelled, we "read" the lands cape so as to de te ct the precise location of a
pictograph site. Our preparation enabled us to identify tell-tale signs. This is especially
difficult when travelling through a lands cape that can look surprisingly undifferentiated to
even an experienced traveller on the lake; becoming lost was never difficult. The charts,
although recent and glossy, are "idealizations." Ultimately, not getting lost depended much
on our ingenuity and ability to find destinations represented by nothing more than dots on
small-scale maps.
To travel at 35 kilometres an hour on Lake of the Woods is to live life in a blur. The
different greens of the coniferous and deciduous trees blend in with the greys of the granite.
Rapidly changing weather conditions can obscure distant landmarks, and make it difficult to
traverse the intricate channels and travel corridors through the lake. At the same time, the
water's surface can change rapidly From a dear turquoise, dark grey, to a deep blue,
depending on the douds, the wind and the rain. The dear distinctions between different tree
species varieties become ever more evident, the granite's strata increasingly marked as we
slow to a haIt. The intricate mosaic of leaves becomes identifiable as a myriad of plants, and
startling contrasts in the colour intensities of the granite rock surfaces emerge as individual
outcrops. As we advance toward our shoreline of interest, and slow the boat's engine until
we almost drift, it becomes easier to examine the rock's surface and to search its physical
features for any traces of red pigment concentration: the pictographs themselves.
An important observation, arising as a consequence of my fieldwork and analytical
work, relates to how archaeologists deal with the sites and images. The action, especially by
archaeologists, of offering tobacco at ail the places with paintings implicitly implies that aU
the sites have, and had, the same meaning for aU of the local indigenous peoples. This
assumption must be questioned, especially since offerings were not found at aU of the sites
examined. It is possible that tobacco left may have blown or washed away. It is equally
possible that not aU places with paintings may have had the same enduring spiritual meaning
among local Aboriginal residents that sorne of them appear to have today. It may be that
offerings are le ft at different sites under different circumstances. Whether the archaeologist
finds offerings there or not may be entirely coincidental timing between his visits and other
individuals visiting and using the site.
Were and are these rock paintings always considered spiritual? Or are the places and
not the painting spiritual? Leaving tobacco at aU of the sites presumes that aU of the places,
Page 478
or the paintings, have and have had the same meaning(s). It is possible that these images
and/ or the rocks on which they exist once had a spiritual significance but, unfortunately, the
oral tradition documenting that character has been lost in the face of acculturation, or is
impossible to securely establish.
Ethnographie literature exists concerning the practice of offering, or giving, tobacco
in exchange for information in a wide a variety of scenarios (see Chapter V), but secure links
between the offering of tobacco with sites do not exist in the literature for this region until
the early twentieth century (Colman 1947; Hilger 1936). Only towards the end of the
twentieth century did Pufahl establish through his informants at the Eagle Lake Reserve,
near the pictograph sites he was studying, that the local indigenous community considered
these sacred sites that "routinely receive[d] offerings of tobacco" (1990: 19). Five years later,
Smith (1995: 33) argued that the use of tobacco was a sign of respect between humans and
the manitous and maintained it was common practice to offer either loose tobacco or
tobacco in the form of cigarettes to EIders.
Now, it is common practice amongst archaeologists to make an offering of tobacco
once the images at a site are distinguished but before recording the site. A pinch of tobacco
is taken out of the packet of loose tobacco and scattered on the surface of the water
immediately in front of the images, or left in a crack or crevice in the rock face in the
immediate vicinity of the images. This is to pay respect to the spiritual character of the site in
line with instructions from Aboriginal EIders. But not ail pictograph sites had evidence of
previous offerings of tobacco when they were examined in 2001. This might be because the
archaeologists left the tobacco on the surface of the water, rather than in a cleft in the rock.
Or it might be that tobacco left by other people, such as local Algonquian-speaking peoples,
had decayed, blown or washed away.
My crew and l always sprinkled tobacco onto the surface of the water but other
archaeologists do different things with the tobacco. When l first started fieldwork in
Northwestern Ontario, Rusak, Pastershank, and Reid (pers. comm. June 1991) advised me to
leave such offerings at sites. Rusak invariably left some tobacco either on the water's surface,
or in a crack, or crevice when she did fieldwork (ibid.).
Regardless of this practice, the vast range and quantity of o fferings , including
tobacco, suggests that some of the physicallocations and/or paintings stiil remain places of
visitation. Some groups of people other than archaeologists are leaving, and have le ft, these
Page 479
offerings. Anthropologieal and ethnographie writings establish that gifts of objects such as
tobacco or other articles are often given to supernatural beings to obtain spiritual benefit,
especially in a ritual context. Archaeologists argue that the offerings found at sites and
discussed since 1885 suggest that each possesses sorne ritual significance. Archaeologists
examining pictograph sites in the Lake of the Woods (Fox 1974; Dewdney and Kidd 1962,
1967; Lambert n.d., Pastershank 1989; Reid 1975, 1976) always seek objects left as offerings
ab ove water.
It is impossible to establish however, the precise nature of these red images, the
pictographs. It is unclear whether they held any religious symbolic significance. It is possible
that the paintings were simply used as forms of communication.
Therefore this practise of leaving offerings by archaeologists may be something of an
'invented tradition'. After all, these archaeologists are not usually indigenous Algonquian
speakers. Their actions may arise from a reading of ethnographic sources, or perhaps the
advice of previous archaeologists. Essentially circularity exists here since the archaeologists
were told (or believed) that painting sites are spiritual places. The action of leaving offerings,
such as tobacco, is a means of paying respect to the Ancients [the antecedents of the current
indigenous peoples] and honouring and respecting the beliefs of contemporary traditional
believers. In actuality, the archaeologists have been blinded by their actions to possible other,
and more secular meanings, of the rock paintings.
Invented traditions are, of course, nothing new. In introducing a well-known
collection of essays on the subject, Hobsbawn (2003: 9) identified three types of 'inventions,'
all of which seem to have emerged in the wake of the indus trial revolution, coinciding with
the dawn of modern archaeology. He identified these overlapping 'inventions' as:
a) those established or symbolizing social cohesion or the membership of groups, real or artificial communities, b) those establishing or legitimising institutions, status or relations of authority, and c) those whose main purpose was socialization, the inculcation of beliefs, value systems and conventions of behaviour.
(Hobsbawn 2003: 9)
From this powerful observation he later argued that:
the study of invented traditions cannot be separated from the wider study of this history of society, nor can it expect to advance much beyond the mere discovery of such practices unless integrated into a wider study
(Hobsbawn 2003: 12) Leaving tobacco as an offering, the archaeologist ascribes an importance that may
not exist to the site. Archaeologists may be creating a tradition by continually depositing
Page 480
tobacco at pictograph sites, especially at those that are newly discovered. We cannot establish
the precise significance of the places where the images occur, and we cannot securely ascribe
a meaning to them. Meanwhile these sites, their images, and the invented traditions are
probably being inexorably incorporated into the lives of the sorne of the local Aigonquian
speaking peoples.
It is very difficult to quantify the level to wruch these sites are used today. It may be
that only a few people from a few reserves use and know of a few sites in sorne parts of the
region. Indeed, the presence of new images at previously known pictograph sites and the
existence of newly discovered sites with images not made using ochre demonstrates that
these images and their sites are being utilised and integrated.
What has been going on here?
My fieldwork was carried out in 2001. The number of images extant on a site bore
little relationship to the presence of offerings. A site could have a large number of images yet
no offerings. The physical proximity of a pictograph site to an Indian reserve also had no
influence. There also was no relationship between the orientation of a site, the landmass on
which it was located and the existence of offerings. Yet the fewer the number of those
"required" physical features present at a pictograph site located on a cliff site, the less the
likelihood of offerings being found. By liberally distributing tobacco are we not manipulating
our own evidence? We are not referring to the artefacts, rather to the relationsrup between
the individuals looking at the site. By following the advice of archaeologists to leave tobacco,
a researcher acknowledges the dignity, legitimacy, and value of that 'other world' to wruch
the site may belong. The archaeologist arguably enters by the act of 'exchange' since no
information can be obtained without the prior offering of tobacco. A complex bridging
argument is involved so that we can assume that the Algonquian-speaking peoples who
currently occupy the region have a long rustory of association with the sites.
Are we perhaps being hoodwinked by the fact that the sites contain 'pictures' or 'art'?
That is after all what the sites are called, in both the popular and academic literature.
Regardless, the principal artefact examined in trus the sis is an image. Images are neither
flakes, nor pot-sherds, nor bones. Images are vectors of information, like lithics, bones, and
ceramics. Each has its unique methods by wruch it can be described and subsequently
Page 481
analysed. The image is a form of communication, and deserves the same consideration, at the
same time as the tool, the bone, and the pot-sherd. AIl must be considered as equaily
important, and possibly even integral to each other's context.
Perhaps the problem is that we do not know the stage at which images were utilised
as forms of expression or as a means of simple communication. It may be that images have
been mislaid or ignored because the emergence of lithics, ceramics, and bones can be more
easily cross-referenced, subsequently documented and incorporated into the archaeological
record. In order to carry out a rigorous examination of an archaeological phenomenon, we
must draw upon increasingly large bodies of disparate data. This is especiaily true of an
image, which is perhaps best understood as a multimedia event, with different multiple
sources, multiple meanings, and sequenced approaches required. Such approaches are
inherent to the application of a range of digital technologies.
This is not surprising, for, as a draft recent report of the Cyber Information
Committee established by the American Council of Learned Societies argues,
a critical mass of information is often necessary for understanding both the context and the specifies of an artefact or event, so that often a very large datas et of multimedia content-image, text, sound, moving image, audio-is required. Humanities scholars are often concerned with how meaning is created, communicated, manipulated, and perceived through discourse. Recent trends in scholarship have broadened the definition of what faIls into the category of discourse, and many scholars formerly comfortable working only with texts now turn regularly to architecture and urban planning blueprints, art, music, video games, ftlm and television, fashion illustrations, billboards, dance, videos, gesture, graffiti, food, rituals, as weil as blogs. While difficult to achieve, the value of critical mass or functional completeness is easily demonstrated. The Shoah archive has a signal authenticity in large part because it is so comprehensive. The tale of what happened to one or two familles, in one or two villages, in one or two countries, during the Holocaust is worth recording and disseminating-but how much greater is the knowledge we gain from the completeness of the record.
(The Draft Report of the American Council of Learned Societies' Commission on Cyberinfrastucture for Humanities and Social Sciences 5 November 2005: 27)
The need for academic rigour is paramount with any object, especiaily with images, and it is
attainable in this digital age.
Images are important sources of information, vectors that, as noted in the quotation,
have been abandoned in favour of text or script. If we continue to charge ahead because we,
as humans, want to know their 'meaning' without using a disciplined, quantitative and
qualitative approach, we will achieve very little. By the same token, each of these data related
Page 482
~ ....
to an Image can be stored in digital format now, and, therefore, can be manipulated
according to well-defined and standardised rules and practices, often formally documented,
e.g., by the IEEE. We can go some way to understand and to document the contexts of
images and how both have changed.
Images are such important vectors of information that some, like the Swastika, or
Madonna, are swamped with different contexts, and with evidence pointing to different
points of view. Others of the very distant or not so distant past ean laek such compelling
detail. The image known as the "Marlboro Man" is a powerful ieon invoking the mythical
Ameriean eowboy and masculine trademark. This image assisted in establishing Marlboro as
one of the world's best-selling cigarette brands (MacKenzie Ward pers comm. 1991). The
image known as the Swastika has a long history, a variety of contexts, and many implicit
meanings throughout the world. Someone could use the word 'Madonna' and think that
everyone knew implicitly they were referring to Leonardo da Vinci's painting ealled the
'Dreyfus Madonna' held by the National Gallery in Washington, De. Yet, it is equally
possible that one could think of a photograph of the pop star called Madonna. Indeed, the
manner in which images are utilised can differ extraordinarily from century to century, and
between cultures and peoples.
Iconographic disjunctures are probably commonplace. In today's world if someone
cannot read text, they will probably rely on pictures or diagrams to help them 'read' and
understand, as mu ch as possible, about these images. Images are very powerful in the
manner in whieh they ean convey a vast amount of information without words. As human
beings we want to 'read' images, establish their meanings and understand them.
To understand the image, the pictograph, as a source of information reqwres a
different way of thinkingl. It involves drawing on all of the disciplines that use images as
sources of information: art history, new art history, anthropology, and history, not just
archaeology. Eaeh uses images in a fundamentally different manner. Yet, the study of rock
images has fallen to the domain of archaeology because they are found on the surface of
rocks, and not on clothes, eanvases, portable objects, or paper documents loeated in archives
or collections.
1 1 cali, as 1 argued in Chapter 1, the images on rock faces shapes not morphs unlike my predecessors Rusak (1991 b), Pastershank (1989), Rajnovich (1987& 1994), Reid (1979), Lambert (1983, 1985, n.d.), Dewdney (Dewdney and Kidd 1962).
Page 483
Any comprehensive interpretation of visual evidence necessitates crossmg
disciplinary boundaries and dealing with different perspectives. This perspective is defined
and exists in the manner by which an individual perceives and organises the world according
to (a) the way in which the individual has been educated and (b) the way in which the culture
that they belong to organises and perceives this world. Every individual is trained and
educated to organise information into categories so that new ideas, concepts, perceptions,
beliefs, essentially information, can be processed, used and stored in manageable units.
Evidence, whether visual, oral or written, can be interpreted, manipulated, and used in a
variety of ways. Each scholar, regardless of discipline, has a number of generic questions to
ask of documentary evidence. Each discipline employs its own techniques to gather
information, draw on disparate types of information, ask a distinct set of questions, and,
perhaps, arrive at different conclusions.
So, the tobacco is scattered and work can begin. It now remains to summarise the
remaining three sections of this chapter. The first is the summary of the findings and a
discussion of whether the goals established at the beginning of the the sis were achieved; the
second section considers the theoretical and methodological implications of this thesis; and
the third and final section considers further research.
Section 1:
The goal of this the sis was to conduct a systematic, detailed, and thorough
examination of the images of the pictograph sites in the Lake of the Woods region to (a)
as certain a possible vocabulary of images, (b) determine whether combinatory rules exist, (c)
establish the life history of each site, and (d) learn whether the images can be related to other
images outside rock image sites and discover whether this can help provide information
about meaning.
As addressed m the conclusion of Chapter II, the information required to
contextualise the pictograph sites in the Lake of the Woods derives from a variety of
disciplines. Considerable agreement exists about the climatic changes and geomorphologic
history of the region. However historians and archaeologists find it difficult to agree about
who has ever lived there, in part, because each discipline uses different types of evidence to
tackle this problem. Our understanding of the archaeological record of the boreal fore st is,
unfortunately, a "rather nasty, messy, and simplistic culture-history" (Hamilton pers. comm.
Page 484
October 2000) and the paucity of work conducted in this region compared to others in
North America is probably a consequence of the physical isolation, vegetation coyer, the
harsh climate, and the sedimentary depositional environment. Unfortunately, the inadequate
archaeological data from the boreal forest region, in conjunction with the scanty historical
references to the individual groups who lived in this region prior to contact, at contact, and
during the early historic period, inhibits scholars from identifying the terri tories of various
groups. The problems resulting from different names for the same groups could perhaps be
resolved by the careful use of historical documents and the identification of synonymous
designations.
However, during the early seventeenth century, the Siouan-speaking Assiniboine
resided seasonally on the western half of the Lake of the Woods and continued to do so until
the middle of the eighteenth century (Heidenreich 1987: Plate 37). Throughout this period
Algonquian-speaking peoples lived in the eastern portion of the Lake of the Woods and
gradually took control of the region as the Assiniboine moved westwards (Heidenreich and
Noël: Plate 40). From a letter written by La Vérendreye in 1736, scholars know that Siouan
speaking peoples were in the region at the same time as the Algonquian speakers. Since the
early part of the eighteenth century, the Assiniboine were allies of the Cree against the Sioux,
although the Assiniboine are linguistically closely related to the Sioux (DeMallie & Miller
2001: 572). It also, so far as discussed in Chapter II, has proved impossible to determine
which specific groups of Algonquians, Cree or Ojibwa, lived in the region between Lake
Superior and Lake Winnipeg. Both groups shared a similar symbolic system. No solid
evidence exists as to whether the Cree or the Ojibwa created the images at the pictograph
sites of the Lake of the Woods region. It is easier to conclude generically that the
Algonquian-speaking peoples made the se images, than to try to ascertain whether one culture
or the other created them. It is possible to posit that the Algonquian-speaking rather than the
Assiniboine-speaking peoples created them, since the rock images in the Lake of the Woods
strongly resemble those of sites elsewhere in Northwestern Ontario where groups of
Algonquian-speaking peoples are known to have resided (This question was considered in
considerable detai! in Chapter II.). The rock images created by the Assiniboine are very
different from those created by Algonquian-speaking peoples in Northwestern Ontario (e.g.
Keyser and I<lassen 2001; Keyser 1977; Greene 2001). Chapter II also clarified that the
Algonquian-speaking peoples used groups of different images for different purposes than the
Page 485
Assiniboine. The physicallocation of sites in the Lake of the Woods also fits into the general
Algonquian pattern of where rock image sites occur in Northwestern Ontario. 1 argued from
culture-historical data, for using general Algonquian cultural data to interpret the rock
images. This warrants using the AIgonquian ethnographic data to interpret the meaning(s) of
these images, and, hence, justifies using homologies for considering the meaning of the
images which occur in Lake of the Woods.
First Goal:
The first goal of establishing a possible vocabulary for the images, unfortunately, was
not achieved. Chapter III revealed that scholars disagreed on a classification scheme for
describing these images. Description of the images is highly subjective, and the vocabulary
used influences interpretation. Each researcher tends to use his/her own system. Images
often were interpreted rather than described, and, so, 1 decided to think very carefully about
how to record them.
As 1 argue in Chapter IV, the archaeologist should first describe, not interpret the
evidence. This operation involves thinking very carefully about how one describes images,
since any description affects subsequent analysis. It entails not imposing any interpretation
based on their physical shape. The rock image shape must described objectively and in detail.
Images can be organised and discussed by their shape. As 1 argued in Chapters IV
and V, this task is crucial since we lack an understanding of the grammar, punctuation,
sequence of painting episodes, degradation, and the information loss regarding painting
episodes. A rock image must be considered primarily as a shape because this reduces the
possibility of investing description with meaning during the recording stage. Understanding
an image's physical outline, form, and structure before discussing its context or meaning is
vital for archaeologists. When describing a shape, reference its form and nothing more.
Meaning is something to be ascribed at a later stage, using either the analogical or the
homological approach, or a combination of both (see Table 1.1 in Chapter 1).
The discipline inherent in the use of digital technologies plays a vital role lU
marshalling the evidence when this method is employed.
Chapter V discussed and applied the culture-historical approach to this data, and
established the total range of images, drawing on the insights provided in using digital
technologies.
Page 486
Image analysis usmg software, discussed at length in Chapter IV, such as Vasari
Image Processing Software (VIPS) and Adobe Photoshop 6.0 indicated the fragility of the
sites and their images, as weil as their changing nature. Adobe Photoshop 6.0, an image
manipulation software package, revealed that a limited range of motifs appeared in the rock
image sites. It also provided sorne insight into the way hum an beings perceive a physical
shape. Close examination of each image revealed that perception of the precise physical
shape of an image could depend on the viewer, rather than on the presence of red ochre.
The VIPS software permitted me to examine each image individually, and confirmed this
observation. The software revealed that shapes often lacked clear, sharply defined edges. It
highlighted the role that the hum an eye and brain play in the process of recognizing images.
The creator did not deliberately blur the shape's boundaries to impart an 'effect'. Post
depositional image degradation, a component of site taphonomy, is creating 'visual tricks' on
the analyst.
This role is most clearly evident when usmg a procedure described in detail in
Chapter V. U sing the VIPS software, a specific image could be isolated from its context on
the surface of the rock and viewed against a black background, divorced from its original
context. The edges of a group of white pixels that once constituted the red image were not
clearly delineated, and the centre of an image was not solid white; the images were not solid
blocks of red on a grayish granite surface. The VIPS software also helped to establish
whether images were superimposed over white mineraI deposits which, in turn, were
superimposed over other red images at sites such as DhKm-5.
We do not understand the graphic grammar, punctuation, the degree of degradation,
or the sequence of painting episodes. This work emphasises the need for sequence,
objectivity, and the requirement for multiple tools [cross-referencing], and has powerful
implications for a discussion of analytic techniques routinely employed in the analysis of
lithics, bones, sherds and images. Above ail, it emphasises the need for rigour in achieving an
understanding of images as a means of communication.
A total of three hundred and eighteen images were recorded; an average number of
eleven images per site. The deposition of white mineraIs, lichen, and rock tripe
encroachment, and exfoliation of the rock surface that have taken place over time has caused
the images either to fade or be destroyed. These images, or shapes, are not static and
timeless. Given that the images are on rocks exposed to the elements, they start to
Page 487
deteriorate and begin changing shape soon after they are created. Consequently, it is
impossible to establish the full range of combinations of images once present. The narrow
range of image types found in the Lake of the Woods area contrasts sharply with the popular
belief that a relatively large vocabulary of images was used.
It is very difficult to classify and describe any image, and the level of description
given to an image dramatically affects the way in which it can be analysed. It is possible to
classify the images differently. The vocabulary used will influence heavily the outcome of any
discussion of perceived style. It is important to take the middle road between the "splitter"
and the "lumper" approach. A nested hierarchy of types, sequentially examined, probably
provides the most options with respect to subsequent classification. l tried to take a middle
road between the "lump ers" and the "splitters" but large numbers of loosely categorised
types of images e.g. 'creatures' in conjunction with a relatively small quantity of images in the
entire datas et made identifying combinations of specific image types problematic. If aIl the
extremities of a shape such as 'creature' were ignored, it would be possible to simply describe
it as a 'creature.' But this gives the impression that aIl the twenty-three shapes that are
'creatures' are physically the same shape. However, if one closely examines the photographs
of each image (see Appendix 6) and takes into account the appendages and whether or not
the image is hollow it is clear that aIl twenty-three images are slightly different
Second Goal:
The second goal was to determine whether combinatory rules can be established for
the images. l applied the contextual approach after the culture-historical approach to the
information in Chapter VI ta consider this question. It is important to establish whether
correlations occurred and patterns existed amongst the various pieces of data to provide a
solid basis for considering possible meanings. It was crucial to discover any different
relationship types between the physical existence of the sites in the larger landscape, the
physical features of the rock's surface and the images themselves so as to provide a solid
basis enabling the identification of their meaning
The pictograph sites in caves and on cliffs should be perceived as different physical
types, since each has its own characteristics. But for their other characteristics they were
examined together. Granite is generally used for sites but the lack of strong correlation
between the geology, the geomorphology in the immediate vicinity of the sites and the
Page 488
~-
landmass suggests that other reasons may have influenced the choice of a particular place.
Sites occurred on islands and on the mainland but not necessarily immediately beside water,
although water is present in close proximity. The fluctuations in water levels demonstrate
that it is not simply the Images that experience time and change. The rock surfaces
themselves have a history.
None of the sites, whether on cliffs or ln caves, has exactly the same physical
structure. These differences may or may not affect the location of the images. Individuals
such as Conway and Conway (1990b), Dewdney and Kidd (1962 & 1967); Lambert (1983,
1985, n. d.), Rajnovich (1989 & 1994), Rusak (1992a), Vastokas and Vasto kas (1973) drew on
a range of ethnographic and anthropological sources including (Coleman 1937; Fiddler and
Stevens 1995; Landes 1968; Wheeler 1977a, n. d.) to identify the key features of rock image
sites on cliffs: deep holes, large cracks, large crevices, rock overhangs, and rock benches or
ledges. These features were considered by previous archaeologists as prerequisites for the
presence of images. Only one of the twenty-five pictograph sites found on a cliff, DgKl-2,
had ail the physical features that should be present according to previous archaeologists.
Granite cliffs did not need to have sorne large or smail crevices, cracks and holes for
paintings to exist. However, those that drew on ethnographic and anthropological sources in
an attempt to establish the factors ind.icating the presence or absence of images at sites
should be applauded for their efforts, since this was a necessary step in the quest for the
identification of the specific origins of the sites. However, a proposition based on such shaky
grounds is likely to be unsound. Perhaps there is no obvious reason why specific places were
chosen. The answer exists in a body of information either lost in the past or now impossible
to obtain. Although it is fascinating to entertain the notion that sorne sites were secular while
others were sacred, this proposition cannot be considered given the evidence at hand.
A strong relationship could not be established between the images and the types of
rocks. A relationship does not occur between the number of images at sites and the type of
land mass, the number of the physical features present at cliff faces, whether or not offerings
were found in 2001, and the number of cliffs used at each site. Images existed on the ceilings
of both of the caves encountered, but do two caves constitute a representative sample from
which to derive a generalization? The lack of positive or negative correlations between the
number of images and these other features, especiaily o fferings , suggests that other factors
may explain why offerings were found at sorne of the sites in 2001. In retrospect, it would
Page 489
have been useful if l had built a component into my research design that documented how
many localities with the physical criteria (granite cliff, caves, cracks, crevices, or holes, etc.)
do or do not contain rock paintings.
It was clearly important to de termine any relationship between the images, shapes,
themselves. l used 'shape' instead of the term morphs.
The widespread nature of the blobs (xq326), and the five types of 'unidentifiable
shapes' (xq744) throughout the geographical area containing pictograph sites indicates that
the sites have been used over a long time period. Essentially, these pictograph images may
have become indistinguishable because images change shape as they weather, fade, and,
perhaps, are reused. The large number of individual images that only occur once, the large
number of variations of different shapes loosely categorised as 'creatures,' 'stick figures,' 'zig
zags,' 'crescents,' 'hand prints,' turtle,' and 'bird,' as well as the large number of blobs, or
decayed images, suggested that the people who used these images, shapes, knew how to
exploit them as a graphic vocabulary to express whatever they wanted to communicate.
Unfortunately, it is impossible to establish the full range of combinations of images.
However, smears, blobs and unidentifiable shapes cannot be ignored as they were in the pasto
Dewdney (1958-1964) and Lambert (n.d.) were mistaken in ignoring smears and not
perceiving them to be indicators of pictographs. Perhaps these archaeologists did not
perceive such images to be interesting or informative from the interpretive perspective;
hardly a valid reason to exclude them. Although these images are smears, they are still
images. Fortunately, there was a change in perception regarding these shapes prior to Rusak's
(1992b) survey of the White Otter Lake pictograph sites. Otherwise, the smear as a legitimate
analytic category may have been ignored for an even longer period of time. These shapes
demonstrate that people used these rock surfaces to express something, and that significant
time has elapsed since they were originally created. This ide a is supported by the presence of
a large number of images that occurred only once and are discussed in Chapter V.
The large number of types of images, shapes, that were loosely categorised as a type,
such as 'creature,' or 'zig-zag,' in conjunction with a relatively small quantity of images in the
datas et makes identifying combinations of specific image types difficult. Images were
identified as being loosely categorised as a type of shape. As l argued in Chapters IV and V,
it is evident that the vocabulary and the level of detail at which these images were described
Page 490
!~
in the earlier archaeologicalliterature has influenced the number and the range of each type
of shape.
It is possible to classify the images differently. Consequently, demonstrating that the
very vocabulary used has heavily influenced the possible outcome of any discussion of
perceived style becomes straightforward. 1 concluded that a nested hierarchy of types left
more classification options open. For example, if ail the extremities of a shape such as
creature were to be ignored, it would be possible to simply describe the shape as a 'creature.'
Unfortunately, this gives the impression that ail the twenty-three shapes that are 'creatures'
are the same. It is occasionaily impossible to identify the creature's species based on its
appendages. However, if one closely examines the photographs of the images and takes into
account the appendages and whether or not the image is hoilow, this could mean that aIl
twenty-three images are slightly different.
Only twenty-seven sites are currently reported; the fewest shapes per site occurred at
DhKm-19, DiKm-50, and DkKn-6, where only one shape existed. The site with the largest
number of images, thirty-seven, was DhI<:m-5. Blobs were the most common type of shape.
The lack of combinations is not the result of the method used to describe the images
since the differences in the physical shapes of each images were not ignored or glossed over.
These differences are important and the descriptive vocabulary used should be selected
carefully since it affects any discussion of the perceived style. Combinations appeared to exist
when the shapes were loosely described as belonging to a particular group or type of images
(see Appendices 6 and 7). However, but when the shapes in question were described in detail
it became abundantly clear that relatively few shape combinations existed at, or between,
sites (see Appendix 16).
Some images existed in some parts of the lake but not in others (see Chapter VI,
particularly Appendix 16). The widespread prevalence of blobs throughout the lake suggests
that blobs rather than smears may be more important indicators of the passage of time.
Smears occurred at pictograph sites in the southern, eastern and southeastern portion of the
lake. This difference in the geographical distribution between the two shapes supports the
ide a that perhaps some smears were placed intentionally at sites and do not always reflect the
passage of time. Smears might be deliberately executed and not a function of taphonomy
primarily because they, unlike blobs, are not 'randomly' distributed. The largest number of
loosely defined groups of shape types occurred in the eastern portion of the lake where
Page 491
seven different shapes existed: 'a group of nine short lines,' 'a group of four diagonal short
lines,' 'a group of four diagonallines,' 'birds', 'crescents', and the shapes loosely categorised
as 'zig-zags'. 'Creatures' constitute the largest number of any shape type and existed in the
southern, eastern, and northeastern portion of the lake. Most of the images are in the eastern
portion of the Lake of the Woods. Relatively few sites exist in the southern and the northern
portion of the region. Two pictograph sites exist in the northwestern portion of the region
but pictograph or petroglyph sites are not present in the southwestern region, possibly
because the geology is different and the area is swampy, sandy, full of muskeg, and bogs (see
Chapter II and V). The petroglyph sites tended to be more numerous in the northwestern
and northern region of the lake. Perhaps sorne of the images in the northeastern and
northwestern pictograph sites occur in the petroglyph sites of the northeastern and the
northern part of the Lake of the Woods. Confirmation of this speculation can only occur
once these sites have been examined and the images described in the same manner as the
pictograph sites were examined.
The homological, rather than the analogical approach was applied in Chapter VII to
establish whether rules existed. Even though the sites could not be dated using absolute
dating techniques, l chose this approach since cultural continuity between the pre-contact
and post-contact past is a strong possibility in this region, and the same techniques and
implements have been employed for a long period. It was established in Chapter II that the
Algonquian-speaking peoples have been using several image types as a means of
communication.
This approach makes connecting archaeological information with that derived from
other disciplines possible, so as to obtain information concerning the practices and beliefs of
either specifie societies, or groups of historically related societies. By relying on a wide variety
of evidence, archaeologists can attempt to establish continuity in the practices and beliefs in
question.
This approach enabled the in-depth analysis of a specifie group and its material
culture, permitting archaeologists to consider what human beings might have thought about
in the past, and the meaning of specifie symbols and objects. Trigger (1995: 452) argued that
written records, the oral heritage, and ethnographie materials always combined with
archaeological data from successive periods, are probably the strongest types of evidence to
be used when devising bridging arguments. Although this approach poses substantial
Page 492
technical challenges, namely in the use of generic approaches to data, scholars should exploit
these materials to establish which beliefs might be present in specifie cultures. It is important
to determine whether similar images used in a variety of different media held similar
meanings. However, continuity of form does not necessarily imply continuity of meaning.
Meaning can also change, especially sin ce icons tend to be polysemie.
The presence of a large range of different shapes that occur only once (discussed in
Chapters V and VII), causes me to speculate that a considerable number of people probably
moved through this region and used these images to convey information. The Lake of the
Woods, as l have repeatedly argued, is, and was, an important part of a primary
transportation corridor that bisects the continent from east to west and from north to south.
The lack of an apparent pattern in these images possibly reflects the diversity of form, a
variety of functions, unknown but of considerable time depth, and a varied ethnie
authorship. It suggests that these images were part of the larger vocabulary of signs used by
everyone and not solely the product of specialised religious practitioners such as the Midé.
This makes it more likely that each image was created independently and understood
separately from the other images at a pictograph site. It is possible that the organization of
images at sites were believed to have supernatural power, and the combination of images
does not reflect any particular set of rules, other than the fact that individuals, maybe
families, used the same place over an unknown period of time. Perhaps the variability in the
form of images reflects a lack of standardisation. More likely, this variability also reflects a
combination of facts that either the images physically changed over time, or that the artists'
choice of shape, icon or shape changed through time or situational context.
Yet these images probably belong to a group for which little ethnographie data exists,
as l established in Chapter II. This, in comparison with the quantity of information for the
two groups of images (the floral image and those utilised by the Midé), indicates that we
really do not have a strong sense of the combinatory rules we seek. One of the largest
problems with the rock images sites is that there is no base line on which we can build our
analysis. Perhaps we, as archaeologists, are unable to see the pattern(s) because we lack the
intellectual context of symbolic meaning and sufficient knowledge or understanding of all the
taphonomic issues.
Page 493
Third Goal:
It would be a considerable breakthrough to establish each site's life history. This is a
reasonable goal but, unfortunately, it was not achieved. Absolute dating of these sites would
have been useful especiaily where the images were superimposed and badly exfoliated.
Unfortunately, direct dating techniques cannot be used. The presence of datable offerings at
DhKm-1, DgKI-17, and nearby to DiKm-S1 and DiKmSO provided some indication of the
length of time over which these sites have been used. This evidence did not provide a strong
indicator of the age of the images. It remains impossible to establish the precise date at
wruch ail but the most recent of these images were created. Nevertheless, some of the new
shapes couid be given approximate ages. The new shapes in modern paint were
superimposed on ochre between 1989 and 2001 at DgKl-2 because Pastershank did not
record them in 1989, and l found them in 2001. It is possible to date the shapes in red, blue,
and white modern paint of DhKm-20 because their creator conveniently provided a date of
1982. These images strongly resemble those images found in paintings and art objects
created by the members of the New Woodland School of Art, founded by Norval
Morrisseau, who drew on Ojibwa beliefs and traditions. Whether these images reflect the re
use of a place for spiritual purposes by an Aboriginal person, or someone of unknown
cultural affiliation mimicking a widely recognised modern artistic tradition is impossible to
establish. Trus is because it is difficult to reconstruct cultural continuity in the recent past in
trus region.
Otherwise, Dewdney, who first recorded many of the sites, relied on informants,
who often gave no indication of the length of time such sites were believed to have existed.
Lichen growth and encroachment were unreliable indicators, since the rate of growth appears
to differ from site to site, and it is impossible to establish how many archaeologists foilowed
Dewdney's practice of removing the lichen covering many of these paintings. Trus prevents
determining a baseline that would make it possible to know wruch images initiaily existed,
and wruch were subsequently added at each site.
It is also impossible to consider whether the images present upon the pictograph
sites have changed over time. The great variability of images suggests that doing trus
stylisticaily, if at ail possible, will be chailenging. Despite tremendous efforts l found that trus
method was problematic given the incomplete documentation regarding these sites, and the
changing methods used in recording them renders application of this method flawed.
Page 494
It is possible to conclude from the presence of blobs at sites throughout the Lake of
the Woods that some of the images have some antiquity. VIPS software and Adobe
PhotoShop 6.0 provided me with the ability to establish whether additional, paler images
existed beside those that were readily visible. Both highlight the importance of looking
beyond the areas of the surface of the rock that had strong, clearly defined red images, since
paler images could exist elsewhere on the rock surface. Furthermore, the presence of both
the pale and the dark red images in ochre, as weil as the modem spray painted images,
indicates that these places were, and continue to be, used.
Perhaps it is possible to date the constituents of the paints, as discussed in Chapters
II and V, but samples could not be removed. It is impossible ta determine either the
implements used to apply the paint or its ingredients. Fingers were possibly used ta apply the
paint of the images found and recorded before 1989. The ethnographie evidence discussed in
Chapter V supports the idea that iron ore, ochre, was used and possibly isinglass from the air
bladder of the sturgeon, a fish that was prevalent in this region of North America. However,
newer images discovered and recorded in 2001 were painted using modem spray paint in red,
green, black, white, and blue.
Clearly, a shift has occurred in the paint and technique used to create the images.
Fourth Goal:
The final goal was to consider whether the images can be related ta other images
outside rock image sites, and if this can help provide some information about meaning. It
should be clear, from Chapter II, that previous researchers have agreed that these images are
polysemie in nature and that the images were used to convey information to other people.
Previous researchers concurred that the images of the pictograph sites are similar to those on
birch bark scroils made by the Midé ritual specialists. Everyone has posited that a detailed
investigation of the images on birch bark scroils, the ethnographie record, and the
pictographs might provide some answers as to the meanings of the images found on the rock
faces. For this reason l applied, in Chapter VII, the same sequence of philosophical
approaches, with their own questions and techniques, to the images of four birch bark
scroils, each of which was produced for different purposes. However, earlier researchers
made evident that an unknown number of images were used by the Algonquian-speaking
peoples who were not ritual specialists (Coleman 1947; Landes 1968; Schoolcraft 1851;
Page 495
Copway 1851; Dewdney 1975). Unfortunately few of these images exist. Perhaps sorne were
used by Roy, an Ojibwa trader, between 1835 and 1839 (Fulford 1992). Over a hundred years
later Landes (1968: 172) observed that images were familiar to ail Ojibwa hunters, trappers,
and traveilers, who used them to indicate boundaries, trails, notices of different sorts, and on
grave markers. Coleman (1947: 63) had observed, while doing her fieldwork, that the totem
of a deceased individu al was placed upon a grave market.
The results of the analysis of the four birch bark scroils reveals the impossibility of
ascertaining the identity of particular songs, mnemonics, without a recorded body of data
reliably connected to the birch bark scroils in question. However, this study indicated that
the images on the pictograph scroils were used differently. Very few of the images on the
birch bark scroils and the rock image sites are the same. Comparison of the images on the
sites and those of the birch bark scroils reveals little similarity. Comparison of the images
found on the birch bark scroils without contextual documentation and those on the rock
image sites indicates that it is impossible to establish what meaning they held for either their
creators or the people who subsequently observed them. The meaning of the images on the
surface of these rock faces cannot be determined. At the same time, since so few of the
images on the birch bark scroils and the sites are similar, sorne level of choice probably
existed in the mind of the creator of the images, in both place selection in the landscape, and
image selection. It is likely that the creator considered each image as an individual entity and
not as a story. Nonetheless, groups of images at sites are currently being reinterpreted as a
"story," rather than as individual entities (see Section 3 of Chapter VII).
If we view images as individual entities, we can explain why so few of the pictograph
sites that had images had ail of the five "required physical features." Few sites had the
"required features" (crevices, benches, holes, overhangs, cracks) because the variables that
archaeologists have deemed crucial were not important for the creation of images. Such
images were probably created over a very long period of time by individuals and possibly not
by ritual specialists. Relatively few offerings were found in 2001 at sites that had relatively
few of the five physical features, possibly because, at one time, these features were
unimportant. They have subsequentlY become important. It is likely that the significance of
these sites has either changed, shifted, or been mistaken over time. At present, their
importance is not only weil known by archaeologists, anthropologists, and art historians but
also by the local indigenous peoples. People, regardless of whether white or native, read
Page 496
books and articles written by art historians, archaeologists, and anthropologists.
Consequently they learn about the importance given to these sites by the practitioners of
each discipline, and the conversations each has with their respective native informants.
Information about these sites is being recycled and exchanged.
Ethnographic information and date able archaeological evidence support the idea that
specific places in the landscape were important in the pasto The presence of new images on
older images at DgKl-2 indicates that physical places in the landscape continue to be
important, despite the numerous changes within Ojibwa and Algonquian societies over the
last two or three hundred years. The discovery of the two sites with new images, DhKm-19
and DhKm-20, suggests that people still use the landscape as more than a place to travel
through, hunt, or live in, but it is still difficult to establish that specific connection with the
spiritual realm. But it is impossible to establish, however, the precise nature and significance
of these images and the pictographs. It is unclear whether they held any religious symbolic
significance. It is possible that the paintings were simply used as forms of communication.
Section 2: The Theoretical and Methodological Implications of this Thesis
Theoretical Implications:
Theories are considered useful in explaining regularities, but a really useful theoretical
framework also will aid the explanation of the variability evident in the data. Theory is
intertwined with method. The theoretical framework archaeological researchers use affects
the way in which they interpret the pasto Therefore, prior to conducting any new research, an
archaeologist must establish the theoretical approach that each of their predecessors used in
their studies. Previous frameworks affect the way data is collected [fieldwork] and
subsequently analysed. Once completed, it is important to establish whether these
philosophical approaches have affected the methodological approaches. Different theories
require different methods, in both fieldwork and analysis.
In the introduction to this thesis, l argued that archaeologists currently use five
different interpretive approaches. Unlike scholars from other disciplines, archaeologists
usually lack textual data. Textual data is often produced within a particular perspective, and
the investigator can be carry forward that perspective to a greater or lesser extent. These five
interpretive approaches, culture-historical, contextual, intuitive, analogical, and homological,
are used to infer human behaviour and meaning from archaeological data. Each prescribes
Page 497
.~.
the types of questions applied and determines the levels of understanding obtained about
whichever form of archaeological evidence is being considered. The same data must be
examined in sequential order using these different approaches, if the potential quantity and
quality of information gained is to be increased.
The different, but complementary theoretical approaches should be employed in a
definite order. My critical review of the previous research conducted on the rock image sites
of the Canadian Shield, specifically the Lake of the Woods, in Chapter III concluded that the
intuitive approach, (narrative, constructivist, or so-called 'humanist'), must be avoided
because of the problems that arise. 1 think it is imperative for scholars to articulate to others
the method through which conclusions were reached. Readers should not be subjected to the
unstated premises of the author, but shown precisely how an interpretation was processed. If
an archaeologist intends to establish the meaning of these images in a more rigorous and
persuasive fashion, the analogical or the homological approaches should be adopted (see
Table 1.1). As argued in the text, archaeologists should use the following sequence of
approaches: culture-historical, contextual followed by either the homological and analogical
approaches or a combination of the latter two.
Methodological Implications:
The first and most obvious methodological implication of this thesis is that images
must be considered as sources of information in the same way as traditional archaeological
artefacts [perhaps canonical] of the lithic, the bone, and the pot-sherd. Lithics, pot-sherds,
and bones do not occur with neither handbooks nor manuals on how to examine them.
Images should be included as an integral part of any archaeological survey and excavation if
they occur in the geographical region together with "conventional" archaeological sites.
Scholars involved in archaeological research should realise that images, although they are not
like "conventional" archaeological artefacts, still cannot be ignored, or treated in a cavalier
fashion. Researchers called "rock art specialists" must in turn be aware of the importance of
'canonical' archaeological techniques, procedures and approaches. It is vital that both groups
consider an image as a source of information and understand its physical outline, form, and
structure before discussing its context or meaning. Once again, 1 stress the importance of
using the theoretical and analytical approaches in a definite sequence.
Page 498
The second methodological implication is that images cannot be adequately
described using numbers. This is weil understood among those attempting to create adequate
metadating conventions for the description of varied bodies of information. (e.g. ]aritz &
Schuh 1992; Rose 2001; Saouter 1998). Otherwise a 'flat' uniformity is achieved at the cast of
losing distinguishing characteristics. Considerable care must be employed when choosing the
vocabulary to describe the shape or form of an image to avoid inadvertently or prematurely
ascribe meaning(s). This is why l described each shape using a small set of clearly delineated
vocabulary (see Chapter IV). The same vocabulary was re-used repeatedly. Once a shape is
carefuily described, it must be written up, preferably in a database of some type (I used
KLEIO lAS), in a referenced, cross referenced, and searchable form. Ascribing meaning
occurs at a different and [much] later level of analysis. Classifying and describing any image is
very difficult, since the level of description given to an image affects the way in which it can
be analysed. It is possible to classify the images differently, and the vocabulary used will
heavily influence the possible outcome of any discussion of perceived meaning. The
vocabulary and the level of detail has had an impact on both the number and range of each
shape type. Therefore, prior to describing the images, the vocabulary used must be carefuily
considered and discussed. The vocabulary used must not be discipline-specific. It is also
important to take the middle road between the "splitter" and the "lumper" approach when
describing images. l found that a nested hierarchy of types le ft the most classification options
open. The precise size of an image was ignored because these images have changed shape
from the onset of their creation. It was noted whether an image such as a line was longer,
shorter, wider, or narrower than the average size of the image at each site so as to provide an
indication of its size. The image shape was dependant, perhaps, on the amount of paint at
hand. It is difficult to establish importance of image from a detailed exarnination of the
ethnographie record of the peoples who lived in this region. The current size of an image
depends upon uncontrollable elements, such as the deposit of mineraIs upon the rock
surface, exfoliation, and plant growth. Nevertheless, it should be noted subsequent to its
description whether a shape is large or smail with respect to the other images at a site.
However, the precise dimensions of the shape need not be ascertained because, during my
analysis of these images, it became clear that the size of an image was unimportant. It is
possible to state simply that a shape called a smail long horizontal line or short vertical
rectangle existed.
Page 499
It is crucial to describe systemically the images from one end of the painted area to
the other without arbitrarily subdividing the images, shapes, into groups defined by rock
fissures or dis crete perceived spatial clusters. Describing the images in smaller, arbitrary parts
may be easier in the field, but this arbitrary subdivision of sites into parts called faces and
panels gives rise to numerous problems, as discussed in Chapter III. l described the images
at a site from left to right to prevent subdivisions suggested by cracks and crevices in the
rock face. l discuss the fieldwork methods and techniques l employed in my 2001 field
season in Chapter IV. Any short cuts taken during fieldwork inevitably leads to problems
during the analysis stage. Finally, the methodology used to describe and decipher rock images
must be extended in stages, sin ce there is no existing narrative that contextualises it.
Section 3: Further research - What should be done next and which way should the field go?
Rock images must be taken seriously as sources of information; symbolism, and
communication; as important human expressions. The original meanings their creators
applied cannot be established or verified. This should not imply that images be "forgotten,"
left unexamined, and unconsidered as part and parcel of the archaeological record of a given
geographical region. Images must be considered as another archaeological source, like a
lithic, a bone, or a pot-sherd, otherwise the problems that l have discovered in their
description and analyses in Chapter II will persist. Images admittedly require the analyst to
think of the archaeological site not just as a two-dimensional, but as a three-dimensional
entity. The examination of these images, the se archaeological artefacts, requires "thinking out
of the box" to a greater degree than usual and often entails dealing with larger quantities of
information, possibly employing techniques borrowed from other disciplines. It might also
require that the archaeologist acknowledge that it might never be possible to determine their
original intended meaning(s). This should not deter archaeologists from examining images
and regarding them as integral to the understanding of the regional archaeological record.
There are several other serious questions that remain for contemplation - they have
been exposed rather than resolved. Sorne argue that the use of digital technologies will 'solve'
formerly unsolvable problems. That might sometimes be the case but, without rigorous
methodological procedures, much will be lost in digital transcription. If we are to examine
the images using powerful technologies, and if we are to have any hope of relating them to
Page 500
hum an experience, then we must have both the infrastructure, and develop our own tools, or
combine a number of methodological tools for use in conjunction with each other. My
experience would suggest that we need to be free of disciplinary restrictions when choosing
tools. It goes without saying that theoretical frameworks must be thoroughly invoked prior
to using any digital tools. The advent of digital technologies has meant, of course, that there
is the possibility that access via digital surrogates to damaged, faded or partial images may
yield further insights. There is also the remote possibility that an examination of other
circumpolar sites using the homological approach might yield sorne insight, but only to the
site in question, and only after the data has been approached with equal rigour.
A secure grasp of 'sophistication' requires a rigorous intellectual base. This, in turn,
requires an understanding of the intertwining of theory and practice. There is simply no
point in collecting data for its own sake, however intrinsically fascinating, or however 'fun'
the field experience. While such experience is a process in itself, it is inherently
impressionistic, and should not be confused with rigorous research. Rigorous research has its
own delights. On the other hand, theory which has not been grounded in rigorous
archaeological practice and secure data is equally problematic.
This gets us back to 'invented traditions' and perhaps archaeologists, unwittingly or
not, have assisted in the creation of many such traditions. Their practices establish them as
members of a specific community. When scattering tobacco at a site, they are affirming
themselves as potential keepers of important information pertaining to the images. They,
possibly unwittingly, aid in the construction of identities for the peoples of the region, and
their relationship to their pasts. By rigorously adhering to systematic exposition, examination,
and analysis of the rock images of Lake of the Woods, we have established that we remain
ignorant of the role these images played in the lives of peoples in the distant pasto Whatever
role they play today is an entirely different concern.
Page 501
Appendix 1: Summary of the information available on the pictograph sites prior to 2001
Borden Number DiKm-3 Alternative Site # 28 (Dewdney August 15th 1953); Return Point Names: Photographs - Yes - Reid (1975); Dewdney (1953); Molyneaux (1979) were they taken? Measurements Dewdney (1953) - Face l = 50 cm (h), Face III = 100 cm /1 m (h); taken by each Reid - 2 metres square iperson Method( s) of Ashdown (1973) - written description and map; Dewdney (1953)-recording used written descriptions, drawings, photographs;
Reid (1975) recorded accessibility, possible cultural affiliation, significance, condition/ endangerment, assessed "representability, interpretability" of site, obtained Borden number for site under Heritage Act, photographed and mapped site. Molyneaux (1979) - photographs
Location of field Ashdown (1973) - unknown notes Dewdney (1953) - New World Archives, Royal Ontario Museum,
Toronto, Ontario; Reid (1975) Ontario Ministry of Citizenship, Culture and Recreation, Thunder Bay, Ontario; Molyneaux (1979) in possession of the author and Ontario Heritage Foundation
Published works Ashdown, J. H. 1973. A Contemporary Explorer's Chart to the Lake of the where site Woods. Kenora: Royal Yacht Club of the Lake of the Woods. referred to Dewdney S. and K. Kidd. 1962 Indian Paintings of the Great Lakes.
Toronto: University of Toronto Press. Dewdney S. and K. Kidd. 1967. Indian Paintings of the Great Lakes. 2nd
Edition Toronto: University of Toronto Press. Reid, C. S. 1976. Environment, Man and MC!)'mC!)'gwC!)'shi: The Dimensions of Past Human Cultural Impact in the Lake of the Woods Area; Regional Archaeologist, Historical Planning and Research Branch, Northwestern Region, Lake of the Woods Planning Area Molyneaux, B. L. 1979 The Whitqish BC!)' Pietograph Pro/eet. Report to the Ontario Heritage Foundation. No licence.
Scholar's name, J. H. Ashdown (traveller) - 1973; Selwyn Dewdney (artist) - 1953 profession, (August 15); C. S. "Paddy" Reid (archaeologist) - 1975; Brian date(s) of study Molyneaux (archaeologist) - 1979 Location, Dewdney (1953) north west shore of an island in Whitefish Bay, number of faces about 3 miles SSE of Sioux Narrows. He subdivided it into three recorded, etc Faces. Noted current water level in relation to the images he recorded.
Reid (1976: 29) accessed by boat to Whitefish Bay, culturally affiliated to Northern Hunters and Fishers, of probable religious significance, good condition, site well known and published in tourist guides. (ibid.: 14) assessed as having "high representability and high interpretability" .
Rock type? Not stated Offerings Dewdney - clothing items deposited at this site (Dewdney -list in present? Field Book 2), "residue of a "prayer raft" found here - now in the
Royal Ontario Museum" in their Ethnological Collections.
Page 502
Borden Number DiKm-4 Alternative Annie Island, Annie Island Pictograph Site, Site # 93, Annie Island Names: (Dewdney J uly 15,h 1960) Photographs - Yes - (Dewdney 1960); Molyneaux (1974, 1979) were they taken? Measurements Dewdney (1960) - a little over 60"; Reid - 2 metres square; taken by each Iperson Method( s) of Ashdown (1973) - written description and map; recording used Dewdney Guly 1960) - field reproductions, written description,
photographs; Molyneaux (1974) - photographie documentation of site, description of erosion of site vis-à-vis future protection or conservation Pelshea (1978) - field reproduction, slides; Molyneaux (1979) - photographs Reid (1976) recorded accessibility, possible cultural affiliation, significance, condition/ endangerment, assessed "representability, interpretability" of site, obtained Borden number for site under Heritage Act, photographed and mapped site;
Location of field Ashdown (1973) - unknown notes Dewdney (1960) - New W odd Archives, Royal Ontario Museum,
Toronto, Ontario; Reid (1975) Ontario Ministry of Citizenship, Culture and Recreation, Thunder Bay, Ontario. Pelshea (1978) unknown where field notes etc located but photographs in Ontario Ministry of Culture, Tourism and Recreation, Thunder Bay, Ontario. Molyneaux (1974) possession of the author Molyneaux (1979) in possession of the author and Ontario Heritage Foundation
Published works Ashdown, J. H. 1973. A Contemporary Explorer's Chart to the Lake if the where site Woods. Kenora: Royal Yacht Club of the Lake of the Woods. referred to Dewdney S. and K. Kidd. 1962 Indian Paintings if the Great Lakes.
Toronto: University of Toronto Press. Dewdney S. and K. Kidd. 1967. Indian Paintings if the Great Lakes. 2nd
Edition Toronto: University of Toronto Press. Molyneaux, B. L. 1974. Trent University Rock Art Research Pro/ect Summary Report. 1974 Field Season. Department of Anthropology, Trent University. Reid, C. S. 1976. Environment, Man and Mqymqygwqyshi: The Dimensions ifPast Human Cultural Impact in the Lake if the Woods Area; Regional Archaeologist, Historical Planning and Research Branch, Northwestern Region, Lake of the Woods Planning Area. Molyneaux, B. L. 1979 The Whitifish Bqy Pictograph Prqject. Report to the Ontario Heritage Foundation. No licence
Scholar's name, Ashdown (explorer) - 1973; Selwyn Dewdney (artist) - 1960 Guly profession, 15); C. S. "Paddy" Reid (archaeologist) -1976; Victor Pelshea date(s) of study (archaeologist) - 1978; Brian Molyneaux (archaeologist) - 1974,
1979.
Page 503
Location, number West side of Annie Island in southern end of Whitefish Bay; of faces recorded, Dewdney (1960) - 48" above watet; one solitary figure; etc Molyneaux (1974:6) single painting that he argued was "the most
expressed shamanic motif in the rock paintings of the Canadian Shield." Reid (1976: 29) accessed by boat to Whitefish Bay, culturally affiliated to Northern Hunters and Fishers, of probable religious significance, good condition, site weIl known and published in tourist guides. (ibid.: 14) assessed as having "high representability and high intefJ.:>retabili!y".
Rock type? Granite Offerings Dewdney (1960) - an "offering" site. In: Dewdney and Kidd (1967: present? 54) discovered "another deposit of dothing, prayer sticks, and
tobacco, aIl as fresh as if they had been put there yesterday."
Borden Number DhKm-4 Alternative Devil's Bay l Pictograph (Lambert n.d.); Site # 92 (Dewdney 1960) Names:
This site was originaIly part of DhKm-1 according to Dewdney's drawings ofDhKm-1. This fact caused problems for Lambert in 1985 sin ce he combined the recordings, description, and analysis of DhKm-1 with DhKm-4.
Photographs - Yes - Dewdney (1960), Lambert (n.d.), Reid (1976), Molyneaux were they taken? f1974 , 1979) Measurements Dewdney (1960) - Face l = 2 feet taIl by 72 inches long; taken by each pers on Method( s) of Ashdown (1973) - written description and map; recording used Dewdney (1960) - written descriptions, drawings, photographs;
Molyneaux (1974) - photographie documentation of site, description of erosion of site vis-à-vis future protection or conservation Reid (1976) - recorded accessibility, possible cultural affiliation, significance, condition/ endangerment, assessed "representability,
/~, interpretability" of site, obtained Borden number for site under Heritage Act, photographed and mapped site. Molyneaux (1979) - photographs Lambert (n.d.) - written description, photographs, dot-for-dot recording;
Location of field Ashdown (1973) - unknown notes Dewdney (1960) - New World Archives, Royal Ontario Museum,
Toronto, Ontario; Molyneaux (1974) - unknown Reid (1975) Ontario Ministry of Citizenship, Culture and Recreation, Thunder Bay, Ontario; Molyneaux (1979) field notes in possession of the author and Ontario Heritage Foundation Lambert (n.d.) - unknown where field notes and photographs located but Borden forms available from Ontario Ministry of Citizenship, Culture, and Recreation.
Page 504
Published works Ashdown, J. H. 1973. A Contemporary Explorer's Chart to the Lake of the where site Woods. Kenora: Royal Yacht Club of the Lake of the Woods referred to Dewdney S. and K. Kidd. 1962 Indian Paintings of the Great Lakes.
Toronto: University of Toronto Press. Dewdney S. and K. Kidd. 1967. lndian Paintings of the Great Lakes. 2nd
Edition Toronto: University of Toronto Press. Lambert, P. J. n.d. The Northwestem Ontano Rock Art Pro/ect: The 1985 Results and Regional Overoiew. Unpublished report for Northwestern Region Conservation Archaeology Report. No number. Molyneaux, B. L. 1974. Trent Universiry Rock Art Research Pro/ect Summary Report. 1974 Field Season. Department of Anthropology, Trent University. Reid, C. S. 1976. Environment, Man and Mqymqygwqyshi: The Dimensions of Past Human Cultural Impact in the Lake of the Woods Area; Regional Archaeologist, Historical Planning and Research Branch, Northwestern Region, Lake of the Woods Planning Area. Molyneaux, B. L. 1979 The Whiteftsh Bqy Pictograph Prqject. Report to the Ontario Heritage Foundation. No licence
Scholar's name, J. H. Ashdown (traveller) - 1973; Selwyn Dewdney (artist) - 1960; C. profession, S. "Paddy" Reid (regional archaeologist) - 1975; Brian Molyneaux date( s) of study (archaeologist) - 1974, 1979; P. J. Lambert (archaeologist) - 1985
(August) Location, number Dewdney (1960) Whitefish Bay, Lake of the Woods. (Dewdneyand of faces recorde d, Kidd 1967: 53) west shore of the - Face l = 3' (above water?). etc Molyneaux (1974) - no comments about images.
Reid (1976) on Lake of the Woods, north side of entrance to Devil's Bay, Access by boat; southern end ofWhitefish Bay; Reid (1976: 30) accessed by boat to Whitefish Bay, culturally affiliated to Northern Hunters and Fishers, of probable religious significance, good condition, site well known and published in tourist guides, (ibid.: 14) assessed as having "high representability and high interpretability".
Rock type? Dewdney (Dewdney and Kidd 1967: 53) granite outcrop Offerings None present?
Borden Number DhKm-l Alternative Dewdney's # 92 A (1960), Names: Devil's Hole (Reid 1976)
This site was originally part of DhKm-4 according to Dewdney's drawings ofDhKm-4. This fact caused problems for Lambert in 1985 since he combined the recordings, description, and analysis of DhKm-1 with DhKm-4.
Photographs - Yes - colour slides- Dewdney (1960); Fox (1974); Molyneaux were they taken? f1974); Reid (1976); Lambert (1985) Measurements Lambert (1985) taken by each pers on
Page 505
Mcthod( s) of Dewdncy (1960) - writtcn descriptions, drawings, photographs; recording used Ashdown (1973) - written description and map.
Fox (1974) - field notes, photographs; Molyneaux (1974) - photographie documentation of site, description of erosion of site vis-à-vis future protection or conservation Reid (1976) recorded accessibility, possible cultural affiliation, significance, condition/ endangerment, assessed "representability, interpretability" of site, obtained Borden number for site under Heritage Act, photographed and mapped site. Lambert (1985) written descriptions, tracings, photographs
Location of field Ashdown (1973) - unknown notes Dewdney (1960) - New World Archives, Royal Ontario Museum,
Toronto, Ontario; Molyneaux (1974) - unknown Fox (1974) - copies of slides and field notes sent to me by W. A. Ross, Regional Archaeologist, Ontario Ministry of Citizenship, Culture and Recreation, Thunder Bay, Ontario - were lost but were found. Reid (1975) Ontario Ministry of Citizenship, Culture and Recreation, Thunder Bay, Ontario; Lambert (n.d.) - unknown where field notes and photographs located but Borden forms available from Ontario Ministry of Citizenship, Culture, and Recreation.
Published works Ashdown,]. H. 1973. A Contemporary Explorer's Chart to the Lake rifthe where site Woods. Kenora: Royal Yacht Club of the Lake of the Woods referred to Molyneaux, B. L. 1974. Trent University Rock Art Research Prqject
Summary Report. 1974 Field Season. Department of Anthropology, Trent University. Reid, C. S. 1976. Environment, Man and Mqymqyqwqyshi: The Dimensions rifPast Human Cultural Impact in the Lake rifthe Woods Area; Regional Archaeologist, Historical Planning and Research Branch, Northwestern Region, Lake of the Woods Planning Area Lambert, P. J. n.d. The Northwestern Ontario Rock Art Pro/ect: The 1985 Results. Unpublished report. Ontario Conservation Archaeology Report. Northwestern Region No number. Ontario Ministry of Citizenship and Culture.
Scholar's name, ]. H. Ashdown (traveiler) - 1973; W. Fox (archaeologist) - 1974; profession, Brian Molyneaux (archaeologist) - 1974, 1979; C. S. "Paddy" Reid date( s) of study (regional archaeologist) - 1976 Location, number Dewdney (1960) Whitefish Bay, Lake of the Woods. of faces recorded, Fox (1974) none recorded just photographed site and recorded etc details as to offerings found.
Lambert (n.d.) - Face II and III = 36" above water; Molyneaux (1974) - no comments Reid (1976: 29) accessed by boat to Whitefish Bay, culturaily affiliated to Northern Hunters and Fishers, of probable religious significance, still in use by local Ojibway, good condition, site weil known and published in tourist guides. (ibid.: 14) assessed as having "high representability and high interpretability".
Rock type? Not stated
Page 506
~, (
Offerings present?
Borden Number Alternative Names: Photographs -were they taken? Measurements taken by each Iperson Method( s) of recording used
Location of field notes
Published works where site referred to
Scholar's name, profession, date(s) of study Location, number of faces recorde d, etc
Rock type? Offerings present?
Borden Number Alternative Names:
Dewdney (1960) found clothing which he termed offerings. Molyneaux (1974: 9) wrote that site found "immediately north of Site 92, these pictographs occur in relation to a very large crevice, the "Devil's Hole". Although there are scattered paintings, predominately smears of pigment, the focal site of this site is this "hole". During the 1974 field season, articles of clothing were discovered in the crack, presumably left as offerings." Fox (1974) found offerings ofbundles of clothingwhich were stuffed into a cleft in the rock surface of DhKm-1
DhKm-18 Site # 104 or Devil's Bay (Dewdney August 19th 1960), Devil's Bay III (Lambert n.d.), Yes - Dewdney (1960); Lambert (n.d.)
12" tall (Dewdney 1960)
Dewdney (1960) - written descriptions, drawings, photographs; Lambert (n.d.) written description, photographs, dot-for-dot recording; Dewdney (1960) - New World Archives, Royal Ontario Museum, Toronto, Ontario; Lambert (n.d.) - unknown where field notes and photographs located but Borden forms available from Ontario Ministry of Citizenship, Culture, and Recreation. Dewdney S. and K. Kidd. 1962 Indian Paintings if the Great Lakes. Toronto: University of Toronto Press. Dewdney S. and K. Kidd. 1967. lndian Paintings rfthe Great Lakes. 2nd
Edition Toronto: University of Toronto Press. Lambert, P. J. n.d. The Northwestern Ontario Roek Art Pro/eet: The 1985 Results and Regional Overview. Unpublished report. Ontario Conservation Archaeology Report. Northwestern Region No number. Ontario Ministry of Citizenship and Culture. Selwyn Dewdney (artist) - 1960; P. J. Lambert (archaeologist) - 1985 (August)
On west side of Devil's Bayon the south side of the Aulneau Peninsula; Dewdney (1960) - Devil's Bay south end ofWhitefish Bay, 96" above water, 49 degrees 14.6' / 94 degrees 4.6' but ocly one image recorded of a hallow winged bird. Not stated Dewdney (1960) - an "offering" site since offerings left below images and collected the ceremonial sticks as he called them and they are currendy held in the Ethnological collection of the ROM. Also Chinaware and enamelware according to Borden form for this site.
DhKm-5 Site # 105 (Dewdney August 19th 1960); Devil's Bay (Molyneaux 1974); Devil's Bay II (Lambert n.d .. )
Page 507
Photographs -were they taken? Measurements taken by each
1 pers on Method(s) of recording used
Yes - Dewdney (1960); Lambert (n. d.); Reid (1976), Molyneaux
(1974, 1979) -----c~~~~-~---Dewdney (1960) 3.5 feet and 36" (two sections);
Dewdney (1960) - written descriptions, drawings, photographs; Lambert (n.d.) written description, photographs, dot-for-dot recording; Molyneaux (1974) - photographie documentation of site, description of erosion of site vis-à-vis future protection or conservation Reid (1976) recorded accessibility, possible cultural affiliation, significance, condition/ endangerment, assessed "representability, interpretability" of site, obtained Borden number for site under Heritage Act, photographed and mapped site. Molyneaux - photographs. (1979)
Location of field Dewdney (1960) - New World Archives, Royal Ontario Museum, notes Toronto, Ontario;
Reid (1975) Ontario Ministry of Citizenship, Culture and Recreation, Thunder Bay, Ontario. Lambert (n.d.)- unknown where field notes and photographs located but Borden forms available; Molyneaux (1974) not known Molyneaux. (1979) - field notes in possession of the author and Ontario Heritage Foundation
Published works Dewdney S. and K. Kidd. 1962. Indian Paintings rf the Great Lakes. where site referred to
Toronto: University of Toronto Press. Dewdney S. and K. Kidd. 1967. Indian Paintings rfthe Great Lakes. 2nd
Edition Toronto: University of Toronto Press. Lambert, P. J. n.d. The Northwestern Ontario Rock Art Pro/ect: The 1985 Results and Regional Overview. Unpublished report. Ontario Conservation Archaeology Report. Northwestern Region No number. Ontario Ministry of Citizenship and Culture. Molyneaux, B. L. 1974. Trent University Rock Art Research Prqject Summary Report. 1974 Field Season. Department of Anthropology, Trent University. Reid, C. S. 1976. Environment, Man and Mqymqygwqyshi: The Dimensions rfPast Human Cultural Impact in the Lake rfthe Woods Areaj Regional Archaeologist, Historical Planning and Research Branch, Northwestern Region, Lake of the Woods Planning Area. Molyneaux, B. L. 1979 The Whitefish Bqy Pictograph Pro/ect. Report to the Ontario Heritage Foundation. No licence.
Scholar's name, Selwyn Dewdney (artist) - 1960 ; C. S. "Paddy" Reid (archaeologist) -profession, 1976; Brian Molyneaux (archaeologist) - 1974, 1979; P. J. Lambert date(s) ofstudy j(archaeologist) -1985 (August) Location, number South of Devil's Bay in Whitefish Bay; Lake of the Woods, of faces recorde d, Dewdney (1960) - 48" above water and 3.5' above water etc Molyneaux (1974: 7) not record images but commented that there
were "at least two representations of Bisons on this site, a very rare motif in the Shield area".
Rock type? Not stated
Page 508
Offerings present?
Borden Number Alternative Names: Photographs -were they taken? Measurements taken by each
1 pers on Method(s) of recording used
Location of field notes
Published works where site referred to
Scholar's name, profession, date(s) of study
Dewdney - an "offering" site, Offering found below paintings that consisted of prayer sticks and Dewdney and his crew "carefully lifted one corner of the neady piled clothing to find it was all clean and in good repair. No attempt been made to foist off second-rate articles on the mysterious healers" (Dewdneyand Kidd 1967: 55)
DhKn-l Site # 106 (Dewdney August 19th 1960), Obabikon Bay (Molyneaux 1974: 11). Yes - Dewdney (1960); Reid (1976)
Dewdney (1960) - Face 1= 15'16", Face II = 50' long by 2" wide; Molyneaux (1974: 11) - none but speculated that "lizard-like" representation" was over 15 feet long. Reid (1976) - 2 metres square Dewdney (1960) - written descriptions, drawings, photographs; Molyneaux (1974) - photographie documentation of site, description of erosion of site vis-à-vis future protection or conservation Reid (1976) recorded accessibility, possible cultural affiliation, significance, condition/ endangerment, assessed "representability, interpretability" of site, obtained Borden number for site under Heritage Act, photographed and mapped site Dewdney (1960) - New World Archives, Royal Ontario Museum, Toronto, Ontario; Reid (1975) - Ontario Ministry of Citizenship, Culture and Recreation, Thunder Bay, Ontario; Molyneaux (1974) - unknown Dewdney S. and K. Kidd. 1962. Indian Paintings if the Great Lakes. Toronto: University of Toronto Press. Dewdney S. and K. Kidd. 1967. Indian Paintings if the Great Lakes. 2nd
Edition Toronto: University of Toronto Press. Molyneaux, B. L. 1974. Trent Universiry Rock An Research Pro/ect Summary Repon. 1974 Field Season. Department of Anthropology, Trent University. Reid, C. S. 1976. Environment, Man and Mqymqygwqyshi: The Dimensions if Past Human Cultural Impact in the Lake if the Woods Area; Regional Archaeologist, Historical Planning and Research Branch, Northwestern Region, Lake of the Woods Planning Area. Molyneaux, B. L. 1981 Archaeological Fieldwork Report: Licence Number 80-F-0383. Department of New World Archaeology. Toronto: Royal Ontario Museum. Selwyn Dewdney (artist) - 1960; Brian Molyneaux (archaeologist) -1974,1979; C. S. "Paddy" Reid (regional archaeologist) -1976.
Page 509
Location, number On the west side of Miles Bayon the south side of the Aulneau of faces recorde d, Peninsula; Dewdney (1960) - Obabikon Narrows (north end of etc Obabikon Bay, Face 1= 22' above water - "serpent", Face II = 18'6"
ab ove water. Molyneaux (1974: 11) wrote that paintings were divided into "two groups, with the most significant motif being a very large lizard-like representation that is over 15 feet in length." Reid (1976: 30) accessed by boat to Miles Bay, culturally affiliated to Northern Hunters and Fishers, of probable religious significance, good condition, sorne fading. (ibid.: 14) assessed as having "high rep_resentability and high interpretability".
Rock type? Molyneaux (1974: 11) granite Offerings Dewdney (Dewdney and Kidd 1967:45) stated that deep cleft existed present? between ledge and "rock wall that almost filled with dirt and rubble.
Lying on the ground were ancient, weathered overcoat and various rags that had rotted beyond recognition".
Borden Number DhKo-l Alternative Site #108 (Dewdney August 20th 1960); Painted Rock Island Names: 1 (Molyneaux 1974); Painted Rock (Reid 1976) Photographs - Yes - Dewdney (1960); Reid (1976); Molyneaux (1974, 1981) were they taken? Measurements Reid (1976) - 1.5 metres square taken by each [person Method(s) of Dewdney (1960) - written descriptions, drawings, photographs, field recording used reproductions (paintings traced by standing on log held by ropes),
"vigorously sponged off' the lichen (Dewdney and Kidd 1967: 46) to record images. Molyneaux (1974) - photographic documentation of site, description of erosion of site vis-à-vis future protection or conservation Reid (1976) recorded accessibility, possible cultural affiliation, significance, condition/ endangerment, assessed "representability, interpretability" of site, obtained Borden number for site under Heritage Act, photographed and mapped site. Molyneaux (1981) - unknown
Location of field Dewdney (1960) - New World Archives, Royal Ontario Museum, notes Toronto, Ontario;
Molyneaux (1974) - unknown Reid (1975) Ontario Ministry of Citizenship, Culture and Recreation; Thunder Bay, Ontario. Molyneaux (1981) -unknown
Page 510
Published works Dewdney S. and K. Kidd. 1962. Indian Paintings if the Great Lakes. where site Toronto: University of Toronto Press. referred to Dewdney S. and K. Kidd. 1967. Indian Paintings if the Great Lakes. 2nd
Edition Toronto: University of Toronto Press. Molyneaux, B. L. 1974. Trent University Rock Art Research Pro/ect Summary Report. 1974 Field Season. Department of Anthropology, Trent University. Reid, C. S. 1976. Environment, Man and Mqymqygwqyshi: The Dimensions if Past Human Cultural Impact in the Lake if the Woods Area; Regional Archaeologist, Historical Planning and Research Branch, Northwestern Region, Lake of the Woods Planning Area. Molyneaux, B. L. 1981 Archaeological Fieldwork Report: Licence Number 80-F-0383. Department of New World Archaeology. Toronto: Royal Ontario Museum.
Scholar's name, Selwyn Dewdney (artist) - 1960 (August 20); Brian Molyneaux profession, (archaeologist) - 1974, 1981; C. S. "Paddy" Reid (regional date(s) of study archaeologist) - 1976. Location, number On north side of Painted Rock Island on the south side of the of faces recorded, Aulneau Peninsula; Dewdney (1960) - part obscured by lichen, 7 etc feet ab ove water, south of Splitrock Island, ledge = approximately
12 feet above water, exposure of face = 3 degrees east of north, crack in face of rock. Molyneaux (1974: 12) wrote "although many of the paintings are indistinct, there are several particularly fine examples of shamanic figures". Molyneaux (1981) - unknown Reid (1976: 30) accessed by boat to Painted Rock Channel, culturaIly affiliated to Northern Hunters and Fishers, of probable religious significance, still in use by local Ojibway, good condition, weIl known and published in tourist guides. (ibid.: 14) assessed as having "high representability and high interpretability".
Offerings Dewdney (1967 publication with Kidd) no offerings present present? Rock type? Molyneaux (1974: 12) granite
Borden Number DgK1-17 Alternative Cyclone Point Pictographs #1 (Pastershank 1989) Names: Photographs - Yes - Pastershank (1989) were they taken? Measurements Pastershank - 35 metres (h) by 36 metres (w) -check that this site? taken by each person Method(s) of Pastershank (1989) dot-for-dot field recording of site, field recording used reproductions, field notes, photographs. Location of field Pastershank (1989) Ontario Ministry of Culture and Recreation, notes Thunder Bay, Ontario. Published works Pastershank, G. 1989. An Archaeological Surot})! if Sabaskong Bqy, Lake where site if the Woods. Ontario Conservation Archaeology Report #13, North-referred to western Region. Kenora: Ontario Ministry of Culture and
Communications.
Page 511
Scholar's name, Georgine Pastershank (archaeologist / biologist) - 1989 profession, date( s) of study Location, number Southeast corner of Sabaskong Penin sula on Cyclone Point, on well-of faces recorde d, travelled water route from Splitrock Narrows into northeast arm of etc Sabaskong Bay, bedrock ledge; Pastershank (1989) - Face l = an
addition al 5-4 metres past a southward bend - Face II = 7 metres from eastern start of cliff - 21 degrees from vertical, Face III = 7 metres from eastern start of cliff.
Rock type noted? No, just fact that site exist on bedrock Offerings None observed
1 present?
Borden Number DgKl-19 Alternative Cyclone Point Pictographs #2 (pastershank 1989). Names: Photographs - Yes - Pastershank (1989) were they taken? Measurements taken by each Iperson Method( s) of Pastershank (1989) dot-for-dot field recording of site, field recording used reproductions, field notes, photographs, Location of field Pastershank (1989) Ontario Ministry of Culture and Recreation; notes Thunder Bay, Ontario Published works Pastershank, G. 1989. An Archaeological Survry 0/ Sabaskong Bqy, Lake where site 0/ the Woods. Ontario Conservation Archaeology Report #13, North-referred to western Region. Kenora: Ontario Ministry of Culture and
Communications. Scholar's name, Georgine Pastershank (archaeologist / biologist) - (1989) profession, date( s) of study Location, number Located on same bedrock as DgKl-17 only 18 metres west. of faces recorde d, Designated by Pastershank (1989: 67) as separate sites based upon etc different styles, elevations and hues of pigments used which
Pastershank argued indicated the two sites were not contemporaneous. Pastershank (1989) One Face divided into two panels that exist because of a "perpendicular, south-facing hiatus of bedrock, nine-centimetres-wide" (ibid.: 69).
Rock type noted? No, just fact that site exist on bedrock Offerings None observed
1 present?
Borden Number DgKl-2
Alternative Site # 198 and Sabaskong Bay II (Dewdney August 20d 1964); Three Names Caverns Pictograph Site (pastershank 1989) Photographs - Yes - Dewdney (1964); Reid (1976), Pastershank (1989) were they taken?
Page 512
Measurements of Pastershank (1989) - entire site measures 48 metres (h) by 18 metres rock art section (w). Pastershank (1989) - Face II = 290 cm by 107 cm, Face VII = (height by width) 27 cm long iftaken Method( s) of Dewdney (1964) - written descriptions, drawings, photographs. recording used Reid (1976) recorded accessibility, possible cultural affiliation,
significance, condition/ endangerment, assessed "representability, interpretability" of site, obtained Borden number for site under Heritage Act, photographed and mapped site. Pastershank (1989) dot-for-dot field recording of site, field reproductions, field notes, QhotograQhs.
Location of field Dewdney (1964) New World Archives, ROM, Toronto; Reid (1975) Notes Ontario Ministry of Citizenship, Culture and Recreation; Thunder
Bay, Ontario. Pastershank (1989) Ontario Ministry of Culture and Recreation, Thunder Bay, Ontario
Published works Dewdney S. and K. Kidd. 1962. Indian Paintings of the Great Lakes. -where site Toronto: University of Toronto Press. referred? Dewdney S. and K. Kidd. 1967. Indian Paintings of the Great Lakes. 2nd
Edition Toronto: University of Toronto Press. Reid, C. S. 1976. Environmen~ Man and Mqymqygwqyshi: The Dimensions ofPast Human Cultural Impact in the Lake of the Woods Areaj Regional Archaeologist, Historical Planning and Research Branch, Northwestern Region, Lake of the Woods Planning Area. Pastershank, G. 1989. An Archaeological Surory of Sabaskong Bqy, Lake of the Woods. Ontario Conservation Archaeology Report #13, North-western Region. Kenora: Ontario Ministry of Culture and Communications.
Scholar's name, Selwyn Dewdney (artist) - 1964; C. S. "Paddy" Reid (regional profession, archaeologist) - 1976. Georgine Pastershank (archaeologist / date(s) of study biologist) - (1989) Location, number Dewdney (Dewdney (1964) only records existence of one face called of faces recorded, Face l by Pastershank (1989), in Dewdney and Kidd (1967: 110) etc states that "only the finger draggings of a smeared area qualify it as a
pictograph site; but they are associated with a low shallow cave where a large lens of rock softer than the prevailing dark, schistose one that has weathered out". Reid (1976: 30) accessed by boat to Sabaskong Bay, culturally affiliated to Northern Hunters and Fishers, of probable religious significance, good condition, well known and published in tourist guides. (ibid.: 15) assessed as having "high representability and high interpretability" . Pastershank (1989) On east side of Sabaskong Peninsula at south end of NE arm of Sabaskong Bay; at the base of the cliff within the rock shelter = 3 tiny roof-laminated caverns; pictographs are generally west facing but N end of rock shelter bends to make a S-facing panel; 7 faces Face l = on N wall (grotto = 18 cm below paintings; Face II = within rock shelter, Face III = within rock shelter, down and to the left of II, Face IV = within rock shelter, Faces V - VI - VII = on bending section of overhang of rock shelter,
Page 513
Rock type?
Offerings present?
Dewdney (Dewdney and Kidd 1967: 110) dark schistose with larger lens of softer rock whlch has weathered out in low shallow cave
~~
Dewdney (1964) - none Pastershank (1989) - fresh tobacco in small crack 53 cm below figure (Morph 10) on Face II
Borden Number DgKl-l
Alternative Names Photographs -were they taken? Measurements of rock art section (height by width) iftaken Method(s) of recording used
Location of field Notes
Published works -where site referred?
Scholar' s name, profession, date(s) of study
Site # 197 and Sabaskong Bay l (Dewdney August 2nd
Sabaskong Bay Pictograph Site (pastershank 1989) Yes - Reid (1976), Pastershank (1989)
1964); The
Dewdney (1964) - written descriptions, drawings, photographs; Reid (1976) recorded accessibility, possible cultural affiliation, significance, condition/ endangerment, assessed "representability, interpretability" of site, obtained Borden number for site under Heritage Act, photographed and mapped site. Pastershank (1989) dot-for-dot field recording of site, field reproductions, field notes, photographs. Dewdney (1964) New World Archlves, Royal Ontario Museum, Toronto, Ontario; Reid (1975) Ontario Ministry of Citizenship, Culture and Recreation; Thunder Bay, Ontario. Pastershank (1989) Ontario Ministry of Culture and Recreation, Thunder Bay, Ontario Dewdney S. and K. Kidd. 1962. Indian Paintings rf the Great Lakes. Toronto: University of Toronto Press. Dewdney S. and K. Kidd. 1967. Indian Paintings rfthe Great Lakes. 2nd
Edition Toronto: University of Toronto Press. Reid, C. S. 1976. Environment, Man and Mqymqygwqyshi: The Dimensions rfPast Human Cultural Impact in the Lake rfthe Woods Area; Regional Archaeologist, Historical Planning and Research Branch, Northwestern Region, Lake of the Woods Planning Area. Pastershank, G. 1989. An Archaeological Survry rf Sabaskong Bqy, Lake rf the Woods. Ontario Conservation Archaeology Report #13, Northwestern Region. Kenora: Ontario Ministry of Culture and Communications. Selwyn Dewdney (artist) - 1964; C. S. "Paddy" Reid (regional archaeologist) -1976; Georgine Pastershank (archaeologist / biologist) - (1989)
Page 514
Location, number Dewdney (1964) only records existence of two Faces; Dewdney and of faces recorde d, Kidd 1967:110) argues "paintings either side of angular rock. On the etc le ft are two likely bears -regardless of the long ears on the one - and
their tracks, and to the right are canoes and tally marks. Bears and tally marks frequently appear on the Miday scroIls, but never in this kind of configuration. l would judge that both groups are by the same hand." Reid (1976: 30) accessed by boat to Sabaskong Bay, culturally affiliated to Northern Hunters and Fishers, of probable religious significance, good condition, weIl known and published in tourist guides. (ibid.: 15) assessed as having "high representability and high interpreta bili ty" . Pastershank (1989: 51-54) divided this face into three Faces. Situated in central southern most area of a first small jutting protrusion from the north west of Sabaskong Peninsula (ibid.: 51). Most northern of aIl the pictograph sites in Sabaskong Bay.
Rock type? Dewdney (Dewdney and Kidd 1967: 110) schistose
Offerings present?
Borden Number
Alternative Names Photographs -were they taken? Measurements of rock art section (height by width) iftaken Method( s) of recording used Location of field Notes
Published works - where site referred?
Scholar's name, profession, date( s) of study Location, number of faces recorde d, etc Rock type noted?
Offerings Ipresent?
Pastershank (1989: 51) observed that "crack in the bedrock at juncture between Face 1 and Face II that presently being used as a tobacco offering ledge. This crack is located 80 centimetres diagonaIly from the morphs.
DkKn-7
Blindfold Lake Pictograph; Rushing River IV Pictograph Site
Yes -Lambert (.n.d.)
Lambert (n.d.) written description, photographs, dot-for-dot recording Lambert (n.d.) - unknown where field notes and photographs located but Borden forms available from Ontario Ministry of Citizenship, Culture and Recreation; Lambert, P. J. n.d. The Northwestern Ontario Rock Art Pro/ect: The 1985 Results and Regional Overview. Unpublished report. Ontario Conservation Archaeology Report. Northwestern Region No number. Ontario Ministry of Citizenship and Culture. P. J. Lambert (archaeologist) - 1985 Ouly)
One of several sites on Blindfold Lake. Lambert (n.d.) stated on Borden form that not noted by Dewdney and Kidd 1967 and not known by local residents Not stated
Not stated
Page 515
Borden Number DkKn-6
Alternative Blindfold Lake Pictograph; Blindfold Site II, Rushing River III, Names Rummery Site Photographs - Yes -Lambert (n.d.) were they taken? Measurements of rock art section (height by width) iftaken Method( s) of Lambert (n.d.) written description, photographs, dot-for-dot recording used recording; Location of field Lambert (n.d.)- unknown where field notes and photographs located Notes but Borden forms available from Ontario Ministry of Citizenship,
Published works Lambert, P. J. n.d. The Northwestern Ontario Rock Art P,o/ect: The 1985 -where site Results and Regional Overview. Unpublished report. Ontario referred? Conservation Archaeology Report. Northwestern Region No
number. Ontario Ministry of Citizenship and Culture. Scholar' s name, P. J. Lambert (archaeologist) - 1985 Guly) profession, date( s) of study Location, number One of several sites on Blindfold Lake of faces recorded, etc Rock type noted? Not stated
Offerings Not stated present?
Borden Number DjKn-l
Alternative Rushing River l, Site # 29 (Dewdney 1958); Blindfold Lake Names (Molyneaux 1974: 14); The Blindfold Lake Pictograph (Lambert n.
d.) Photographs - Yes - Dewdney (1958); Molyneaux (1974); Reid (1976); Lambert were they taken? ICn.d) of Face l and Face II; Rusak 1993 Measurements of rock art section (height by width) iftaken
Page 516
Method( s) of Ashdown (1973) - written description and map recording used Dewdney (1958) - written descriptions, drawings, photographs;
Molyneaux (1974) - photographic documentation of site, description of erosion of site vis-à-vis future protection or conservation Reid (1976) recorded accessibility, possible cultural affiliation, significance, condition/ endangerment, assessed "representability, interpretability" of site, obtained Borden number for site under Heritage Act, photographed and mapped site. Lambert (n.d.) written description, photographs, dot-for-dot recording; Rusak (1991) photographs and dot-for-dot recording
Location of field Ashdown (1973) - unknown Notes Dewdney (1958) - New World Archives, Royal Ontario Museum,
Toronto, Ontario; Molyneaux (1974) unknown Reid (1975) Borden forms, Ontario Ministry of Citizenship, Culture and Recreation; Thunder Bay, Ontario; Pelshea (1978) unknown where field notes etc located. Lambert (1985) - unknown where field notes and photographs located but Borden forms available from Ontario Ministry of Citizenship, Culture and Recreation, Thunder Bay, Ontario. Rusak (1993) field recordings, field notes in Ontario Ministry of Citizenship, Culture and Recreation, Thunder Bay, Ontario, photographs unknown
Published works Ashdown, J. H. 1973. A Contemporary Explorer's Chart to the Lake if the -where site Woods. Kenora: Royal Yacht Club of the Lake of the Woods referred? Dewdney S. and K. Kidd. 1962 lndian Paintings if the Great Lakes.
Toronto: University of Toronto Press. Dewdney S. and K. Kidd. 1967. Indian Paintings if the Great Lakes. 2nd
Edition Toronto: University of Toronto Press. Molyneaux, B. L. 1974. Trent University Rock Art Research Prqject Summary Report. 1974 Field Season. Department of Anthropology, Trent University. Lambert, P. J. n.d. The Northwestern Ontario Rock Art Pro/eet: The 1985 Results and Regional Overview. Unpublished report for Northwestern Region Conservation Archaeology Report. No number. Reid, C. S. 1976. Environment, Man and Mqymqygwqyshi: The Dimensions ifPast Human Cultural Impact in the Lake if the Woods Area; Regional Archaeologist, HistorÏcal Planning and Research Branch, Northwestern Region, Lake of the Woods Planning Area
Scholar' s name, J. H. Ashdown (traveller) - 1973; Selwyn Dewdney (artist) - 1958; profession, Molyneaux (1974); C. S. "Paddy" Reid (regional archaeologist) -date( s) of study 1976; Victor Pelshea (archaeologist) - 1978; P. J. Lambert - 1985
(July); Jacqueline Rusak (archaeologist) - 1993
Page 517
Location, number Molyneaux (1974:14) stated that the range and the strong colours of of faces recorde d, the paintings at this site "makes it a significant site in the area." etc Reid (1976: 29) accessed by boat or from dirt road from Highway
17, cultural1y affùiated ta Northern Hunters and Fishers, of probable religious significance, good condition and covered by high water in summer months. (ibid.: 14) assessed as having "high representability and high interpretability"
Rock type noted? Not mentioned
Offerings present?
Borden Number
Alternative Names Photographs -were they taken? Measurements of rock art section (height by width) iftaken Method(s) of recording used
Location of field Notes
Dewdney wrote that offerings were found at this site Molyneaux (1974: 14) wrote that the rock shelf at this site was "used for the deposition of offerings, indicating an active, contemporary belief in the sacred qualities of this site".
DiKp-1
Site # 70 (Dewdney Aug 12th 1959); Sunset Channel; 25thJuly 1975 l(Molyneaux 1975 for the TRAP study), Cliff Island Yes - Reid (1976), Molyneaux (1975)
Dewdney (Aug 12th 1959) - 50 cm (h);
Ashdown (1973) - written description and map Dewdney (1959) - written descriptions, drawings, photographs; Reid (1976) recorded accessibility, possible cultural affiliation, significance, condition/ endangerment, assessed "representability, interpretability" of site, obtained Borden number for site under Heritage Act, photographed and mapped site. Molyneaux (1975) assessed to establish the condition regarding moisture access, seepage deposit formation, seepage deposit flaking, exfoliation, and biological growth. Site its physical condition with regard to these criteria. Site location recorded, scale drawings of the images, general site data and condition and description of undertaken. Ashdown (1973) - unknown. Dewdney (1959) - New World Archives, Royal Ontario Museum, Toronto, Ontario; Molyneaux (1975) TRAP Project. Field notes possession of the Canadian Conservation Institute Reid (1975) Borden forms, Ontario Ministry of Citizenship, Culture and Recreation, Thunder Bay, Ontario.
Page 518
Published works Ashdown, J. H. 1973. A Contemporary Explorer's Char! to the Lake of the -where site Woods. Kenora: Royal Yacht Club of the Lake of the Woods referred to? Dewdney S. and K. Kidd. 1962 Indian Paintings of the Great Lakes.
Toronto: University of Toronto Press. Dewdney S. and K. Kidd. 1967. Indian Paintings of the Great Lakes. 2nd
Edition Toronto: University of Toronto Press. Lawson, A. C. 1885. Ancient Rock Inscriptions on the Lake of the Woods. American Naturalist XIX (7): 654-657. Reid, C. S. 1976. Environment, Man and Mqymqyqwqyshi: The Dimensions ofPast Human Cultural Impact in the Lake of the Woods Area; Regional Archaeologist, Historical Planning and Research Branch, Northwestern Region, Lake of the Woods Planning Area. Molyneaux, B. L. 1975. CCI/TRAP Pictograph Condition Inspection Survey. Ottawa: Canadian Conservation Institute.
Scholar's name, A. C. Lawson (geologist) - 1885; Selwyn Dewdney (artist) - 1959 profession, (August 12);]. H. Ashdown (traveIler) - 1973; C. S. "Paddy" Reid date( s) of study (archaeologist) - 1975 Location, number On south side of Cliff Island, Sunset Channel, north on Aulneau of faces recorded, Peninsula, near DiKo-2. etc Reid (1976: 29) accessed by boat to Sunset Channel, culturaIly
affùiated to Northern Hunters and Fishers, still in use by local Ojibway, good condition, weIl known and published in tourist guides. (ibid.: 14) assessed as having "high representability and high interpretability"
Rock type noted? Not mentioned
Offerings Not mentioned present?
Borden Number DhKm-3
Alternative Site # 69 (Dewdney August 8th 1959) Names Photographs - Yes, Dewdney (1959), Reid (1976), Molyneaux (1979), were they taken? Measurements of rock art section (height by width) iftaken Method( s) of Dewdney (August 8th 1959) - written descriptions, drawings, recording used photographs.
Reid (1976) recorded accessibility, possible cultural affiliation, significance, condition/ endangerment, assessed "representability, interpretability" of site, obtained Borden number for site under Heritage Act, photographed and mapped site. Molyneaux (1979) -Photography
Page 519
Location of field Dewdney (1959) - New World Archives, Royal Ontario Museum, Notes Toronto, Ontario.
Reid (1975) Borden forms, Ontario Ministry of Citizenship, Culture and Recreation, Thunder Bay, Ontario. Pelshea (1978) unknown where field notes etc located. Molyneaux (1979) in possession of the author and Ontario Heritage Foundation
Published works Dewdney S. and K. Kidd. 1962. Indian Paintings if the Great Lakes. -where site Toronto: University of Toronto Press. referred to? Dewdney S. and K. Kidd. 1967. Indian Paintings if the Great Lakes. 2nd
Edition Toronto: University of Toronto Press. Reid, C. S. 1976. Environment, Man and Mqymqygwqyshi: The Dimensions ifPast Human Cultural Impact in the Lake iftbe Woods Area; Regional Archaeologist, Historical Planning and Research Branch, Northwestern Region, Lake of the Woods Planning Area. Molyneaux, B. L. 1979 The Whitefish Bqy Pictograph Pro/ect. Report to the Ontario Heritage Foundation. No licence.
Scholar' s name, Selwyn Dewdney (artist) - 1959; C. S. "Paddy" Reid (archaeologist) profession, - 1976; P. J. Lambert (archaeologist) - 1985; Victor Pelshea date(s) of study 1 (archaeologist) - 1978; Brian Molyneaux (archaeologist} - 1979. Location, number Dewdney (1959) Three Faces subdivided by bands of white mineraIs of faces recorde d, on rock surface. Site on Picture Rock Island in Whitefish Bay. On etc cliff face but immediately beside water.
Reid (1976: 30) Southern end ofWhitefish Bay; accessed by boat to Whitefish Bay, culturally affiliated to Northern Hunters and Fishers, of probable religious significance, good condition, site weIl known and published in tourist guides. (ibid.: 14) assessed as having "high representability and high interpretability".
Rock type noted? Not stated
Offerings Dewdney (1959) no offerings present present?
Borden Number DgKm-l
Alternative names Splitrock Bay (Reid 1976)
Photographs - Yes, (Reid 1976) were they taken? Measurements of rock art section (height by width) iftaken Method(s) of Reid (1976) recorded accessibility, possible cultural affiliation, recording used significance, condition/ endangerment, assessed "representability,
interpretability" of site, obtained Borden number for site under Heritage Act, photographed and mapped site.
Location of field Reid (1975) Borden forms, Ontario Ministry of Citizenship, Culture Notes and Recreation, Thunder Bay, Ontario
Page 520
Published works Reid, C. S. 1976. Environment, Man and MqymqyglJJqyshi: The Dimensions
-where site rifPast Human Cultural Impact in the Lake rifthe Woods Area; Regional referred? Archaeologist, HistorÏcal Planning and Research Branch,
Northwestern Region, Lake of the Woods Planning Area. Scholar's name, C. S. "Paddy" Reid (archaeologist) - 1976 profession, date( s) of study Location, number Reid (1976: 31) accessed by boat ta Splitrock Bay, culturally affiliated of faces recorde d, ta Northern Hunters and Fishers, of probable religious significance, etc good condition, site well known and published in tourist guides.
(ibid.: 15) assessed as having "high representability and high interpretability".
Rock type noted? Not stated
Offerings Not stated 1 present?
Borden Number Dewdney Site # 94 - no Borden number
Alternative names
Photographs - Yes, Dewdney (1960) were they taken? Measurements of rock art section (height by width) iftaken Method(s) of Dewdney (1960) - written descriptions, drawings, photographs. recording used Location of field Dewdney (1960) - written descriptions, drawings, photographs; Notes Published works Dewdney (1960) - New World Archives, Royal Ontario Museum, -where site Toronto, Ontario. referred? Scholar's name, Selwyn Dewdney (artist) - 1960 profession, date( s) of study Location, number of faces recorded, etc Rock type noted?
Offerings Offerings in a crack below site (Dewdney and Kidd 1967: 44) 1 present?
Borden Number DiKn-l
Alternative names Hay ter Peninsula, Dewdney # 91
Photographs - Yes, Dewdney - (15th July 1961), Molyneaux (1979), Reid (1976) were they taken?
Page 521
Measurements of rock art section (height by width) iftaken Method( s) of Dewdney (1961) - written descriptions, drawings, photographs; Reid recording used (1976) recorded accessibility, possible cultural affùiation,
significance, condition/ endangerment, assessed "representability, interpretability" of site, obtained Borden number for site under Heritage Act, photographed and mapped site. Molyneaux (1979)
Location of field Dewdney (1961) - New World Archives, Royal Ontario Museum, Notes Toronto, Ontario;
Reid (1975) Borden forms, Ontario Ministry of Citizenship, Culture and Recreation, Thunder Bay, Ontario Molyneaux (1979) - photographs
Published works Dewdney S. and K. Kidd. 1962. Indian Paintings rif the Great Lakes. -where site Toronto: University of Toronto Press. referred? Dewdney S. and K. Kidd. 1967. Indian Paintings rif the Great Lakes. 2nd
Edition Toronto: University of Toronto Press. Molyneaux, B. L. 1979 The Whiteftsh Bqy Pictograph Prrject. Report to the Ontario Heritage Foundation. No licence. Reid, C. S. 1976. Environment, Man and Mqymqygwqyshi: The Dimensions rifPast Human Cultural Impact in the Lake rifthe Woods Area; Regional Archaeologist, Historical Planning and Research Branch, Northwestern Region, Lake of the Woods Planning Area.
Scholar's name, Selwyn Dewdney (artist) - (15th July 1961); Brian Molyneaux profession, (archaeologist) 1979; C. S. "Paddy" Reid (archaeologist) - 1976; date(s) of study Location, number Dewdney (1961) Northeast tip of Hay ter Peninsula at north entrance of faces recorde d, to Whitefish Bay, Lake of the Woods. Three Faces each subdivided etc into parts (a) and (b) because a group of paintings existed above
ledge and group of paintings immediately beside water's edge. Rock type noted?
Offerings Dewdney (Dewdney and Kidd 1967: 52) wrote that found "no present? prayer sticks; but an odd chine [china] cup and other odds and ends
were visible in a horizontal crack nearby".
Borden Number DjKn-2
Alternative names Route Bay (Lambert n.d.), Rushing River III
Photographs - Yes, Reid (1976); Lambert (n.d.) were they taken? Measurements of rock art section (height by width) iftaken
Page 522
Method( s) of Lambert (n.d.) written description, photographs, dot-for-dot recording used recording;
Reid (1976: 14) recorded accessibility, possible cultural affiliation, significance, condition/ endangerment, assessed "representability and interpretability" of site, obtained Borden number for site under Heritage Act, photographed and mapped site.
Location of field Lambert (n.d.) unknown where field notes and photographs located Notes but Borden forms available from Ontario Ministry of Citizenship,
Culture and Recreation; Published works Lambert, P. J. n.d. The Northwestern Ontano Rock Art Project: The 1985 -where site Results and Regional Overview. Unpublished report. Ontario referred? Conservation Archaeology Report. Northwestern Region No
number. Ontario Ministry of Citizenship and Culture. Reid, C. S. 1976. Environment, Man and Mqymqygwqysht~· The Dimensions rfPast Human Cultural Impact in the Lake rfthe Woods Area; Regional Archaeologist, Historical Planning and Research Branch, Northwestern Region, Lake of the Woods Planning Area.
Scholar' s name, C. S. "Paddy" Reid (archaeologist) - 1976; P. J. Lambert profession, (archaeologist) - 1985 Guly) date(s) of study Location, number Reid (1976: 25), accessed by boat or 500 yards through bush from of faces recorde d, dirt road, culturally affiliated to Northern Hunters and Fishers, of etc probable religious significance, good condition, and sorne danger of
defacement as well known. (ibid.: 14) assessed as "high representability and high interpretability"
Rock type noted? Not stated
Offerings Not stated 1 present?
Borden Number Dewdney's Site #95 - no Borden number
Alternative names Picture Rock Point; Site # 95 (Dewdney 1960)
Photographs - Yes, Dewdney (1960) were they taken? Measurements of rock art section (height by width) iftaken Method( s) of Dewdney (1960) - written descriptions, drawings, photographs; recording used Reid (1976: 13) recorded accessibility, possible cultural affiliation,
significance, condition/ endangerment, assessed representability and interpretability" of site, obtained Borden number for site under Heritage Act, photographed and mapped site.
Location of field Dewdney (1960) - New World Archives, Royal Ontario Museum, Notes Toronto, Ontario.
Page 523
Published works Dewdney S. and K. Kidd. 1962. Indian PaintingJ qfthe Great LakeJ. - where site Toronto: University of Toronto Press. referred? Dewdney S. and K. Kidd. 1967. Indian Paintings 0/ the Great Lakes. 2nd
Edition Toronto: University of Toronto Press. Reid, C. S. 1976. Environment, Man and Mqymqygwqyshi: The Dimensions 0/ Past Human Cultural Impact in the Lake 0/ the Woods Area; Regional Archaeologist, Historical Planning and Research Branch, Northwestern Region, Lake of the Woods Planning Area
Scholar' s name, Selwyn Dewdney (artist) -1960; C. S. "Paddy" Reid (archaeologist) -profession, 1976; date( s) of study Location, number Dewdney (1960) Western Peninsula. Reid (1976: 26) by boat to of faces recorded, western Peninsula culturally affiliated to Northern Hunters and etc Fishers, Major pictograph site, still in use by local Ojibway. Fair
condition because local bigots splashed paint when Ojibway use became known. (ibid.: 13) assessed " as high representability and high inteq.,retability"
Rock type noted? Not stated
Offerings Dewdney and Kidd (1967: 44) "found offerings on water lapped present? ledge: neatly folded clothing and a towel, topped by a little pile of
tobacco."
Borden Number No Borden number
Alternative names Site # 270 (Dewdney 1967) on Horseshoe Island
Photographs - Yes, Dewdney (1967) were they taken? Measurements of rock art section (height by width) iftaken Method(s) of Dewdney (1967) - photographs; mentioned in a notebook recording used Location of field Dewdney (1967) - New World Archives, Royal Ontario Museum, Notes Toronto, Ontario. Published works None -where site referred? Scholar's name, Selwyn Dewdney (artist) -1967 profession, date(s) of study Location, number Dewdney - Horseshoe Island, Whitefish Bay of faces recorded, etc Rock type noted? Not stated
Offerings Not stated 1 present?
Page 524
Appendix 2: What sorne of the Tools of Adobe PhotoShop 6.0 can reveal.
This appendix has severai screenshots taken while utilising the different toois
available in Adobe Photos hop 6.0. Each illustrates what happened to the image of DiKp-
1 when examined to establish the edges of the images, including any that may have faded.
The image below of DiKp-l was loaded into Adobe:
Figure Appendix 2.1: DiKp-l
Figure Appendix 2.2: The tiff file called DiKp-l as seen once loaded in the software.
It was subsequently examined, once it had been zoomed in using the tool called hue and
or saturation and the following screenshot resulted.
Page 525
Figure Appendix 2.3: The shapes visible using the desaturate tool.
It is clear in the screenshot above that the desaturate tool highlights the presence of red
pigment on the surface of the rock.
Figure Appendix 2.4: Manipulation of the colour red using the red colour channel.
Here colour is removed ta see whether the edges of the red shapes can be determined.
The brightnessj contrast tool was the next tool to be used. The screenshot below (figure
appendix 2.5) is the same tiff before it was manipulated using the brightness and contrast
tool.
Page 526
Figure Appendix 2.5: The tiff when it was being manipulated.
The screenshot below (figure appendix 2.6) can occur once with the manipulation of the
brightness/ contrast tool. The red image on the rock surface becomes very visible since
the tool highlights the differences.
Figure Appendix 2.6: The colour is removed.
The screen shot below (figure appendix 2.7) is what the tiff fùe of DiKp-l might look
like if the levels of the red channel of the colour balance tool are manipulated and either , reduced or increased beyond its level when the fùe was imported into the software.
Page 527
~ ..
Figure Appendix 2.7: The image me during the process ofusing the red colour channel.
The red colour channel is very useful as the screenshot below (figure appendix 2.8)
illustrates since the redness of the ochre images becomes highlighted and emphasised
distinguishing itself From the red of the granite.
Figure Appendix 2.8: The colours are being manipulated using the red colour channel.
The green colour channel is useful as the screenshot below (figure appendix 2.8)
demonstrates since the red images are highlighted.
Page 528
.r---. Figure Appendix 2.9: The use of the green colour channel.
The green colour channel is equally as useful as the red colour channel but it is clear if the
figure below is compared with the one above that a small decrease in the green colour
causes the red images on DiKp-l to have clearer outlines (see figure appendix 2.10).
Figure Appendix 2.10: The images have clearer edges.
The blue colour channel was subsequently used to examine these red images and to
determine whether they were clearer or harder to distinguish by altering the quantity of
blue in the tiff flle.
Page 529
Figure Appendix 2.11: The images have clearer edges.
The figure above illustrates the shapes of the pictographs at the maximum level of colour
saturation using the blue colour channel. While the figure below shows what occurred to
the same images when the levels of blue were considerably reduced.
Figure Appendix 2.12: The same images.
Page 530
Appendix 3: A Blank Copy of Form A and Form B for Sites on CHffs and Caves.
Form A for Both CHffs and Caves:
Borden Number Recorder Date
Place Name Researcher's site number Alternate Site Names Previous Site Names Spatial Co-ordinates (lat/long) NTSmap Elevation Geographic Notes Present Day Nearby Settlements Land Masses Access Regional Drainage Water Systems (direction of flow) Uses of Body ofWater Period of use for body of water (in years) Presence of Water (on # of sides) Body of water surrounding/ adjacent to site Freeze/Partially freeze in winter Water Level Fluctuation (yes/no) Cause of water level fluctuation Extent of Fluctuation (inches per annum) Control ofWater Levels by whom/what? Se as on ofWater Level Fluctuation Major Changes in Regional Water Level (in years) Visible Water Levels (number, height) Location (cliff faces/rock outcrops/ caverns/ out of water) General Notes
Form B for Pictographs on Cliff Faces:
Borden Number
Recorder Date Units of measurement Type of rock Number of cliff(s)
Page 531
Height of top of cliff(s) from present day water level Number of visible water lines Height of top of cliff(s) from visible water line Vegetation of top of cliff(s) Vegetation on side/ surface of cliff(s) Vegetation on base of cliff(s) Rock overhangs Proximity of rock overhangs to paintings Large cracks Proximity of large cracks to paintings Large crevasses Proximity of crevasses to paintings Rock shelf/ shelves Proximity of rock shelf/shelves topaintin~s Rocks present at base of cliff(s) Offerings Number of angles of painted face + location (base, middle, top) Orientation of angles (NSEW) Methods of Recording General Notes (exfoliation, white mineraI deposit, lichen and rock tripe encroachment)
Form B for Pictographs inside Caves:
Borden Number
Recorder Date Units of measurement Type of rock Number of cave(sl refer to field drawing Height of cave from p~esent day water level Number of visible water lines Height of top of cave from visible water line Width of cave Vegetation of to~ of cave Vegetation on side/ surface of cave Vegetation on base of cave Rock overhangs Large cracks Proxirnity of large cracks to paintings Large crevasses Proximity of crevasses to paintings Rock shelf/ shelves
Page 532
Rocks present at base of cave Offerings Orientation of cave opening(s) Method(s) of recording General Notes (exfoliation, white mineraI deposit, lichen and rock tripe encroachment)
Page 533
Appendix 4: The Mod File used by Kleio lAS
Below is the physical structure of the data in the DAT file. The structure of the data
represents the logical relationship between the parts of the data. The structure starts out with
the most abstract level at the top, and becomes more concrete. The elements must be
identified first.
data base name=design;first=i;overwrite=yes;write=structure element name=assigned_no;image=digital;type=image element name=access;write="access_to_site" element name=alcname;write= "alternative_name" element name=annotation;write="interesting .. jnfo_re_tale" element name=author;write="author_oCtale" element name= bhosadsi;write="body _oC watecsurrounding/ adjacencto_site" element name=birchbark;write= "birchbark" element name= bark_contxt;write=" contexC oCbirchbark" element name=bordenn;write="borden_number" element name=claimed_status;write=" specific........geographical_region" element name=colour;write="colour" element name=date;write="date_oCfield_recording" element name=date_info_collected;write="date_information_collected" element name=exfoliat;write= "exfoliation"; element name=extent;write="extent_of_fluctuation" element name=fielinfo;write="fieldnotes" element name=format;write=" format" element name=freefw;write="freeze/partially_freeze_in_winter" element name=gen_notes;write=" general_notes" element name=geo~not;write="geographical"notes" element name=group_belon~to;write="group_tale_belon~to" element name= hcavprwl;write= "heighC oCtop _ oCcavern_from_presencday _ wateclevel" element name=hcliprwl;write="heighcoCtop_oCcliffs_from_present_day_wateclevel" element name=holefluc;write="maintenance_oCcurrencwater" element name= horeflec;write="water_reflection_on_rocks( clifConly)" element name=id element name=idprofyr;write=" scholar' s_name,_profession_and_date( s L oCstudy" element name= l_crevas;write= "large_crevasses" element name= landmass;write="land_masses" element name= larcraks;write= "large_cracks" element name=lac1ong;write="latitude_and_longitude" element name= location;write= "location_ oCimages" element name=maintyr;write= "maintenance _ oCcurrent_ water_leveCyears" element name=majchhol;write="major_changes_in_regionaCwater_levels_years" element name=mater_used;write="materials_used_to_create_image" element name=meascav;write="measurements_of_cavern_if_taken" element name=measclif;write="measurements_oCclifCiCtaken"
Page 534
element name=meth_rec;write="methods_oCrecording" element name=scrollid;write= "birch_bark_scrolLidentity _num ber" element name=n_cavern;write="number_of_cavern" element name=n_cliffs;write="number_oCcliffs" element name=n_tclvwl;write= "heighcoCclifCfrom_ visible_ watecline" element name=n_ vwlin;write= "number_oC visible_ water_lines" element name=no_angle;write="number_oCangles"; element name=no _story _pub;write=" numbec oCstories_published" element name=offering;write=" offerings" element name=offnoted;write="offerings_noted" element name=ori~ed;write="original_edition" element name=ori~pdate;write="originaLpublication_date" element name=ori~pub;write=" original_publisher" element name=ori~title;write="originaLtitle" element name=page;write+ "page_to_page" element name=personid;write="identity_oCperson" element name=phychart;write=" physical_ characteristics" element name=pLname;write= "place_name" element name=prealtname;write="previous_alternative_name" element name=predpeop;write=" presencday _peoples" element name=predsett;write= "presencday _settlements" element name=preduse;write="period_oCuse_in_years" element name=prehosid;write= "presence_oC water-numbecoCsides" element name=primary _editor;write= "primary _editor" element name=projname;write= "projeccname" element name=prxpaint;write=" proximity _ oCrock_ overhangs_to paintings" element name=prxrochg;write="proximity_oCrock_overhangs_to_one_another" element name=pub_ work;write="published_ work" element name=px_cre;write="proximity_oCcrevasses_to_another" element name=px_lc;write="proximity_oClarge_cracks_to_one_another" element name=px_lcpai;write="proximity_oClarge_cracks_to_paintings" element name=rcprcsh;write="rocks_on_clifCshelf/ shelves" element name=regdrsys;write=" regionaLdrainage_ water_systems" element name=rocbacav;write="rocks_acbase_oCcavern(s)" element name=rocbacli;write="rocks_acbase_oCcliff(s)" element name=rockfall;write="rock_fall" element name=rocovhgs;write="rock_overhangs" element name=rocprecre;write=" rocks_presencin_ crevasses" element name=rocshelf;write"rockshelf' element name=roctacli;write="rocks_actop_oCcliff(s)" element name= roctocav;write = "rocks_at_top_oCcavern(s )" element name= season;write=" season_ oC waterlevel_fluctuation" element name= second_editor;write=" secondary _editor" element name= second_pdate;write=" secondary _publication_date" element name= second_pub;write=" secondary _publisher" element name=shape;write="shape_oCtopos" element name=tiffnums;write="tifCftle_numbers" element name=type_ data;write="type_oCdata_recorded"
Page 535
element name=typerock;write=ltype_oCrock" element name:=urut_m;write=" urut_ of_measurement" element name=useofho;write= "uses_oCbody _oC water" element name=vegbcav;write:= "vegetation_acbase_oCcavern( s)" element name=vegbcli;write= "vegetation_acbase_ ° Ccli ff( s)" element name=vegsscav;write:="vegetation_on_side/ surface_oCcavern(s)" element name=vegsscli;write:= "vegetation_on_side/ surface_oCcliff( s)" element name:=vegtcav;write:="vegetation_actop_oCcavern"; element name=vegtcli;write="vegetation_actop_oCcliff(s)" element name=visiwale;write="visible_ wateclevels"; element name=why;write=" cause_of_ water_leveLfluctuation" element name=projname;write="project_name" element name=persorud;write= "identity _ oCperson" element name:=date;write=" date_oCfield_recording" element name= bordenn;write= "borden_number" element name=urut_m;write="urut_of_measurement" element name=measclif;write="measurements_oCclifCiCtaken" element name= fielinfo;wri te=" fieldnotes" element name=pub_ work;write="published_ work" element name=idprofyr;write=" scholar' s_name:>-profession_and_date( s L oCstudy" element name=phychart;write="physicaLcharacteristics" element name=typerock;write="type_oCrock" element name=offnoted;write="offerings_noted" element name=shape;write=" shape_oCtopos" element name=mater_used;write="materials_used_to_create_image" element name=centre_ oCimage;write=" centre_ oCimage" element name=shape;write="shape_oCtopos" element name=mater_used;write="materials_used_to_create_image" element name=centre_oCimage;write="centre_oCimage" element name=shape;write="shape_oCtopos" element name=mater_used;write="materials_used_to_create_image" element name=centre_oCimage;write="centre_oCimage" element name=projname;write=" project_name" element name=personid;write="identity_oCperson" element name=tiffnums;write="tifCfùe_numbers" element name= bordenn;write= "borden_number" element name=date;write=" date_ oCfield_recording" element name=pl_name;write="place_name" element name=alt_name;write=" alternative_name" element name=pre_name;write="previous_name" element name=lac1ong;write="latitude_and_longitude" element name=geo~not;write="geographical"notes" element name=predsett;write= "presencday _settlements" element name=predpeop;write= "presencday _peoples" element name= landmass;write= "land_masses" element name=access;write=""access_to_site" element name=regdrsys;write=" regionaLdrainage_ water_systems" element name=useofho;write="uses_oCbody _oC water"
Page 536
r--..
element name=preduse;write="period_oCuse_in_years" element name=prehosid;write= "presence_ oL water-number_ of_sides" element name= bhosadsi;write= "body _oL watecsurrounding/ adjacent_to_site" element name=maintyr;write="maintenance_oCcurrencwater_leveLyears" element name=holefluc;write="maintenance_oCcurrencwater" element name=why;write="cause_of_water_level_fluctuation" element name=extent;write="extencof_fluctuation" element name=season;write=" season_ oC waterleveLfluctuation" element name=majchhol;write="major_changes_in_regional_water_levels_years" element name=visiwale;write="visible_ wateclevels" element name= location;write= "location_ o Limages " element name=projname;write="project_name" element name=personid;write=" identity _ oCperson" element name=date;write=" date_oCfield_recording" element name=tiffnums;write="tifCfile_numbers" element name= bordenn;write= "borden_number" element name=unit_m;write="unit_of_measurement" element name=typerock;write="type_oCrock" element name=n_cavern;write="number_of_cavern" element name=cavprwl;write= "heighC oCtop _oC cavern_from_presencday _ water_level" element name=n_ vwlin;write="number_oC visible_ water_lines" element name=n_tcavwl;write= "height_ oCcave_from_ visible _ watecline" element name=vegtcav;write="vegetation_actop_oCcavern" element name=vegsscav;write= "vegetation_on_side/ surface_oCcavern(s)" element name=vegbcav;write= "vegetation_acbase_oCcavern( s)" element name=rocovhgs;write=" rock_ overhangs" element name=prxrochg;write="proximity_oCrock_overhangs_to_one_another" element name=prxpaint;write= "proximity _oCrock_ overhangs_to paintings" element name=larcraks;write="large_cracks" element name=px_lc;write= "proximity _ oClarge_cracks_to _ one_another" element name=px_lcpai;write="proximity_oClarge_cracks_to_paintings" element name= Lcrevas;write= "large_crevasses" element name=px_ cre;write= "proximity _oC crevasses_to _another" element name=rocshelf;write"rockshelf' element name=rocbacav;write="rocks_acbase_oCcavern( s)" element name=roctocav;write="rocks_at_top_of_cavern(s)" element name=rcprcsh;write= "rocks_on_clifCshelf/ shelves" element name=rocprecre;write=" rocks_presencin_crevasses" element name= offering;write = "offerings" element name=oritangle;write="orientation_oCangles" element name=rockfail;write=" rock_fail" element name=exfoliat;write="exfoliation" element name= gen_notes;write=" generaLnotes" element name=projname;write="project_name" element name=personid;write= "identity _oCperson" element name=date;write="date_oCfield_recording" element name= bordenn;write= "borden_number" element name=unit_m;write= "unit_ of_measurement"
Page 537
element name=meascav;write="measurements_oCclifCiCtaken" element name=fielinfa;write="fieldnates" element name=pub_ work;write= "published_ work" element name=idprofyr;write=" scholar' s_name,_profession_and_ date ( s LoCstudy" element name=phychart;write="physicaLcharacteristics" element name=typerock;write="type_oCrock" element name=offnoted;write="offerings_noted" element name=shape;write="shape_oCtopos" element name=mater_used;write="materials_used_to_create_image" element name=centre_oCimage;write="centre_oCimage" element name= shape;write= "shape_oCtopos" element name=mater_used;write="materials_used_to_create_image" element name=centre_oCimage;write="centre_oCimage" element name=id element name=format;write="format" element name=date_info_collected;write="date_information_collected" element name= second_pdate;write=" secondary _publication_date" element name=author;write="author_of_tale" element name=primary _editor;write= "primary _editar" element name=type_ data;write="type_oCdata_recorded" element name=page;write+"page_to_page" element name=annotation;write="interestingjnfo_re_tale" element name=group_belong_Jo;write="group_tale_belong_Jo" element name=claimed_status;write=" specific~eographicaLregion" element name=id element name=m_content;write="myth_content" element name=scene;write="scene" element name=people element name= sex;write=" sex_ oCdepicted_person" element name=m_pname;write="name_oCpersons_in_myth" element name=m_associated_characteristics;write="associated_characteristics_in_myth" element name=m_status;write= " status_oCperson_my th" element name=m_clothing;write="clothingjn_myth" element name=m_scene;write=" scenes_in_myth"; element name=action element name=setting element name=person;write="person" element name=id element name= format;write = " format" element name=date_info_collected;write="date_information_collected" element name=author;write="author_oCtale" element name=primary _ editor;write= "primary _editor" element name=type_ data;write="type_oCdata_recorded" element name=page;write="page_to_page" element name=annotation;write=" interestingjnfo _re_tale" element name=group_belong_Jo;write="group_tale_belong....to" element name=claimed_status;write=" specific~eographicaljegion" element name=id
Page 538
element name=form_oCobject element name=collection element name=archive element name=ori~collect;write="original collector" element name=why element name=collected_date element name=author element name=visible element name=group_belon~to element name=claimed_status element name=materials_used_to_construct_object element name=shape;write=" shape_oCtopos" element name=mater_used;write="materials_used_to_create_image" element name=centre_oCimage;write="centre_of_image" element name=shape;write="shape_oCtopos" element name=mater_used;write="materials_used_to_create_image" element name=centre_ oCimage;write=" centre_ of_image" element name=o_content;write="oraLevidence_content" element name=o_scene;write="scenes in_oraLevidence";order=multiple element name=o_people;write="people_in_oraLcontent" element name=o_sex;write="sex_oCdepicted_characters";type=category;category=gender element name=o_pname;write="name_oCperson" element name=o_associated_characteristics;write="associated_characteristics_o" element name=o_status;write="status_of_person_oral" element name=o_clothing,o_status; element name=o_action;write="oraLsituation" part name=i;write="image";
position = number;arbitrary=rock_art,birchbark,second_data part name=rock_art;arbitrary=rock_face,cave_face part name=rock_face;arbitrary=rphy _cont,rcuc work,rprev _ work,rimage part name=rphy_cont;write="rock_face_physical_context";
"Prayer Sticks" (26 bundles of 12, 12 cm = 101 (E),5
1 small long by 12.5 - 17 cm wide, tied together = 139 (SE),
6 = 120 blob,4 with white cotton string, half of each stiek
(SE), 7 = 99 small painted blue and the other half painted red,
(E),8 = 150 blobs large diameter sticks always tied together
alongside and small diameter sticks always tied Pastershank DgKl-
No Yes No Yes (SE), 9 =
and 9 images together - never mixed, 1 bundle of 12 in in 1989 -2 140 (SE), 10
attached to the left-hand cave, refer to field sketches tobaeco = 100 (E), 11
each other, and digital and manual photographs). More = 120 (SE),
12 = 120 1 offerings were left on the rock bench under
(SE), 13 = unidentifia overhang when came back later to record
160 (SSE), ble shape more photographs, modern paintings also
1 14 = 130
visible at site. Returned August 5th 2001 with a team of divers who discovered the
(SE), 15 = rocks immediately below and the vieinit)'
1 189 (S\'V) around the rock image site. They 1
discovered that it was a mud bottom and i
that there was a pile of rocks but no i offerings were evident only Mountain Dew 1 soft drink can, a gendeman's bicycle, and
another cave to the right and 2.14m (7 feet)
1 below site un der water.
Page 543
') ')
Tobacco, money (coins) - on shelf/ crevasses. August 5/01 = (with
divers) Medicine sticks found under water in same bundles = 12 sticks tied together
with string, red and blue just like site
DgKl- 1 -109 immediately opposite Nestor Falls - are a
None 17
Yes Yes No Yes degrees (E)
No blobs 13 images couple of feet below the water level but not recorded
immediately beside/below rock art site; Broken plate but can't find other part -same maker's mark as Horseshoe Island pot - Alfred Meaken; child's fishing rod;
i
collection of shirts about 6.1 m (20 feet) among large boulders below the water level
June 12/01 - multiple articles of clothing (socks, shirts, jeans, jersey shirt, polyester sport shirt, aIl colours) - aIl above water level , stuffed in crevasses and cracks as
weIl as numerous sample-size alcohol bottles, coins, tobacco, empty small-size soft drink bottle, toweL July 22/01 (with
with two finished - no seam at top, cork with hand-
clothing. tooled top, has a smell when opening
DhK 110 degrees short
container - juniper smell, on bottom 25%, Molyneaux
m-l Yes Yes Yes Yes
(ESE) vertical 17 images
wedged in groove in rock about 1.52 - 1.83 in 1974-
1
lines m (5-6 feet) from pot), lots of coins in the
clothing; sticking up Fox 1974-
on top rock (George V quarter, American penny),
bundles of right hand
plate made by same company as bowl from clothing;
new Horseshoe Island site - Alfred side
Meaken, Royal Ironstone China, England, Queen Victoria's crown, lion and unicorn) (AC). Huge pile of miscellaneous clothing, different styles and different periods to the depth of 15.85 m (52ft) from the surface of
the water. lncluded fabric and leather clothes, moccasins, pile leaning against cliff
1 face.
Page 545
\ )
Stick, peeled, which looks !ike beaver stick but ends are a bit weathered so hard to be 100% sure - 3/4" diameter and 1/5' long. Tobacco between 2 and 3. Two cowrie
shells = one 7.5 cm - other 5 cm (hidden in sedge). One pie ce of red clotho One cap
DhK Yes Yes No Yes
150 degrees 10 blobs 33 images
(red) embroidered "Sail Caribbean" - None m-3 (SE) deteriorating cloth, very faded, not recently recorded
left (probably over 1 year old). Ali on rock shelf. Soil under areas 2 and 3 has been
disturbed recently (by stick?) and between a bush at north-east edge, also under large cowrie shell and one small area of moss 6
inches.
1 = 200
DhK degrees
None m-4
No Yes No No (SSW), 2 = No blobs 5 images No recorded
182 degrees (S)
10 different types of
Difficult to blobs,l
estab!ish group of
Dewdney in 1
because of blobs of
1960 found DhK Yes accumulatio Yes Yes 120 degrees different
37 images Yes - coins (coin) on 2ledges, tobacco on 2
m-5 (ESE) sizes, 1 ledges, decayed fruit - food prayer
n of white sticks and mineraI
round none later
cleposits blob with short !ines
sticking out either
sicle
Page 546
î ')
1 blob, one faint
unidentifia ble shape, a group of Dewdney in 1
DhK 230 degrees small 1960 found '
m-18 Yes Yes No No
(SW) blobs and 6 shapes No prayer
! lines that sticks and cross each none later
other at 1
ninety degree angles
Left side = Dewdney in
250 degtees 1960 found overcoat DhKn
No Yes No No (WSW),
2 blobs 10 images Yes - Right side = feather Oarge bird),
and various -1 Right side = partially smoked cigarette weathered
240 degrees and rotten
(WSW) "rags" ,
y es = in the various cracks in the cliff face Dewdney in righ t cliff - tobacco (5 piles), cigarettes, snuff can lids, 1960 found
119 (SE); left 8 blobs, 1
3 sample-size Crown Royal botdes, alcohol chinaware DiKn-
Yes Yes Yes No cliff - 1 Oeft)
collection 16 images botdes, clothing, money (25 cents = 1977, and other
be identified as rock Mainland - Aulneau is identified as
outcrop which is ice Peninsula - Hay ter DiKn-l hornblende, gabbro,
eroded Archean granite Peninsula diorite, periodite, and
with patches of thin serpentine. None of
drift and scattered them is the same age.
boulders.
Archean Early More than eighty Precambrian, schistose percent of the rock can
volcanic with some be identified as rock small quantities of outcrop which is ice
Island - Cliff Island DiKp-l sediments that are iron eroded Archean granite
formation. Y ounger with patches of thin sediments possibly drift and scattered
exist here. boulders.
Archean Early Precambrian but rock
Glaciofluvial deposits of is identified as
sand and gravel measure Mainland DkKn-6 hornblende, gabbro,
between one and three diorite, periodite, and
metres in depth. serpentine. None of
them is the same age. Archean Early
Precambrian but rock Glaciofluvial deposits of
is identified as sand and gravel measure
Mainland DkKn-7 hornblende, gabbro, between one and three
diorite, periodite, and metres in depth.
serpentine. None of them is the same age.
Archean Early Drift and rock
Precambrian but rock consisting of till and
is identified as stratified deposits that
Mainland DjKn-l hornblende, gabbro, measure between one to
diorite, periodite, and three metres thick in
serpentine. None of depressions.
them is the same age.
Page 567
More than eighty Archean Early
percent of the rock can Precambrian but rock
be identified as rock Island - Unnamed is identified as
outcrop which is ice island across from DiKm-3 hornblende, gabbro,
eroded Archean granite Tamarack Island diorite, periodite, and
with patches of thin serpentine. None of
drift and scattered them is the same age.
boulders. More than eighty
Archean Early percent of the rock can
Precambrian but rock be identified as rock
is identified as outcrop which is ice
Island - Annie Island DiKm-4 hornblende, gabbro, eroded Archean granite
diorite, periodite, and with patches of thin
serpentine. None of drift and scattered
them is the same age. boulders.
More than eighty Archean Early
percent of the rock can Precambrian but rock
be identified as rock Island - Picture Rock
is identified as outcrop which is ic~
DhKo-l hornblende, gabbro, eroded Archean grarute Island
diorite, periodite, and with patches of thin
serpentine. None of drift and scattered
them is the same age. boulders.
More than eighty Archean Early
percent of the rock can Precambrian but rock
be identified as rock Mainland - Penin sula -
is identified as outcrop which is ic~
DjKr-23 hornblende, gabbro, eroded Archean grarute Western Peninsula
diorite, periodite, and with patches of thin
serpentine. None of drift and scattered
them is the same age. boulders.
More than eighty Archean Early
percent of the rock can Precambrian but rock
be identified as rock Island - U nnamed is identified as
outcrop which is ic~ island but locally called DiKm-51 hornblende, gabbro,
eroded Archean grarute Horseshoe Island diorite, periodite, and
with patches of thin serpentine. None of
drift and scattered them is the same age.
boulders. More than eighty
Archean Early percent of the rock can
Precambrian but rock be identified as rock
Mainland - Peninsula-is identified as
outcrop which is ice DiKq-l0 hornblende, gabbro,
eroded Archean granite Western Peninsula diorite, periodite, and
with patches of thin serpentine. None of
drift and scattered them is the same age.
boulders.
Page 568
Archean Early More than eighty Precambrian, schistose percent of the rock can
volcanic with sorne be identified as rock
DhKm-19 small quantities of outcrop which is ice Mainland - Aulneau
sediments that are iron eroded Archean granite Peninsula formation. Y ounger with patches of thin sediments possibly drift and scattered
exist here. boulders.
Archean Early More than eighty
Precambrian but rock percent of the rock can
is identified as be identified as rock
DhKm-20 hornblende, gabbro, outcrop which is ice Mainland - Aulneau
diorite, periodite, and eroded Archean granite Peninsula
serpentine. None of with patches of thin drift and scattered
them is the same age. boulders.
Archean Early More than eighty
Precambrian but rock percent of the rock can
is identified as be identified as rock
DgKm-2 hornblende, gabbro, outcrop which is ice
Mainland diorite, periodite, and
eroded Archean granite
serpentine. None of with patches of thin drift and scattered
them is the same age. boulders.
Archean Early More than eighty
Precambrian but rock percent of the rock can
is identified as be identified as rock
Island - Unnamed DiKm-50 hornblende, gabbro,
outcrop which is ice island but locally called
diorite, periodite, and eroded Archean granite
Horseshoe Island serpentine. None of
with patches of thin drift and scattered
them is the same age. boulders.
Archean Early More than eighty
Precambrian but rock percent of the rock can
is identified as be identified as rock
DiKm-49 hornblende, gabbro, outcrop which is ice Unnamed island across
diorite, periodite, and eroded Archean granite from Tamarack Island
serpentine. None of with patches of thin drift and scattered
them is the same age. boulders.
Page 569
'î 1
) /
Appendix 9: Possible Correlations between the Type of Track. Reserves. Offerings in 2001 and before. and Type of Site.
This appenclix presents the data regarding the type of track (routes taken by boat inclicated on charts by the Department of Oceans
and Fisheries), whether the route may or may not exist nearby a site, the presence or absence of reserves nearby, whether offerings were
present in 2001 and before, and whether the site exists on a cliff or in a cave. The word 'near' implies that site is within 6 miles, or
5.21385744 nautical miles, or 9.6506404 kilometres by boat from the nearest reserve. The journey would only take up to 15 to 20 minute by
boat.
DgKl-1
Knownreported on May
29 th 1963 but first recorded by
Dewdney in 1964
5 minutes from South track
that runs through
Sabaskong Bay
Sabaskong Bay lndian Reserve 35
D
Yes - tobacco
Page 570
Pastershank in 1989 - tobacco
Cliff
') )
Yes = children's and adults clothes (trousers, skirts, t-shirts, socks, chenille sweater, flannel clothes - some old, some new), dried food stuff, bandages, earring (mother of pearl with porcupine quills and
silver and blue beads), earring - plastic heart (never in pairs), tobacco, empty tobacco container ("Copenhagen Tobacco"), beer
Known - first bottle glass, clear glass, "Prayer Sticks" (26 bundles of 12,12 cm
recorded by long by 12.5 - 17 cm wide, tied together with white cotton string,
Dewdney in Sabaskong half of each stick painted blue and the other half painted red, large
1964, then 5 minutes from
Bay Indian diameter sticks always tied together and smalt diameter sticks always
Pastershank in DgKl-2
Pastershank in South track in
Reserve 35 tied together - never mixed, 1 bundle of 12 in the left-hand cave,
1989 - tobacco Cliff
1989 but found Sabaskong Bay
D refer to field sketches and digital and manual photographs). More
in 2001 with new offerings were left on the rock bench under overhang when came
images back later to record more photographs, modern paintings also
visible at site. Returned August 5th 2001 with a team of divers who discovered the rocks immediately below and the vicinity around the rock image site. They discovered that it was a mud bottom and that
there was a pile of rocks but no offerings were evident only Mountain Dew soft drink can, a gentleman's bicycle, and another
cave to the right and 2.14 m (7 feet) below site under water. Known- by
Dewdney but he Yes - tobacco, money (coins) - on shelf/crevices. August 5/01 =
did not record it. It was first
(with divers) - Medicine sticks found under water in same bundles -
recorded by 5 minutes from Sabaskong 12 sticks tied together with string - red and blue just like site
DgKl-Pastershank in South track in
Bay Indian immediately opposite Nestor Falls - are a couple of feet below the None recorded Cliff
17 1989 but Sabaskong Bay
Reserve 35 water level but not immediately beside/below rock art site; broken
informed by Ron D plate but can't find other part - same maker's mark as Horseshoe
Kelly and Ron Island pot - Alfred Meaken; child's fishing rod; collection of shirts
Seymour, from about 6.1 m (20 feet) among large boulders below the water level
nearby reserve Known by
Dewdney but he 5 minutes from
Sabaskong DgKl- did not record.
South track in Bay Indian
No None recorded Cliff 19 Recorded by
Sabaskong Bay Reserve 35
Pastershank in D 1989
---
Page 571
') ! ')
Known - first 5 -10 minutes Sabaskong
DgKm-recorded by Reid
from South lndian No None recorded Cliff 1 track in Reserve
in 1975 Sabaskong Bay 45F
June 12/2001 - Yes = multiple articles of clothing (socks, shirts, jeans, jersey shirt, polyester sport shirt, all colours) - all ab ove water level , stuffed in crevices and cracks as well as numerous sample-size
(bottom blown into a mould, containing lead drop shot, neck of 1960 - clothing. Known - first bottle hand-finished - no seam at top, cork with hand-tooled top,
DhKm- recorded by main track
None has a smell when opening container - juniper smell, on bottom Molyneaux in
called East 1974 - clothing; Cliff 1 Oewdney in Track through
nearby 25%, wedged in groove in rock about 1.52 - 1.83 m (5-6 feet) from Fox 1974-1960
Whitefish Bay pot), lots of coins in the rock (George V quarter, American penny), bundles of
plate made by same company as bowl from new Horseshoe Island clothing;
site - Alfred Meaken, Royal lronstone China, England, Queen Victoria's crown, lion and unicorn). Huge pile of miscellaneous
clothing, different styles and different periods to the depth of 15.85 m (52ft) from the surface of the water. lncluded fabric and leather
clothes, moccasins, pile leaning against cliff face.
Yes - Stick, peeled - looks like beaver stick but ends are a bit weathered so hard to be 100% sure - 3/4" (1.88 cm) diameter and
10 minutes 1/5' (15.24 cm) long. Tobacco between 2 and 3. Two cowrie shells Known- first from main = one 7.5 cm - other 5 cm (hidden in sedge). One piece of red
DhKm- recorded by track called None clotho One cap (red) embroidered "Sail Caribbean" - deteriorating None recorded Cliff 3 Dewdney in East Track nearby cloth, very faded, not recently left (probably over 1 year old). Ali on
1959 through rock shelf. Soil under areas 2 and 3 has been disturbed recently (by a Whitefish Bay stick chewed by a beaver?) and between bushed at north-east edge,
also under large cowrie shell and one small area of moss 14.4cm (6 inches) from it.
Known -first 5 minutes from
main track DhKm- recorded by
called East None
No None recorded Cliff 4 Dewdney in Track through
nearby 1960
Whitefish Bay
Page 572
') ")
Known -fu:st 5 minutes from
Dewdneyin main track
DhKm- recorded by called East
None Yes - coins (coin) on 2 ledges, tobacco on 2 ledges, decayed fruit - 1960 found Cliff
5 Dewdney in Track through
nearby food prayer sticks and 1960
Whitefish Bay none later
Known -first 5 minutes from
Dewdneyin main track
DhKm- recorded by called East
None No
1960 found Cliff
18 Dewdneyin Track through
nearby prayer sticks and 1960
Whitefish Bay none later
Sabaskong
10 minutes Bay Indian
Dewdneyin Reserves
Known - first from main
35H & 32 1960 found
DhKn-recorded in 1960
track called C&the Yes - Right side = feather Oarge bird), partially smoked cigarette
overcoat and Cliff
1 South Track various by Dewdney
through lndian
weathered and Reserve on
Sabaskong Bay Obabikon
rotten "rags"
Bay
1 0 minutes of Y es = in the various cracks in the cliff face tobacco (5 piles), Dewdneyin 1960 found
Known- main track None
cigarettes, snuff can lids, 3 sample-size Crown Royal botdes, aIcohol chinaware and
DiKn-1 recorded in 1960 called East nearby
botdes, clothing, money (25 cents = 1977, 1989,1995; 1 cent = other odds and
Cliff by Dewdney Track through 1989, 10 cents = 1994). 2 Jack pine trees up top with gifts on them -
Whitefish Bay dress shirt, towel (on ground), cloth tied around tree. ends. Nothing
found later
5 minutes from
Known - first the alternative
DiKp-1 field drawing by track called the None
No No Cliff Lawson in 1885
east-west route nearby above Aulneau
Peninsula.
DkKn-6 Known in 1985 Not near main None
No No Cliff by Lambert track nearby
DkKn-7 Known in 1985 Not near main None
No No Cliff by Lambert track nearby
---- -
Page 573
\) ~ /
Known-Dewdneyin
recorded by Not near main None 1958 and
DjKn-l No Molyneaux in Cliff Dewdney in track nearby 1974 recorded 1958
offerings
Reported in 20 minutes from main
1946 but first track called None
Dewdney in DiKm-3 recorded by
East Track nearby No 1953 found a Cliff
Dewdney in through
"prayer raft" 1953
Whitefish Bay At least 20
Known- first minutes from Dewdneyin
DiKm-4 recorded by main track None Yes-2 silver coins on rocks at base of cliff that were under the 1960 found
Cliff Dewdneyin called East nearby surface of the water clothing, tobacco 1960 Track through and prayer sticks.
Whitefish Bay 10 minutes
from Lake of the alternative Woods
Known - drawn route through Indian
DhKo-by Howard T. area and 20 Reserve 35
1 Thompson on minutes from and Lake Yes - beer bottle glass under moss in vertical crevice None Cliff
1
the 9th South track in of the November 1936 Sabaskong Bay Woods
and 10 minutes Indian from Main Reserve 35]
track 10 minutes
Known-from main
Lake of the DjKr-
Dewdney first track called
Woods IR Yes = money, bullcts (33), tobacco - alliocatcd in crevices Dewdney found
Cliff 23 recorded in 1960
North-South 31B
offering in 1960 track through Big Narrows
Page 574
') ')
Known-first 5 minutes from
DiKm-recorded by main track
None 51
Dewdneyin called East nearby
No None Cliff 1966 or track through
afterwards ~'hitefish Bay 10 minutes
Known-from main
Lake of the 1 Dewdney found DiKq-
Dewdney first track called
Woods IR No clothing, towel
Cliff 10 North-South and tobacco in
recorded in 1960 track through
31B 1960
Big Narrows Sabaskong
Less 5 minutes Bay Indian
New - found in from main Reserve
DhKm- 2001 but track called and
19 paintings in South Track Sabaskong No Not applicable Cliff
modern paint through Bay Indian Reserves
Sabaskong Bay 35H and 32
C New - found in 5 minutes from
DhKm-2001 but with main track
None 20
date of 1982 but called East nearby
Yes - tobacco, something in water too deep to reach Not applicable Cliff paintings in Track through
modern paint %itefish Bay
New - found in 5 minutes from
Sabaskong main track
DgKm- 2001 but called South
Indian No Not applicable Cliff 2 paintings in Reserve
ochre Track through
45F Sabaskong Bay
New - found in 5 minutes from
DiKm- 2001 but main track
None 50 paintings in
called East nearby
No Not applicable Caye
ochre track through %itefish Bay
Page 575
")
20 minutes New - found in from main
DiKm- 2001 but track called None No Not applicable Caye
49 paintings in East Track nearby ochre through
Whitefish Bay
Page 576
"j )
Appendix 10: Indian reserves, offerings in 2001 and before, type of site, type oflandmass, and orientation of sites.
This appendix presents the data required ta discuss whether a correlation exists between the physical proximity of the pictograph
sites to the nearby reserves and the presence of offerings recorded either in the past or in 2001. It also enables the data regarding the type of
site, its land mass, and orientation to be examined and compared.
DgKl-l
DgKl-2 Sabaskong Bay lndian Reserve
35D
Yes - tobacco
Yes = children's and adults clothes (pants, skirts, t-shirts, socks, chenille sweater, flannel clothes - some old, some new), dried food stuff, bandages, earring (mother of pearl with porcupine quills and
silver and blue beads), earring - plastic heart (never in pairs), tobacco, empty tobacco container ("Copenhagen Tobacco"), beer bottle glass, clear glass, "Prayer Sticks" (26 bundles of 12, 12 cm long by 12.5 - 17 cm wide, tied together with white cotton string,
half of each stick painted blue and the other half painted red, large diameter sticks always tied together and small diameter sticks
always tied together - never mixed, 1 bundle of 12 in the left-hand cave, refer to field sketches and digital and manual photographs). More offerings were left on the rock bench under overhang when came back later to record more photographs, modern paintings
also visible at site. Returned August 5th 2001 with a team of divers who discovered the rocks immediately below and the vicinity around the rock image site. They discovered that it was a mud
bottom and that there was a pile of rocks but no offerings were evident only Mountain Dew soft drinks can, a gentleman's bicycle, and another cave to the right and 2.14m (7 feet) below site under
Yes - tobacco, money (coins) - on shelf/crevices. August 5/01 = (with divers) - Medicine sticks found under water in same bundles
Sabaskong Bay - 12 sticks tied together with string - red and blue just like site
Mainland -DgKI-Indian Reserve
immediately opposite Nestor Falls - are a couple of feet below the None recorded 1 - 109 degrees (E) Cliff Cyclone 17
35D water level but not immediately beside/below rock art site; broken
Point plate but can't find other part - same maker's mark as Horseshoe Island pot - Alfred Meaken; child's fishing rod; collection of shirts about 6.1 m (20 feet)among large boulders below the water level
4 - Angle 1 on left when facing cliff -continues to the
DgKl-Sabaskong Bay right to Angle 4 Mainland Indian Reserve No None recorded (at base). 1 - 90 Cliff - Cyclone 19
35D degrees (E), 2 - 93 Point degrees (E), 3 - 90
degrees (E), 4 -162 degrees (SSE)
DgKm-Sabaskong
Indian Reserve No None recorded 81 (E) Cliff Mainland 1 45F
June 12/01 - Yes = multiple articles of clothing (socks, shirts, jeans, jersey shirt, polyester sport shirt, all colours) - all above water level , stuffed in crevices and cracks as well as numerous
glass bottle (bottom blown into a mould, containing lead drop shot, neck of bottle hand-finished - no seam at top, cork with Dewdney in 1960 -
DhKm-hand-tooled top, has a smell when opening container - juniper clothing. Molyneaux Mainland -
1 None nearby smell, on bottom 25%, wedged in groove in rock about 1.52 - 1.83 in 1974 - clothing; 110 degrees (ESE) Cliff Aulneau
m (5-6 feet) from pot), lots of coins in the rock (George V Fox 1974 - bundles Peninsula quarter, American penny), plate made by same company as bowl of clothing;
from new Horseshoe Island site - Alfred Meaken, Royal Ironstone China, England, Queen Victoria's crown, lion and unicorn) (AC). Huge pile of miscellaneous clothing, different styles and different
periods to the depth of 15.85 m (52ft) from the surface of the water. Included fabric and leather clothes, moccasins, pile leaning
against cliff face.
Page 578
') ')
Yes - Stick, peeled -looks like beaver stick but ends are a bit weathered so hard to be 100% sure - 3/4"(1.88 cm) diameter and
1/5' (15.24 cm) long. Tobacco between 2 and 3. Two cowrie shells = one 7.5 cm - other 5 cm (hidden in sedge). One piece of
Island -DhKm-None nearby
red clotho One cap (red) embroidered "Sail Caribbean" -None recorded 150 degrees (SE) Cliff unnamed
3 deteriorating cloth, very faded, not recendy left (probably over 1 year old). Ali on rock shelf. Soil under areas 2 and 3 has been
island
disturbed recendy (by a stick chewed by a beaver stick?) and between bushed at north-east edge, also under large cowrie shell
and one small area of moss 6 inches from it. .
DhKm- 1 1 = 200 degrees Mainland -
4 None nearby No None recorded (SSW) , 2 = 182 Cliff Aulneau
degrees (S) Peninsula
DhKm- Yes - coins (coin) on 2ledges, tobacco on 21edges, decayed fruit-Dewdney in 1960 Mainland -
5 None nearby
food found prayer sticks 120 degrees (ESE) Cliff Aulneau
and none later Peninsula
DhKm-Dewdney in 1960 Mainland -
18 None nearby No found prayer sticks 230 degrees (SW) Cliff Aulneau
and none later Peninsula Sabaskong Bay
Indian Dewdney in 1960 Left side = 250 Reserves 35H Mainland -
DhKn-and 32 C and Yes - Right side = feather (large bird), partially smoked cigarette
found overcoat and degrees (WSW), Cliff Aulne au
1 the Indian
various weathered Right side = 240 Peninsula
Reserve on and rotten "rags" degrees (WS\X~
Obabikon Bay
Yes = in the various cracks in the cliff face tobacco (5 piles), Dewdney in 1960
right cliff - 119 Mainland -cigarettes, snuff can lids, 3 sample-size Crown Royal botdes,
found chinaware and (SE); le ft cliff - 1 Aulneau
DiKn-1 None nearby alcohol botdes, clothing, money (25 cents = 1977, 1989, 1995; 1 other odds and ends.
(left) = 132 (SE), 2 Cliff Peninsula cent = 1989, 10 cents = 1994).2 Jack pine trees up top with gifts
Nothing found later (right) = 164 - Hay ter
on them - dress shirt, towel (on ground), cloth tied around tree. (SSE) Peninsula
1 - 142 degrees (SE), 2 - 164
Island -DiKp-l None nearby No No
degrees (SSE), 3 -Cliff Cliff
150 degrees (SE), Island
4 - 140 degrees (SE)
Page 579
DkKn-6
DkKn-7
DjKn-1
DiKm-3
DiKm-4
DhKo-1
DjKr-23
DiKm-51
None nearby
None nearby
None nearby
None nearby
None nearby
Lake of the Woods Indian Reserve 35 and
Lake of the Woods Indian
Reserve 351
Lake of the Woods IR 31B
None nearby
No
No
No
No
Yes-2 silver coins on rocks at base of cliff that were under the surface of the water
Yes - beer bottle glass under m05S in vertical crevice
Yes = money, bullets (33), tobacco - alliocated in crevices
No
Page 580
)
No
No
Dewdney in 1958 and Molyneaux in
1974 recorded offeri~
Dewdney in 1953 found a "prayer
raft"
Dewdney in 1960 found clothing,
tobacco and prayer sticks.
None
Dewdney found offering in 1960
None
322 (NW) Cliff
222 degrees (SW) Cliff
1 = 140 (SE), 2 = . 142 (SE), 3 = (SE) 1 Cliff
WSW-225 degrees
90 degrees - East
Almost due north, 4 degrees
4 (1 being highest of the 4,4 being the lowest). 1 =
Yes - tobacco, something in water too deep to reach
No
No
No
Page 581
''j
Dewdney found clothing, towel and
tobacco in 1960
Not applicable
Not applicable
Not applicable
Not applicable
Not applicable
2 (base) 1 = 148 Mainland -Peninsula
(SE), 2 = 120 Cliff - Western
(ESE) Peninsula
Mainland -1 - 154(SE) Cliff Aulneau
Peninsula
1 (base of cliff Mainland -face) 54 degrees Cliff Aulneau
(NE) Peninsula
236 degrees (SW) Cliff Mainland
Island -Unnamed island but
214 (SSW) Cave locally called
Horseshoe Island
NNW, another Island -
Unnamed opening at back
island leads up to top of
Cave DiKm-3 opening
across from
up and facing Tamarack
WSW Island
'') / )
Appendix 11: Data regarding the type of access to the pictograph site, vegetation at base of site, water levels, and the number of
cliffs used.
This appendix presents the data regarding the type of access to images at each site, the vegetation at the base of each site, the water
level at each site in 2001, the number of cliff faces used, and the total number of images present. It is important to indicate the water level
because it should not be assumed it is the same at each site.
Boat and foot since it was possible to climb on the .
DgKI-2 k b h' f f Present water level None- but rock and water 13.73 m 9 Images roc enc ln ront 0
site
DgKl-17 Boat Present water level None - water 4.572 metres 1 13 images 1 1
DgKl-19 Boat Present water level None - water 6.096 metres 6 images
DgKm-1 Boat Present water level None - water 4 metres 7 images
)hKm-1 Boat Present water level None - water 7 mettes i 17 images 1 1bmnm
From ledge to top of cliff is ' applicable Boat and foot since an . b h 1 d 1 12.19 metres. Ledge has to be 1 because site
DhKm-3 individual had to climb Present water level Jumpedr us ,pa e dCory ka ys, reached by climbing steep cliff 33 images not k f se ge, moss an roc d f . 1 liff . d' 1 up roc ace. compose 0 vertJca c Imme late y
paintings) paintings) & below No 8 blobs 36 images
water) paintings)
DiKm-Yes (paintings Yes
Yes (above & 3
Cliff on the No (through No No 5 blobs 29 images below water)
overhang) paintings)
DiKm-Yes
Yes (adjacent Yes (ab ove & Cliff No No (adjacent No No blobs 4 images
4 paintings)
paintings) below water) --- ---- -
Page 586
') ')
DhKo- Yes (below Yes
Yes (through 3 blobs and 1 spiky blob with Cliff No (through No 29 images No
1 paintings) paintings)
paintings) seven lines sticking out
DjKr- Yes (below Yes
Yes (through 2 blobs and two small round
Yes (ab ove & Cliff No (through No blobs one immediately on top of 8 images
23 paintings) paintings)
paintings) the other
below water)
DiKm-Yes (above Yes Yes (ab ove &
Yes (ab ove & 51
Cliff No & below (through below No 1 blob 6 images below wa ter)
paintings) paintings) paintings)
DiKq-Yes
Cliff No No (through No No 7 blobs 11 images No 10
paintings) 1
DhKm-Cliff No
Yes (below No
Yes (above No No blobs 1 image No
19 paintings) paintings)
DhKm- Yes (ab ove Yes (ab ove
Yes (ab ove & 20
Cliff paintings)
No & through No No No blobs 4 images below water)
paintings)
Yes (paintings
DgKm-above the Yes -check Yes
Yes (ab ove 2
Cliff overhang and with field (through paintings)
No 4 blobs 6 images No on the notes paintings)
overhang) cave -
DiKm-Cave
cave - not cave - not cave - not cave - not not No blobs 1 image
cave - not 50 applicable applicable applicable applicable applica applicable
ble cave -
DiI<Jn-Cave
cave - not cave - not cave - not cave - not not No blobs 2 images
cave - not 49 applicable applicable applicable applicable applica applicable
ble
Page 587
.~ ')
Appendix 13: Sites grouped according to the number of physical attributes present.
This appendix demonstrates that only one site had aU Eve of the physical attributes argued by previous archaeologists as important
for the presence of paintings at pictograph sites on cliffs. The table below clearly indicates that the most sites have only four of the Eve
physical attributes considered prerequisite: deep holes, large crevices and cracks, rock overhangs, rock benches, and rock ledges. Rocks at
the base of cliffs and caves are not considered as one of the prerequisites since only Rusak (1992a) recorded and considered this features as
important.
DgKl-2 DgKl-19 DhKo-l DiKm-3
DhKm-3 DgKm-l DgKl-l
DhKm-5 DiKm-51 DhKm-4
DhKm-18 DiKn-l DhKm-20
DhKn-l DiKm-4 DhKm-19
DkKn-6 DkKn-7 DiKp-l
DgKl-17 DjKn-l DiKq-l0
DgKm-2 DjKr-23
DhKm-l
Page 588
') ')
Appendix 14: Sites grouped according to number of physical attributes present plus offerings.
Data in this appendix demonstrates any correlation between the presence of the physical attributes and the presence of offerings in
2001. The table indicates that the presence of the five physical attributes did indicate that offerings were present in 2001 (those sites that
have offerings are in italics). The cave sites DiKm-49 and DiKin-50 are not included in this table because not ail the physical prerequisites
are associated with caves. Furthermore neither site had offerings.
DgKl-2 DgKl-19 DhKo-1 DiKm-3
DhKm-3 DgKm-l DgKI-1
DhKm-5 DiKm-51 DhKm-4
DhKm-18 DiKn-1 DhKm-20
DhKn-1 DiKm-4 DhKm-19
DkKn-6 DkKn-7 DiKp-l
DgKl-17 DjKn-l DiKq-l0
DgKm-2 DjKr-23
DhKm-1
Page 589
'1 !
-)
Appendix 15: Sites grouped according to number of physical attributes present and with rocks at base of cliff face.
This appendix demonstrates that only one site had ail five of the physical attributes argued by previous archaeologists as important
for the presence of paintings at pictograph sites on cliffs, and that most sites have only four of the five physical attributes. The sites with
rocks at their base are in italics. Rocks do not exist in the cave site called DiKm-49 but they do occur at the base of DiKm-50 but neither
site is included in this table because this table is concerned with the sites on cliff faces.
DgKI-2 1
DgKI-19 DhKo-l DiKm-3 1
DhKm-3 DgKm-1 DgKJ-l
DhKm-5 DiKm-51 DhKm-4
DhKm-18 DiKn-1 DhKm-20
DhKn-1 DiKm-4 DhKm-19
DkKn-6 DkKn-7 DiKp-1
DgI<l-17 DjKn-1 DiKq-l0
DgKm-2 DjKr-23
DhKm-l
Page 590
') ') /
Appendix 16: The "presence and absence" tables and the dendrograms from the Pearson correlation and the Euclidean distance method to consider the re1ationship between the physical attributes of the sites, the different sites and the images, and the relationship between the sites and sorne of the largest shape types.
The information entered in the tables below was subsequently examined using SPSS to confirm the existence of clusters. Through
SPSS clusters were analysed and the relationship between the different sites and image types was investigated. The sequence of information
foilows that of the chapter. The relationship between different aspects of the sites, the relationship between the physical attributes of the
sites, and finaily the relationship between sites and some of the largest shape types identified in Chapter V.
a) The data presented below was used to create the foilowing dendrograms, and to indicate the weak relationship, perhaps differing
insignificantly from randomness, between these sites and the different features such as to the type of land mass, access to sites by boat,
proximity to lndian reserves, offerings left nearby the images in 2001, vegetation at base, proximity to site and number of cliffs at a site if
the site was on a cliff face.
In the "type of site" column, number 1 designates a cliff, while 0 means that the site is a cave. In the "land mass" column, '1' stands
for mainland and '0' indicates an island. The 'l'in the "access" column indicates a site accessible only by boat, and the '0' indicates that the
images could be reached by foot as weil as boat. The '1' under "proximity to lndian reserve" shows that an lndian reserve existed near the
site in 2001, while a '0' indicates no proximate reserve. As discussed in Chapter VI the word 'near' implies that site is within 6 miles, or
5.21385744 nautical miles, or 9.6506404 kilometres by boat from the nearest reserve. The journey would only take up to 15 to 20 minute by
boat. The 'l'in the "offerings le ft at sites in 2001" column indicates that offerings were found there in 2001, while '0' indicates that no
offerings were found. The 'l'in the "vegetation at base" column indicates that plant matter grew at the base of the site, while the '0'
indicates that vegetation was not present at the base of the site. The '1' under "proximity of main track to site" indicates that the site was
close by to an officiaily charted water route, and the '0' indicates that it was not. The 'l'in the "number of cliffs" column indicates images
occurring on one cliff face, while '0' shows that multiple cliff faces were used. A '-9' placed in this column shows paintings occurred inside a
cave rather than on a cliff face.
Page 591
') ')
-_.- -------
proximity offerings proximity
Borden latitude longitude
geographical type land acces to Indian left at sites
vegetation at of main numberof number range of site mass s Base track to c1iffs
Hierarchical Cluster Analysis - Dendrogram using Average Linkage (Between Groups) using Euclidean Distance Method Rescaled Distance Cluster Combine
CAS E 0 5 10 15 20 25 Label Num DiKq-10 22
DgKm-2 25
DgKl-19 4
DgKm-1 5 --' 1-
DgKl-17 3
3 DjKr-23 20
DgKl-1 1
DgKl-2 2
DhKm-19 23 1
1 DkKn-6 14
Tl-DkKn-7 15
DhKm-4 8
Di'Km-3 17
3 DiKm-51 21
DiKp-1 13
DhKm-1 6 J ~ DhKm-5 9
DiKm-4 18
DhKm-18 10 ===rt-r-DhKm-20 24
DhKm-3 7
DhKo-1 19
DiKn-1 12
DhKn-1 11
DjKn-1 16
DiKm-50 26
DiKm-49 27
Page 593
) ; î
/
Hierarchical Cluster Analysis - Dendrogram using Average Linkage (Between Groups) using Pearson Correlation Method Rescaled Distance Cluster Combine
CAS E o 5 10 15 20 25 Label Num DiKq-lO 22
DgKm-2 25
DgKl-19 4
DgKm-l 5
DhKm-19 23
DgKl-17 3
DjKr-23 20
DgKl-1 1
DgKl-2 2
DkKn-6 14
DkKn-7 15
DhKm-4 8
DhKm-18 10
DiKm-3 17
DiKm-51 21
DiKp-1 13
DhKm-1 6
DhKm-5 9
DiKn-1 12
DiKm-4 18
DhKo-1 19
DiKm-50 26
DiKm-49 27 J DhKn-1 11
DhKm-3 7
DhKm-20 24
DjKn-1 16
Page 594
î »
b) The table below indicates a weak relation between the site and the five required physical features considered necessary for the presence
of a pictograph site. The following dendrograms confirm this weak relationship.
A '-9' indicates cave sites without cliffs. The '1' indicates the presence of a feature at a site, while the '0' points to its absence. --- -- ---------- -- _ .. _- ----- -- ------ -- -- --_ ....... _-
Examining through hierarchical cluster analysis both of the dendrograms below (Euclidean distance and Pearson correlation), demonstrates
only a weak correlation between the sites and the 'stick figure.'
Page 614
')
Hierarchical Cluster Analysis - Dendrogram using Average Linkage (Between Groups) using Euclidean Distance Method Rescaled Distance Cluster Combine
CAS E o 5 10 15 20 25 Label Num +---------+---------+---------+---------+---------+
DiKm-4 17 -DhKo-l 21 -DhKm-3 6
DhKm-3 7 -DjKn-1 13
DgKm-1 4 -DjKn-1 14
DgKm-1 3 - -DhKo-1 18 - r---DhKo-1 19 -DhKm-1 5
DhKn-1 11 - 1--
DiKp-1 12 -DgKm-1 1
DhKm-18 10
DhKm-3 8 '-----
DiKm-3 15
DiKm-3 16 ~ DgKm-1 2
DhKm-18 9
DhKo-1 20
Page 615
') / ')
Hierarchical Cluster Analysis - Dendrogram using Average Linkage (Between Groups) using the Pearson Correlation Method Resca1ed Distance C1uster Combine
CAS E o 5 10 15 20 25 Label Num
DiKm-4 17 -
DhKo-l 21 -DhKm-3 6
DhKm-3 7 -DjKn-1 13
DhKo-1 18 -DhKo-1 19 -DhKm-1 5
DhKn-1 11 -DiKp-1 12 - - r---DgKm-1 1
DhKm-18 10
DhKm-3 8
DgKm-1 4 -DjKn-1 14
DgKm-1 3 - -DiKm-3 15
DiKm-3 16
DgKm-1 2
DhKm-18 9 1
DhKo-1 20
Page 616
'1 / ')
j) The 'smear':
The number '-9' indicates that exfoliation, white mineraI deposits, lichen or rock tripe encroachment made it impossible to establish
whether the shape was either hollow or solid.
Borden Latitude Longitude Geographical Range centre of shape red ochre number
(ii) Hierarchical Cluster Analysis - Dendrogram using Average Linkage (Between Groups) using Pearson Correlation Method Rescaled Distance Cluster Combine
CAS E 0 5 10 15 20 25 Label Num
DjKr-23 61
DgKm-2 71
DhKm-3 8
DkKn-6 41
DiKm-3 58
DhKm-5 23
DhKm-5 15
DhKm-3 12
DiKp-1 37
DkKn-7 45
DhKm-3 9
DiKn-1 32
DiKp-1 36
DiKq-10 69
DgKm-2 72
DhKm-3 11
DiKq-10 67
DiKq-10 68
DiKq-10 65
DiKq-10 66
DjKn-1 51
DiKq-10 64
DiKp-1 39
DiKp-1 40
Page 627
'j -)
DhKn-1 27
DiKp-1 38
DhKm-3 13
DgKm-2 70
DgKm-2 73
DhKm-3 10
DhKo-1 60
DiKq-10 63
DiKm-3 54
DhKo-1 59
DjKn-1 49
DjKn-1 52
DiKn-1 34
DjKn-1 46
DiKn-1 30
DiKn-1 33
DiKn-1 28
DiKn-1 35
DgKl-2 1
DhKm-3 7
DhKm-1 5
DhKm-3 6
DhKm-1 4
DhKm-18 25
DjKr-23 62
DgKm-1 3
DhKm-5 21
DhKm-5 22
Page 628
') ')
DhKm-5 19
DhKm-5 20
DhKm-5 17
DhKm-5 18
DhKm-5 16
DiKm-3 56
DiKm-3 57
DgKl-19 2
DiKm-3 53
DiKm-3 55
DjKn-1 48
DjKn-1 47
DkKn-6 44
DkKn-6 43
DkKn-6 42
DiKn-1 31
DiKn-1 29
DhKm-3 14
DjKn-1 50
DhKn-1 26
DhKm-5 24
Page 629
'î ')
m) Possible re1ationship between sites, various physical features, and groups which have large quantities of shapes.
This table contains the information required to establish a possible connection between a site, the five required physical features,
and the shapes which occur in large numbers: 'creatures,' 'zigzags,' 'crescents,' 'blobs,' 'smears,' 'parallellines,' birds,' 'turtles,' 'hand print,'
and 'three horizontallines.' An '-9' indicates cave sites without cliffs.
The following dendrograms were obtained through hierarchical cluster analysis of this data. The dendrograms below, (on the next few
pages) labelled (i) and (ii), confirm a very weak correlation and perhaps differ insignificantly from randomness between the site, various
physical attributes at each site, and the shapes that occur in large numbers: 'creatures,' 'zigzags,' 'crescents,' 'blobs,' 'smears,' 'parallellines,'
birds,' 'turtles,' 'hand print,' and 'group of three horizontallines.'
Page 632
.'j ) ')
Hierarchical Cluster Analysis - Dendrogram using Average Linkage (Between Groups) using Euclidean Distance Method
CAS E Label Num
6
25
19
5
4
9
2
11
12
21
15
20
10
14
3
13
18
22
24
8
16
23
17
1
7
o 5 10
J
J
.. -
Rescaled Distance Cluster Combine 15 20 25
1 -
1
1
~ l-
I--
-
1-.. -
Page 633
'\ 'î
Hierarchical Cluster Analysis - Dendrogram using Average Linkage (Between Groups) using Pearson Correlation Method Resca1ed Distance Cluster Combine
CAS E o 5 10 15 20 25 Label Num DiKq-10 22
DgKm-2 25
DgKl-19 4
DgKm-1 5
DhKm-19 23
DgKl-17 3
DjKr-23 20
DgKl-1 1
DgKl-2 2
DkKn-6 14
DkKn-7 15
DhKm-4 8
DhKm-18 10
DiKm-3 17
DiKm-51 21
DiKp-1 13
DhKm-1 6
DhKm-5 9
DiKn-1 12
DiKm-4 18
DhKo-1 19
DiKm-50 26
DiKm-49 27 J DhKn-1 11
DhKm-3 7
DhKm-20 24
DjKn-1 16
Page 634
'\ ;
'~ /
Appendix 17: The "Presence and Absence" Tables for the Images of the Birch Bark Scrolls.
The smail number of results for each of the largest types of shape precludes any statistical analysis from being conducted using
SPSS. For each of these tables a "1" indicated the presence of a feature while a "0" indicated its absence. This is because there is not a
sufficient quantity of data on which to perform cluster analysis. However, the table below indicates that the 'creature' which occurred only
on birch bark scrail number 1, were not exactly the same.
The table below for indicates that the 'bird' shapes on birch bark scrails were not exactly the same and that they occurred only on birch
bark scrail number 1.
Page 635
) )
The table below presents the information on the shape called 'two parallel verticallines', which only occurs on birch bark scroll1.
Page 636
~ .. !
Cited Bibliographie References:
Abercrombie, M., C. J. Hickman, and M. L. Johnson. 1985 The Penguin Dictionary if Biology. Seventh Edition. Penguin Books Ltd.: Harmondsworth, England.
Anonymous 1990a. On Record. A Long, Hard Journey Shown at Annie Island. Ontario Rock Art Conservation Association Newsletter Autumn: 10.
1990b On Record. The Grand Medicine Society at Picture Rock Island. Ontario Rock Art Conservation Association Newsletter Spring: 3.
Anfinson, S. F. 1990 Archaeological regions in Minnesota and the Woodland Period. In: The Woodland Tradition in the Western Great Lakes: Papers presented to the Eldon Johnson, edited by G. E. Gibbon, pp. 135-166. Publications in Anthropology # 4. University of :Minnesota, Minneapolis.
Anfinson, S.F. (Ed.) 1979 A Handbook if Minnesota Prehistoric Ceramics Occasional Papers in Minnesota Anthropology No 5, Minnesota Historical Society, Fort Snelling, Minnesota.
Angel, M. 2002 Preserving the Sacred: Historical Perspectives on the Ojibwa Midewiwin. University of Manitoba Press, Winnipeg.
Arsenault, D. 1996 A Unique Pictograph site in the Context of Political and Ideological conflicts. In: Papers if the Twenty-Seventh Algonquian Conference edited by D. Pentland, pp. 1-10. University of Manitoba, Winnipeg.
1997 L'Impact du Tourisme sur les Sites Sacré en Amérique Précolombienne. Téoros Été: 21-25.
1998 Esquisse du paysage sacré algonquien: Une étude contextuelle des sites rupestres du Bouclier canadien. Recherches Amérindiennes au Québec XXVIII (2):19-39.
Arsenault, D. ,L. Gagnon, C. A. Martijn, and A. Watchman 1995 Le projet Nisula: recherche pluridisciplinaire autour d'un site a pictogrammes (DeEh-1) en Haute Côte Nord. Paléo-Québec 23:17-57.
Arsenault, D. and L. Gagnon 1998 Pour une approche sémiologique et contextuelle en archéologie rupestre du Bouclier Canadien. In: Essais archéologiques et ethnohistoires en hommage à Charles A. Martijn, edited by R. Tremblay L'Éclairer et l'Ambassadeur. Recherches amérindiennes au Québec. Paléo-Québec 27: 213-241.
Page 637
Arthurs, D. 1978 Sandy Lake Ware in Northwestern Ontario: a Distributional Study. Manitoba Archaeological Quarter!J 2 (1 and 2): 57-64.
1979a The Renshaw Site, DaJI-I, An Archaic Site on the Western Lake Supetior Shore. On file with the Historie Planning and Research Branch. Ontario Ministry of Culture and Communication, Toronto.
1979b The Duncan Site Rescue Excavation, 1979. Preliminary Report. On file with the Historie Planning and Research Branch. Ontario Ministry of Culture and Communication, Toronto.
1979c Iron Range Lake Archaeological Survry-1979. On file with the Historie Planning and Research Branch. Ontario Ministry of Culture and Communication, Toronto.
1984 An Archaeological survey of the north-western shore of Lake Superior 1977. Conservation Archaeology Report, North Central Region Report # 16. Ontario Ministry of Citizenship and Culture, Thunder Bay.
1986 Archaeological Investigations at the Long Sault site (Manitou Mounds) Conservation Archaeology Report, North-western Region Report, # 7, Ontario Ministry of Citizenship and Culture, Thunder Bay.
Ashdown, J. H. 1973 A Contemporary Explorers Chart to the Lake rf the Woods. Royal Yacht Club of the Lake of the Woods, Kenora.
BGS 1101, 1986 Brock University Geological Society, Radiocarbon Laboratory, Department of Earth Sciences, St. Catharines, Ontario.
Bahn, P. G., R. G. Bednarik and J. Steinbring 1996 Conservation: A Canadian Tragedy. International Newsletter on Rock Art 13: 14-17.
Bahr, D. 1991 Interpreting Sacramental Systems: The Midewiwin and The Wi:Gita. Wicazo Sa Review 7 (2) : 18-25.
Balikci, A .. 1956 Note sur le Midewiwin. Anthropologica. 2:165-217.
Barbeau, C. M. 1930 The Origins of Floral and Other Designs among the Canadian and Neighbouring Indians. Proceedings rfthe International Congress rf Ameticanists 23: 512
Barnett, P.J., A. P. Henry, and D. Babuin Quaternary geology of Ontario, west central sheet/ Map 224 [map]. 1: 1 000,000. Sudbury, Ontario: Ontario Geological Survey, 1991.
Page 638
. /
Bates, R. L. and]. A. Jackson
Bell, R.
1984 Dictionary of Geological Terms. 3rcl Edition. Anchor Press/Doubleday, Garden City, New York.
1881 Report on Hudson's Bay and some of the Lakes and Rivers Lying to the West of it. In: Report of Progress for 1879-80. Geologieal and Natural History Survey of Canada. Montreal: Dawson Brothers
Bhar, D. 1991 Interpreting Sacramental Systems: The Midewiwin and The Wi:gita. Wicazo Sa Review 7(2): 18-25.
Bianchi, N. 0 and F. Rothhammer 1995 Reply to Torroni and Wallace. American Journal ofHuman Genetics 56: 1236-1238 .
Binford, L. R. 1968 Methodologieal considerations of the archaeological use of ethnographic data. In Man the Hunte1) edited by R. B. Lee and 1. DeVore, pp. 268-273. Aldine, Chicago.
1977 For Theory Building in Archaeology. Academic Press Inc., New York.
1981 Bones: Ancient Men and Modern Myths. Studies in Archaeology. Academic Press Inc., New York.
Bird,]. D. 1973 The Wisconsin deglaciation of Canada. Introduction and summary. Arctic and Alpine Research 5: 165-168.
Bishop, C. A.. 1976 The emergence of the Northern Ojibwa: social and economic consequences: American Ethnologist 2:39-54.
1983 The Western James Bay Cree: Aboriginal and Early Historic Adaptations: Prairie Forum 8(1): 47-55.
Bishop, C. A.. and M. E. Smith 1975 Early historie populations in northwestern Ontario: archaeologieal and ethnohistorical interpretations. American Antiquiry 40:54-63.
Bishop, C. A.. and A. J. Ray 1976 Ethnohistoric Research in the Central Subarctic: Some conceptual and methodologieal problems. The Western Canadian Journal of Anthropology 6(1):116-144.
Blackmore, S. and E. Tootill. 1984 The Penguin Dictionary of Botaf!J. Penguin Books: London.
Blessing, F. K.
Page 639
~ ..
1963 Birchbark Mide Scrolls from Minnesota. Minnesota Archaeologist 29:91-143.
1977 The Ojibway Indians Observed: Papers of Fred K. Blessing, Jr., On The Ojibway Indians From The Minnesota Archaeologist. Occasional Publications in Minnesota Anthropology, No. 1. publisher
Braaten, A. W. and S. P. Ellingson 1992 Instructional Resources: Art from the American Indians of the Plains and the Woodlands Region. Arl Education 45(6):25-33.
Brand, R. 1979 Petroglyph Casting. A Method of Collecting and Recording Rock Art. Canadian Rock Arl Research Association Newsletter (5):24.
Brasser, T. J. 1976 BoJou, Neijee! Prqftles of Canadian Indian Arl. Ottawa: National Museums of Canada.
1999 Notes on a recently discovered Indian Shirt from New France. American Indian Arl Magaifne 25(2)46-55.
Brenner, B. 1994 Interpretations of Rock Art. Manitoba Archaeological Journal. 4(1 and 2):43-76.
Brown, J. S. H. and R. Brightman 1988 'The Orders of the Dreamed": Geor;ge Nelson on the Cree and the NOrlhern Ojibwa Religion and My th, 1823. University of Manitoba Press, Winnipeg.
Brose, D. S. 1970 The Archaeology of Summer Island: Changing settlement patterns in northern Lake Michigan. Museum of Anthropology, U niversiry of Michigan, Anthropological P apers 41.
1978 Late Prehistory of the Upper Great Lakes Area. In: Handbook of the Norlh American Indians, edited by B. G. Trigger, pp. 569-582. Volume 15, Smithsonian Institution Press, Washington D. C.
Bryson, R. A. 1966 Air Masses, Streamlines and the Boreal Forest. Geographical Bulletin 8(3):228-269.
Bryson, R. A. and W. Wendland 1967 Tentative climatic patterns for sorne late glacial and post-glacial episodes in central North America. In: Lift. Land and Water, edited by W. ]. Mayer-Oakes, pp. 271-298. University of Manitoba Press, Winnipeg.
Bryson, R. A., W Wendland, M. Ives,]. D. and]. T. Andrews 1969 Radiocarbon ischrones in the disintegration of the Laurentide lce Sheet. Arctic and Alpine Research 1: 1-14.
Page 640
Blocker, H. G. 1994 The Aesthectics if Primitive Art. University Press of America, Lanham, London and New York.
Bolnick, D. and D. G. Smith 2003 Unexpected patterns of Mitochrondrial DNA Variation among Native Americans from the Southeastern United States. American Journal if Pf?ysical Anthropology 122:336-354.
Boone, E. 1994 Introduction: Writing and recording knowledge. In: Writing Without Words: Alternative Literacies in Mesoamerica and the Andes, edited by E. Hill Boone and W. D. Mignolo, pp. 3-26. Duke University Press, Durham and London.
Boone, E.H. and W. D. Mignolo 1994 W riting without words: Alternative Literacies in Mesoamerica and the Andes. Duke University Press, Durham and London.
Buisson, D., M. Menu, G. Pincon, and Ph. Walter 1989 Les objets colorés du Paléolithique superior cas de la grotte de La Vache (Ariège). Bulletin de la Société Pré-Historique Françaises. 86(6):184-191.
Bushnell, D. I. 1927 Burials of Algonquian, Siouan, and Caddoan tribes of the west of the Mississippi. Bureau if American Ethnology. Bulletin 83.
Cadzow, D. A. 1926 Bark Records of the Bungi Midewin Society. Indian Notes and Monographs 3: 123-134.
Canadian Hydrographie Services. "Sabaskong Bay, Lake of the Woods, Southeastern Channel" [hydrographie ehart]. 1:40,000. Sheet 6214. Ottawa: Department of Oceans and Fisheries, 1989.
Cameron,]. 1982 Petrogfyphs if the Lake if the Woods: Recording Techniques and Results. Unpublished paper subrnitted to the Ontario Heritage Foundation in Fulfilment of the requirements for Licence 82-50.
Cameron, A. and]. Cameron. 1979 Recent Adventures in Rock Art. Newsletter if the Canadian Rock Art Research Associates. Manitoba Chapter 2(1):4-8.
Carroll, F. M. 2001 A Good and Wise Measure: The Search Jor the Canadian-American Boundary, 1784-1842. University of Toronto Press, Toronto, Buffalo and London
Callahan, K. L.
Page 641
1998 Shamanism, Dream Symbo/ism, and A/tered States in Minnesota Rock Art [www document] http://www.geocities.com/Athens/Orade/2S96/mnra1.html Viewed 17th Nov, 17, 1998.
Chisholm, F. 1982 Polyfilm Mapping of the Main Petroglyph Panel, Mud Portage, Channel Site, Ontario. News/etter of the Canadian Rock Art Research Associates, Manitoba Chapter. 5 (1 and 2): 37-46.
Chilton, E. S. 1998 The Cultural Origins of Technical Choice: Unravelling Aigonquian and lroqiuan Ceramic Traditions in the Northeast. In: The Archaeology of Social Boundaries, edited by M. T. Stark, pp. 132-160. Smithsonian Institution Press, Washington and London.
Clark, G. A. 1999 Highly Visible, Curiously Intangible. Science New Series 283 (5410): 2029-2032.
Clelland, C. E. 1982 The Inland Shore Fishery of the Northern Great Lakes: Its Development and Importance in Prehistory. American Antiquzjy 47: 761-784.
Closs, M. P. 1986 Tallies and the Ritual Use of Number in Ojibway Pictography. In Native American Mathematics, edited by M. P. Closs, pp. 181-211. University of Texas Press, Austin.
Clottes,]. 1993 Post-stylistic? In: Rock Art Studies: The Post-Srylistic Era or Where do we go lrom here?, edited by M. Lorblanchet and P. G. Bahn, pp. 19-25. Papers presented in Symposium A of the 2nd AURA Congress, Cairns 1992. Oxbow Monograph 35, Oxford.
Clottes, ]., M. Menu, and Ph. Walter 1990a La préparation des peintures magdaléniennes des caverns aréigeoisis. Bulletin de la Société Pré-Historique Françaises. 71 :170-192.
1990b New Light on the Niaux Paintings. Rock Art Research. 7(1): 21-26.
Coleman, Sister Bernard, OSB 1929 Religion and Magic Among Cass Lake Ojibwa. Primitive Man 2: 52-55.
1937 The Religion of the Ojibway of Northern Minnesota. Primitive Man. 5: 1-25.
1947 Decorative Designs of the Ojibwa of Northem Minnesota: Catholic University of America Press, Washington D. C.
Collingwood, R. G. 1939 An Autobiograpl?J. Clarendon Press, Oxford.
Page 642
Conkey,M. 1987 New Approaches in the Se arch for Meaning? A review of the research in Palaeolithic 'art'. Journal if Field Archaeology 14: 413-430.
Conway, T. n.d. a Constellations Depicted in Ancient Ojibwa Rock Art. Unpublished Paper.
n.d. b Star charts, Art and Artifacts: A New Approach to Rock Art in Northern Ontario. Paper presented at the 1978 Symposium of the Ontario Archaeological Society.
n.d.c An Ancient Rock Art Site at Diamond Lake, Temagami District, Ontario. Unpublished paper.
1975 An Archaeological Survey of the North-eastern Shore of the Lake Superior. In: Collected Papers-Canadian Archaeological Papers, pp.17 -28. Historical Sites Branch, Division of Parks, Research Report #6.
1979 Rock Art and Conservation in Northeastern Ontario. Canadian Rock Art Research Association 1977. Papers from the Fourth Biennial Conference of the Canadian Rock Art Research Associates, Heritage Record 8: 127-145.
1984 An Archaeological Assessment if Obabika Lake. Manuscript. First Draft. Unpublished paper for Ontario Ministry of Citizenship and Culture, Sault Ste. Marie.
Conway, T. and J. Conway 1989 An Ethno-Archaeological Study of Algonkin Rock Art ln Northeastern Ontario, Canada. Ontario Archaeology 49: 43-59.
1990a An Ethno-Archaoelogical Study of Algonkin Rock Art ln Northeastern Ontario, Canada. Ontario Archaeology 49: 43-59.
1990b Spirits on Stone: The Agawa Pictographs Heritage Discoveries Inc., Echo Bay.
Cooper, J. M. 1936 Notes on the Ethnology of the Otchipwe of the Lake of the Woods and of Rainy Lake. Catholic University if America, Anthropology Series. 3:1-28.
Copway, G. 1851 The traditiona! history and charactenstic sketches if the Ojibwqy nation [microform] B. B. Mussey, Boston.
Courant, P., S. Fraser, M. Goodchild, M. Hedstrom, C. Henry, P. B. Kaufman, J. McGann, R. Rosenzweig, J. Unsworth, and B. Zuckerman.
2005 The Draft Report of the American Council of Learned Societies' Commission on Cyberinfrastucture for Humanities and Social Sciences [pdf document] http://www.acls.org/cyberinfrastructure/cyber report.htm Viewed Jan, 1 2006.
Page 643
. /
Couraud, C. 1983 Pour une etude methodoloque des colorants préhistoriques. Bulletin de la Société Pré-Historique Françaises 80 (4): 104-110.
Dailey, R. C. 1958 The Midewiwin, Ontario's First Medical Society. Ontario History 50 (3):133-138.
Davidson, D. S. 1941 Decorative Art of the Tetes de Boule of Quebec. lndian Notes and Monographs. 10 (9): 115-149 .
Davies, J. C. Bag Bqy; Ontario Map 4222 [map]. Scale 1 Inch to V2 mile, 1: 31 680. Geological Survey Precambrian Geology Series. Ottawa: Geological Survey of Canada, 1982.
Western Peninsula; Ontario Map 2443 [map]. 1 Inch to 1/2 mile, 1: 31 680. Geological Survey Precambrian Geology Series. Ottawa: Geological Survey of Canada, 1983.
Davis, W. 1992 Masking the Blow: the Seenes 0/ Representation zn Late Prehistorie Egyptian Art. University of California Press, Berkeley.
Dawson, K. C. A. 1973 Archaeological Report Kaministikwia River 1973. Unpublished. Lakehead University, Thunder Bay, Ontario.
1974 The McClusky Site National Museum of Man Mercury Series Paper No # 25, Archaeological Survey of Canada, Ottawa.
1975 Historie Populations of Northwestern Ontario. In Papers o/the SeventhAlgonquian Conference, edited byW. Cowan, pp.157-174. Carleton University, Ottawa.
1976a Albany River SU17Jf!YJ Patricia District, Ontario. Paper No # 51, Mercury Series, National Museum of Man, Archaeological Survey of Canada, Ottawa, Canada.
1976b Algonkians 0/ Lake Nipigon: An Archaeological SU17Jf!Y Paper No # 48, Mercury Series, National Museum of Man, Archaeological Survey of Canada, Ottawa, Canada.
1977 Northwestern Ontario Historie Populations. Man in the Northeast. 13: 67-77.
1981 The Wabinosh River Site and the Laurel tradition in northwestern Ontario. Ontario Archaeology 36:3.
Page 644
~-1
1983 Prehistory of the Interior Forest of Northern Ontario. In Boreal Forest Adaptations: the Nor/hem Aigonkians, edited by A. T. Steegman, pp. 55-84. Plenum Press, New York.
1987 Northwestern Ontario and the Early Contact Period: the Northern Ojibwa from 1615-1715. Canadian Journal if Anthropology 17: 143-180.
DeMallie, R. J. and D. R. Miller 2001 Assiniboine. In: Handbook if North American Indians, edited by R.]. DeMallie, pp. 572-595. Volume 13, Part 1. Plains. Smithsonian Institution, Washington D. C.
Denny,]. P. 1989 Algonquian connections to Salishan and northeastern archaeology. In Papers if the Twentieth Aigonquian C01iference, edited by W. Cowan, pp. 86-107. Carleton University, Ottawa.
1991 The Algonquian migration to Salishan and northeastern archaeology. In Papers if the Twenry-Second Aigonquian Conference, edited by W. Cowan, pp. 103-124. Carleton University, Ottawa.
Densmore, F. 1910 Chippewa Music I. Washington Government Printing Office. Smithsonian Institution, Bureau of American Ethnology. Bulletin 45.
1913 Chippewa Music II. Bulletin 53 of the Bureau of American Ethnology. Government Printer, Washington.
1941 The Native Art of the Chippewa. American Anthropologist. New Series 43(4)1:478-681.
1974 [1928] How Indians Use Wiid Plants for Food, Medicine and Crafts. New York: Dover Publications. Originally published as Uses of Plants by the Chippewa Indians (pp.275-397) in the Forty-fourth Annual Report of the Bureau of American Ethnology to the Secretary of the Smithsonian Institution, 1926-1927 by the United States Government Printing Office, Washington.
1979 [1929] Chippewa Customs. With an introduction by Nina Marchetti Archabal. Originally published by the Smithsonian Institution Bureau of American Ethnology as Bulletin No. 86 in 1929. Minnesota Historical Society Press, St. Paul.
Dewdney, S. 1958 The Quetico Pictographs. The Beaver Summer: 15-22.
1963 Indian Rock Art. Popular Series #4. Saskatchewan Museum of Natural History, Regina.
1965 Stone Age Paintings. A Heritage Area Publication. Province of Manitoba, Department of Mines and Natural Resources Parks Branch.
Page 645
1970a. Dating Rock Art in the Canadian Shield. Royal Ontario Museum, Art and Archaeology Occasional Paper 24. University of Toronto Press, Toronto.
1970b Pictographic Pursuits of Kitchigami's Shore. Archaeological Newsletter, Rqyal Ontario Museum. New Series, Number 59.
1970c Ecological Notes on the Objiway shaman-artist. Artscanada. August: 17-28.
1971 Insights on Vision Sites: A Matter of Relevance. Archaeological Newsletter, Rqyal Ontatio Museum, New Series, No. 69.
1972 The Year of the Pictographic Puns. Archaeological Newsletter, Rqyal Ontario Museum, New Series, No. 82.
1975a Walls and Waves: A Dating Dilemma. Archaeological Newsletter, Rqyal Ontario Museum, New Series, No. 121.
1975b The Sacred Scrolls if the Southem Ojibwcry. University of Toronto Press, Toronto and Buffalo.
1977. Search for Images of Forgotten Dreams. In Relevés et travaux récents sur l'art rupestre amérindien, edite pars G. Tassé et S. Dewdney. Laboratoire d'archéologie de l'Université du Québec à Montréal, PaléoQuébec 8: 5-29.
1979a. Rock Art Research in Canada-The First Hundred Years. Papers from the Fourth Biennial Conference of the Canadian Rock Art Research Associates, Heritage Record 8: 1-4.
1979b Verbal Versus Visual Approaches to Rock Art Research. Papers from the Fourth Biennial Conference of the Canadian Rock Art Research Associates, Heritage Record 8: 325-339.
Dewdney S. and K. Kidd 1962 lndian Paintings rfthe Great Lakes. University of Toronto Press, Toronto.
1967 lndian Paintings rf the Great Lakes. 2nd Edition. University of Toronto Press, Toronto.
1975 lndian Paintings if the Great Lakes. 3nd Edition. University of Toronto Press Toronto.
Dickason, O. P. 1972 lndian Arts in Canada. Department of Indian Affairs and Development, Ottawa.
Dieder, M. & 1. Herbich. 1998 Habitus, Techniques, Style: An Integrated Approaches to the Social Understanding of Material Culture and Boundaries. In The Archaeology if Social
Page 646
. ~ ..
Boundaries, edited by M. T. Stark, pp. 232-263. Smithsonian Institution Press, Washington and London.
Dondis, A. D. 1973 A Primer if Visual Literary. The MIT Press, Cambridge, Mass. and London, England.
Duncan, K. C. 1991 So many bags, so little known: Reconstructing patterns of Evolution and Distribution of Two Algonquian Bag Forms. Arctic Anthropology 28 (1): 56-66.
Dunning, R.W. 1959 Social and Economic Change Among the Northern Ojibwa. University of Toronto Press, Toronto .
Elson,J. A. 1965 Soils of the Lake Agassiz region. In S oils in Canada; Geologica~ Pedological and Engineering S tudies, edited by R. F. Legget, pp. 51-79. Royal Society of Canada. Special Publications No. 3, .University of Toronto Press, Toronto.
Eshleman, J. A, R. R. Malhi, J. J. Johnson, F. A Kaestle, J. A Lorenz, and D.J. Smith 2004 Mitochrondrial DNA and prehistoric settlements: Native Migrants on the Western Edge of North America. Human Biology 15: 55-75.
Evans, C. E. 1961 Prehistoric Blackduck-Historic Assiniboine: a Reassessment. Plains Anthropologist 6: 271-275.
Faulkner, G. F., N. T. Prince, and J. S. Neptune 1998 Beautifully Beaded Northeastern Native American Beadwork. American lndian Art Magaifne 24 (1): 32-45.
Fiddler, Chief T. and J. R. Stevens 1995 Killing the Shaman. 6th Printing. Penumbra Press, Ottawa.
Fieber, F. 1978 Birchbark Biting. Beaver 308 (4): 56-57.
Fiedel, S. J. 1987 Algonquian Origins. A problem in archaeological linguistic correlation. Archaeology if Eastern North America 15: 1-11.
1990 Middle Woodland Algonquian expanslOn: a refined model. North American Archaeologist 11: 209-230.
Frank, 1. 2000 The Theory if Decorative Art: An Anthology if European and American Wnftngs 1750-1940. Yale University Press, New Haven and London.
Page 647
Friedl, E. 1944 A Note on Birchbark Transparencies. American Anthropologist. New Series. 46 (1) Part 1: 149-150.
Friend, D. 1983 The EIder Site: A Preliminary Report Canadian Rock Art Research Associates Newsletter 5 (1 & 2): 1-36.
Fox, W. A. 1974 FallArchaeological Survry North Central and Northwestern Regions. Historical Planning and Research Branch. North Central Region. Manuscript on file. Ministry of Culture, Thunder Bay.
Flannery, R. 1931 A Stu4J if the Distribution and Deveiopment if the Memegwecio Concept in Algonquian Folklore. An Essay. Unpublished. Submitted to the Faculty of the Graduate School of Arts and Sciences, of the Catholic University of America, in partial fulfillment of the requirements for the degree of the Master of Arts.
Fulford, G. 1988 Ojibwa Picture Writing: A Stu4J on the Process if Symbol Information. MA. Thesis, University of Western Ontario.
1989 A Structural Analysis of Mide Song Scroils. In Actes du vingtième congrés des algonquinistes, edited by W. Cowan, pp. 132-153. Carleton University, Ottawa ..
1990 A Structural Analysis of Mide Chants. In Papers if the Twenry-First Aigonquian Conference, edited by W. Cowan, pp. 126-158. Carleton University, Ottawa.
1992 The Pictographie Account Book of an Ojibwa Fur Trader. In Papers if the Twenry-Third Algonquian Conference, edited by W. Cowan, pp.190-233. Carleton University, Ottawa.
Gardner, J. S. 1981 General Environment. In Handbook if the North American Indians, edited by J. Helm, pp. 5-14. Volume 6, Smithsonian Institution, Washington D. C.
Garte, E.J. 1985 Living Tradition in Ojibwa Beadwork and Quillwork. In Papers if the Sixteenth Algonquian Conference, edited by W. Cowan, pp. 9-24. Carleton University, Ottawa.
Gibbon, G. E. and C. A. H. Caine 1980 The Middle to Late Woodland Transition in Eastern Minnesota. Midcontinental Journal if Archaeolo!!J 5 (1): 57-72.
Gibbons, G.
Page 648
1916 Address by Sir George Gibbons, K. C. before the Canadian Bar Appropriation. Toronto, 1916, In: Papers relating to the Work if the International Joint Commission. Ottawa. pp. 7-17. International]oint Commission: Washington, D.C., and Ottawa, Canada.
Goddard,1. 1975 Algonquian, Wiyot, and Yurok: proving a distant genetic relationship. In Linguistics and AnthropoloJ!)', edited by M. D Kinkade, K. L. Hale, and O. Werner, pp. 249-262. Peter de Ridder Press, Lisse.
1978 Central Algonquian Languages. In Handbook if the North American Indians, edited by B. G. Trigger, pp.: 583-587. Volume 15. Smithsonian Institution, Washington D. C.
1979 Comparative Algonquian. In The Languages 0/ Native America: Historical and Comparative Assessment, edited by L. Campbell and M. Muthun, pp. 70-132. University of Texas Press, Austin, Texas.
1994 The West-to-East cline in Algonquian dialectology. In Papers of the Twenry-Fijth Aigonquian Conference, edited by W. Cowan, pp. 187-211. Carleton University, Ottawa.
Gombrich, E. H. 1984 The S tory 0/ Art. 14,h Edition. Phaidon, London.
Goodby, R. G. 1998 Technological Patterning and Social Boundaries: Ceramic Variablility in Southern New England, A. D. 1000-1675. In: The ArchaeoloJ!)' if Social Boundaries, edited by M. T. Stark, pp. 161-182. Smithsonian Institution Press, Washington and London.
Goods, B. M. 1990 Pigment Analysis of the Rice River Pictographs, Manitoba. The Rock Art Quarterfy 1 (3-4): 113-115.
Grant, C. 1967 Rock Art if the American Indian. Thomas Y. Crowell Company, New York.
Granzberg, G. and J. Steinbring 1995 The Line, Tree, and Grele in Rock Art and Pictography: Sorne Selected Examples from the Southwest. In Rock Art in the Americas, edited by ]. Steinbring, pp. 43-54. Papers presented to Symposium B of the First AURA Congress, Darwin 1988, Oxbow Monograph 45, Oxbow Books, Oxford.
Greenburg, A. M. and]. Morrison 1982 Group Identities in the Boreal Forest: the Origin of the Northern Ojibwa. Ethnohistory 29 (20): 75-102.
Greene, C. S.
Page 649
2001 Art Until 1900, edited by R. ]. DeMallie, pp. 1039-1061. In Handhook 0/ North American lndians, Volume 13, Part 1. Plains. Smithsonian Institution, Washington D. C.
Grim,]. A. 1983 The Shaman Patterns 0/ Religious Healing Among the OJïhwqy Indians. University of Oklahoma Press, Norman, Oklahoma.
Haas,M .. 1960 Sorne genetic affiliations of Algonkian. In Culture in History: Essqys in Honor 0/ Paul Radin, edited by S. Diamond, pp. 977-992. Columbia University, New York.
1965 Is Kutenai related to Algonkian? Canadian Journal o/Linguistics 10:77-92.
Hall, R .. L. 1979 In Search of the ideology of the Adena-Hopewell. In: Hopewell Archaeology: The Chillicothe Conference, edited by David S. Brose and N'omi Greber, pp. 258-265. Ohio State University Press, Kent, Ohio.
Hall, T. D. and C. Chase-Dunn 1993 The World-Systems Perspective and Archaeology: Forward into the Pasto Journal qfArchaeological Research 1 (2): 121-143.
Hallowell, A. 1. 1936 The Passing of the Midewiwin ln the Lake Winnipeg Region. American Anthropologist 38(1): 32-51.
1955 Culture and Experience. University of Philadelphia, Philadelphia, Pennsylvania.
1992 The Ojibwa 0/ Berens River, Manitoba. Ethnography into History. Preface and Afterward by Jennifer S. H. Brown. Case Studies in Cultural Anthropology. Harcourt Brace Janovich College Publications, Orlando.
Halverson, C. 1988 An Archaeological Survry 0/ Stephen, Cameron, Flint and Kakagi Lakes in Northwestern Ontario. Ontario Conservation Archaeology Report #12, North-western Region, Ontario Ministry of Culture and Communications, Kenora.
Halverson, C. 1992 Turtles and Tourists: Excavations at the Nestor Falls Site (DgKI-3) 1989 and 1990. Ontario Conservation Archaeology Report #16, North-western Region, Ontario Ministry of Culture and Communications, Kenora.
Hamilton, S. 1981 The Archaeology 0/ the Wenesaga Rapids, Archaeological Research Report # 17. Toronto: Ontario Ministry of Culture and Recreation, Toronto.
Page 650
1988 The concept of the Cultural Edge on the Boreal Forest of Manitoba and Northwestern Ontario. In Boreal Forest and Sub-Arctic Archaeology, cdited by C. s. "Paddy" Reid, pp. 37-65. No.6, Occasional Publications of the London Chapter, Ontario Archaeological Society Inc., London.
1996 Pleistocene Landscape Features and PIano Archaeological Sites upon the Kaminstiquia River Delta, Thunder Bqy District. Lakehead University Monographs in Anthropology # 1, Thunder Bay.
2000 A Rock Art Panel from the Petawanga Lake Site (EjJc-1), Albat!) River: Northwestern Ontm;o. Preliminary Report prepared for the Eabametoong First Nation.
Hamilton, S., R .. Morrisseau, and Chief Theron McCrady 1995 New Solitudes: Conflicting World Views in the Context of Contemporary Northern Resources Development. Canadian Journal of Archaeology 19: 3-18.
Hare, F. K. 1976 Late Pleistocene and Holocene climates. Some persistent problems. Quaternary Research 6:507-517.
Hare, F. K. and M. K. Thomas 1979 Climate Canada. 2nd Edition. Wiley, Toronto.
Hartley, A. H. 1980 The Expansion of the Ojibway and Place-names into the Lake Superior Region in the Seventeenth Century. Names 28 (1): 43-68.
Harris, R. C. 1987 Historical Atlas of Canada. From the Beginning to 1800. Volume 1 University of Toronto Press, Toronto, Buffalo, and London.
Haselberger, H. 1961 Method for Studying Ethnological Art. Current Anthropology 2(4): 341-384.
Haywood, N. A. 1983 The Quetico Provincial Park Archaeological Survry 1979. Conservation Archaeology Report. North Central Region Report # 11. Thunder Bay: Ontario Ministry of Citizenship and Culture.
Hegman,M. 1998 Technology, Style and Social Practices: Archaeological Approaches. In: The Archaeology of Social Boundaries, edited by M. T. Stark, pp. 264-280. Smithsonian Institution Press, Washington and London.
Heidenreich, C. E 1987. France Secures the Interior, 1740-1755 [map] In: R. C. Harris (Ed.) Historical Atlas of Canada. From the Beginning to 1800. Volume 1. Toronto, Buffalo, and London: University of Toronto Press, Plate 37
Page 651
Heidenreich, C. E. and F. Noël 1987 Re-establishment of Trade, 1654-1666 [map] In: R. C. Harris (Ed.) Historical Atlas of Canada. From the Beginning to 1800. Volume 1. Toronto, Buffalo, and London: University of Toronto Press, Plate 40.
Hickerson, H. 1956 The Genesis of a Trading Post Band: The Pembina Chippewa. Ethnohistory 3: 289-343.
1962 The Southwestern Chippewa: An Ethnohistorical Study. American Anthropological Association. Memoir 92.
1963 The Sociohistorical Significance of Two Chippewa Ceremonials. American Anthropologist 65: 65-83.
1966 The genesis of bilaterality among two divisions of the Chippewa. American Anthropologist 68: 1-26.
1967 Land Tenure of the Rainy Lake Chippewa at the beginning of the 19th century. Smithsonian Contributions to Anthropology. 2: 41-63.
1970 The Chippewa and their Neighbours: A Stucfy in Ethnohistory. Holt, Rinehart and Winston Inc., New York.
Hilger, Sister M. Inez OSB 1936 Chippewa Customs Primitive Man IX (2): 17-24.
1991 [1951] Chippewa Child Life and its Cultural Background. Originally published by the Smithsonian Institute Bureau of American Ethnology as Bulletin 146. Minnesota Historical Society Press, St Paul.
Hill, C. G. 1982 Archaeological Survey of South Geraldton District. In: Studies in West Patricia Archaeology No.3, edited by W. A. Ross, pp. 69-132. Archaeological Research Report # 19. Ontario Ministry of Citizenship and Culture, Toronto.
1983 Archaeology East of Nipigon. Conservation Archaeology Report. With contributions by W. David Hems. North Central Region Report # 13, Ontario Ministry of Citizenship and Culture, Thunder Bay.
Hinshelwood, A. 1996 Boreal Forest fire ecology and archaeological site formation: An example from Northern Ontario. Ontario Archaeology 62:63-92.
Hlady, W. M. 1970 Manitoba-The Northern Woodlands. In Ten Thousand Years: Archaeology in Manitoba, edited by W. M. Hlady, pp. 93-121. Manitoba Archaeological Society, Winnipeg.
Page 652
Hobsbawn, E. 2003 Introduction: Inventing Traditions. In The Invention if Tradition, edited by E. Hobsbawn and T. Ranger, pp. 1-14. University of Cambridge Press, Cambridge, New York, Melbourne, Madrid, Cape Town.
Hodder,1. 1985 Postprocessual Archaeology. Advances in Method and Theory 8: 1-26.
1987 The Contextual Analysis of Symbolic Meanings. In The Archaeology if Contextual Meanings, edited by 1. Hodder, pp. 1-10. Cambridge University Press, Cambridge, New York, Sydney and Melbourne.
1991 Interpretative Archaeology and Its Role. American Antiquity 56 (1): 7-18.
Isajiw, W. W. 1974 Definitions of Ethnicity. Ethnicity 1: 111-124.
International Joint Commission 1917 Lake if the Woods Riference Washington: Government Printing Office, Washington D. C. and Ottawa.
Hoffman, W.]. 1883 Sixty-ninth Regular Meeting, April 17th
, 1883. Transactions if the Anthropological Society if Washington, Volume 2 (February 7th
, 1882-May 15,h 1883): 126-148.
1888 Pictographs and the Ojibwa Religion. American Anthropologist 1: 209-231.
1889 Notes on Ojibwa Folk-Lore. American Anthropologist 2 (3): 215-224.
1891 The Midewiwin or the "Grand Medicine Society". In Seventh Annual Report if the Bureau ifEthnology to the Secretary if the Smithsonian Institution 1885-1886, edited by].W. Powell, pp. 149-299. Government Printing Press, Washington.
Jackson, R. (Editor) 2000. The Explorer's Guide to Lake if the Woods. Lake of the Woods Museum & Aulneau Adventure Tours, Kenora.
Jaritz, G. 1993 Images. A Primer if Computer- Supported Ana/ysis with KLEIO 145. Halbgraue Reihe zur Historischen Fachinformatik. Serie A: Historische Quellenkunden, Band A 22. Max-Planck-Institut für Geschichte St Katherinen: In Kommission bei Scripta Mercaturae Verlag.
Jaritz, G. and B. Schuh 1992 Describing the Indescribable. In: Images and Manuscripts in Historical Computing, edited by Manfred Thaller, pp. 143-153. Halbgraue Reihe zur Historischen Fachinformatik. Serie A: Historische Quellenkunden, Band 14. Max-Planck-Institut für Geschichte St Katherinen: In Kommission bei Scripta Mercaturae Verlag.
Page 653
i-···
Johnson, D., L. Kershaw, A. MacKinnon, and J. Pojar 1995 Plants of the Western Boreal Forest and Aspen Parkland Lone Pine Publishing, Renton, Vancouver and Edmonton.
Johnson, E. 1969a The Prehistoric Peoples of Minnesota Minnesota Historical Society, St Paul.
1969b Preliminary Dates on the prehistoric use of wild rice. Minnesota Archaeologist 30:
31-34.
Johnson, E. and T. Ready 1992 Ceramic funerary masks from McKinstry Mound 2. Midcontinental Journal of Archaeology 17(1): 16-45.
Johnston, B. 1976 Ojibwa Heritage. McClelland and Stewart Inc., Toronto.
1982 Ojibwa Ceremonies. McClelland and Stewart Inc., Toronto.
Jones, L. 1979 Rock Art and Ethnography. Papers from the Fourth Biennial Conference of the Canadian Rock Art Research Associates, Heritage Record 8: 82-97.
Jones, S. 1997 The Archaeology of Ethnicity: Constructing Identities zn the Past and the Present. Routledge, London and New York.
Jones, T. E. H.
1966 Pictographs on Kipahigan Lake, Manitoba. Manitoba Archaeological Newsletter. III (1):3-4.
1968 Problems in Recording Rock Paintings in the Precambrian Shield. Proceedings if the 28'h International Congress if Americanists (1968). Stuttgart-Munich Pp.: 109-116.
1981a The Aboriginal Rock Paintings if the Churchill River. Anthropological Series. Saskatchewan Museum of Natural History, No. 4. Saskatchewan Department of Culture and Y outh, Regina.
1981b Fish Guts and Rock Paintings: Ethnography Informing Archaeology. Canadian Rock An Research Associates Newsletter 15:44-49.
Keeper, F. H. 1912 Letter of Reference. International Joint Commission re: Levels of the Lake of the Woods. Brief for Province of Ontario and Dominion of Canada.
Kaye, L. Kakagi Lake, Ontario / Map 2447 [map]. Series Scale 1 inch to 1/2 mile. Geological Survey Precambrian Shield Geology Ottawa: Geological Survey of Canada, 1981.
KelsaIl,]. P. E.S. Telfer, and T. D. Wright 1977 The iffeets 0/ fire on the eeoloJ!)l 0/ the boreal Jorest, with partieular rejerenee to the Canadian North: a review and seleeted bibliograpf?y Canadian Wildlife Service, Oceasional Paper No 32, Ottawa Supply and Services, Ottawa, Canada.
Kenyon, W. A. 1970 The Armstrong Mound on Rainy River, Ontario. Canadian Historie Sites Oeeasional Papers in ArehaeoloJ!)l and History 3: 65-84. National Historie Sites Service, Ottawa.
Keyser,]. D. 1977 Writing-On-Stone: Roek Art on the Northwestern Plains. Canadian Journal 0/ ArehaeoloJ!)l 1: 15-80.
Keyser,]. D. and M. A. I<lassen 2001 Plains Indian Rock Art. Seattle and London: University of Washington Press and UBC Press, Vancouver and Toronto.
Kidd, K. E. 1965 Birch-Bark Scrolls in Archaeological Contexts. Ameriean Antiquity 30(4):480-483.
Kinietz, C. 1940 Birch Bark Records among the Chippewa. Indiana Aeademy 0/ Science, Proceedings 49: 38-40.
King,]. C. H. 1991 Collecting in the Sub-Aretie Today: the Entrenehment of Sorne Nineteenth and Twentieth Century Prejudices. Aretie ArehaeoloJ!)l 28(1): 134-144.
Koezur, P. and]. V. Wright 1976 The Potato Island Site, District of Kenora, Ontario. National Museum 0/ Man, Mercury Series, Paper # 51, National Museums of Canada, Ottawa. Pp.: ii-51.
Kohl,]. G.
Page 655
1985 [1860] Kitchi-Gami Lift among the Lake Superior Ojibwqy. Originally published in 1860 London: Chapman and Hall. Reprinted. Minnesota Historical Press, St. Paul.
Knowlton, N. 1992 Passport-to-the-Past Volunteers on ORACA's "Images on Stone" Project. Ontario Rock Art Conservation Association Newsletter. Summer/ Autumn Pp.: 9-10.
La V érendreye, Pierre Gaultier de Varennes de 1927 Journals and Letters if Pierre Gaultier de Varennes de la Vérendrtye and his Sons. Burpee, L. J. (editor), The Champlain Society, Toronto.
Lambert, P. J. n.d. The Northwestern Ontario Rock Art Pro/eet: The 1985 Results. Unpublished report for Northwestern Region Conservation Archaeology Report. No number. Ontario Ministry of Citizenship and Culture, Toronto.
1983 The Northwestern Ontario Rock Art Pro/eet: The 1982 Results. Northwestern Region Report #2. Ontario Ministry of Citizenship and Culture, Toronto.
1985 The Northwestern Ontario Rock Art Pro/eet: The 1984 Results. Northwestern Region Conservation Archaeology Report #2. Ontario Ministry of Citizenship and Culture, Toronto.
Lanford, B. L. 1984 Winnebago Bandolier Bags. Ameriean Indian Art Maga~ne 9(3): 30-37.
Landes, R. 1968 Religion and the Midewiwin. University of Wisconsin Press, Madison.
Lanoue, G. 1989 Images on Stone: A Theory on Interpreting Rock Art. Art Center, Rome.
Lawson, A. C. 1885 Ancient Rock Inscriptions on the Lake of the Woods. Ameriean Naturalist XIX (7): 654-657.
Lay ton, R. 1991 The Anthropology if Art. Second Edition. Cambridge University Press, Cambridge.
Leacock, E. 1973 The Montagnais-Naspaski Band. In Cultural Eeology: Readings on the Canadian Indians and Eskimo, edited f:y B. Cox, pp. 81-100. McClelland and Stewart, Toronto.
Lemaître, S. 1995 Les peintures rupestres au Québec. International Newsletteron Rock Art 11: 23-29.
Lenius, B. J. and D. M. Olinyk
Page 656
1990 The Rainy River Composite: Revisions to Late Woodland Taxonomy. In The Woodland Tradition in the Western Great Lakes: Papers presented to Bidon Johnson, edited by G. E. Gibbon, pp. 77-112. Publications in Anthropology # 4. University of Minnesota, Minneapolis.
Lenz, M.]. 1986 The Sttiff 0/ Dreams: Native American Dolis. Museum of the American Indian, New York.
Leroi-Gourhan, A. 1982 The Dawn o/Paleolithic Art. Cambridge University Press, Cambridge.
Lessard, F. D. 1986 Great Lakes Indian "Loom" Beadwork. American Indian Art Maga;;jne 11 (3):54-61.
Lewis-Williams,]. D. and T. A. Dowson 1994 Vision and Power in the Neolithic: Evidence from the Decorated Monuments. Current Anthropology 34: 55-65.
Lévesque, R. 1968 Des Petroglyphes au Québec. Vie en Arts Vol 52: 18-23.
Lieberman, D. E., B. A. Wood, and D. R. Philbeam. 1996 Homoplasy and early Homo: analysis of the evolutionary relationships of H. habilis sensu stricto and H. Rufoldensis.Journal o/Human Evolution 30: 97-120.
Lipsett, B.M. 1970 Comparative Study to Determine the Origin of sorne Canadian Shield Rock Paintings. In Ten Thousand Years: Archaeology in Manitoba, edited by W. M. Hlady, pp. 181-321. Manitoba Archaeological Society, Winnipeg.
Long,]. 1971 V l2Yages and travels 0/ an Indian intepreter and tradefj describing the manners and customs 0/ the North American Indians; with an account 0/ posts situated on the river Saint Laurence) Lake Ontario) &c. To which is added a vocabulary 0/ the Chippewqy language ... a list 0/ words in the Iroquois) Mohegan) Shawanee) and Esquimeaux tongues) and a table) shewing the analogy between the Algonkin and Chippewqy languages. London, 1791. Reprinted by Coles Publishing Company, Toronto.
Lorblanchet, M. ,M. Labeau,]. L. Vernet, P. Fitte, H. Valladas, H. Cachier, and M. Arnold 1990 Palaeolithic Pigments in the Quercy, France. Rock Art Research 7 (1): 4-20.
Lorenz, J. G. and D.G .Smith 1994 Distribution of the 9-Bp Mitochondrial DNA Region V Deletion among North American Indians. Human Biology 66: 777-788.
1996 Distribution of Four Founding Mtdna Haplogroups among Native North Americans. American Journal 0/ Physical Anthropology 101: 307-323.
Page 657
1997 Distribution of sequence variation in the mtDNA control region of native North Americans. Human Biology 69: 749-776.
Lothson, G. A. 1976 The J effèrs Petrog!Jphs Site. A S urvry and Ana!Jsis of the Carvings. Minnesota Prehistoric Archaeology Series No. 12. Minnesota Historical Society, St. Paul.
Lowie, R. H. 1910 The Assiniboine. Anthropological P apers of the American Museum of N atural History. 4 Part 1.
Lugenbeal, E. N. 1976 The Archaeology of the Smith Site: a Stucfy of the Ceramics and Culture History of Minnesota Laurel and Blackduck Unpublished Ph.D. Dissertation Department. of Anthropology, University of Wisconsin, Madison.
1978 The Blackduck Ceramics of the Smith Site (21KC3) and their Implications for the History of the Blackduck Ceramics and Culture in Northern Minnesota. Midcontinental Journal of Archaeology. 3 (1): 45-68.
Lyford, C. A 1943 The Crcifts of the Ojibwa (Chippewa). Office of Indian Affairs, Indian Handicrafts. No. 5. Phoenix Indian School, Phoenix.
Lyman, R. L., and M. J. O'Brien 2001 The Direct Historical Approach, Analogical Reasoning, and Theory ln
Americanist Archaeology. Journal of Archaeological Method and Theory 8 (4): 303-342.
McAndrews, J. H. 1969 Paleobotany of a Wild Rice Lake in Minnesota. Canadian Journal of Botatry 47: 1671-1679.
McPherron, A. 1967 The Juntunen Site and the Late Woodland History of the Upper Great Lakes Area. Anthropological Papers # 30, Museum of Anthropology, University of Michigan, AnnArbor.
Macfie, J. A. 1992 The Stories on the Cliffs. Ontario Rock Art Conservation Association Newsletter. Spring Pp. 7-9. Originally published by SYLVA-Your Lands and Forest Review, Ontario Department of Lands and Forests November-December 1959.
MacNeish, R. S. 1954 The Stott Mound and Village, Near Brandon, Manitoba. In Annual Report of the National Museum of Canada 1952-1953, pp. 20-65. Bulletin No 132, Ottawa: National Museums of Canada.
Page 658
1958 An Introduction to the ArchaeoloJ!Y if Southeast Manitoba, Bulletin # 157, National Museum of Canada, Ottawa.
Mackenzie, A. 1902 [1793] V l!Yages jrom Montreal through the continent if North America to the Frozen and Pacific oceans in 1789 and 1793 with an Account if the Rise if the State if the Fur Trade. George N. Morang and Co. Ltd., London.
Maddox, J. L. 2003 [1923] Shamans and Shamanism. New York: Dover Publications. Originally published by The Macmillan Company, New York under the title The Medicine Man: A Sociological Study of the Character and Evolution of Shamanism.
Margry, P. 1888 Relations et mémoires inédits pour servir à l'histoire de la France dans les pcrys d'outremer [micro forme] :Tirés des Archives du Ministère de la marine et des colonies. Challamel Aîné, Paris.
Malhi, R. S., B. A. Schultz, and D. G. Smith 2001 Distribution of mitochrondrial DNA lineages among Native American tribes of northeastern North America. Human BioloJ!Y 73: 17-55.
Malru, R. S.,]. A. Eshleman,]. A. Greenburg, D. A. Weiss, B. A. S. Shook, F. A. Kaestle,]. G. Lorenz, B. M. Kemp,]. R.Johnson, and D. G. Smith
2003 The structure of diversity witrun New World Mitochrondrial DNA halplogroups: Implications for the prehistory of North America. American Journal if Human Cenetics 70: 905-919.
Malhi, R. S., H. M. Mortensen,]. A. Eshleman, B. M. Kemp,]. G. Lorenz, F. A. Kaestle,]. R. Johnson, C. Gorodezky, and D. G. Smith
2003 Native American mtDNA prerustory in the American Southwest. American Journal ifPhysicalAnthropology 120: 108-124.
Malhi, R. S., K. Breece, B. A. S. Shook, F. A. Kaestle, J. C. Chatters, S. Hackenburger, and D. G. Smith
2004 Patterns of mtDNA diversity in Northwestern North America. Human Biology 76: 33-54.
Mallory, E. S. 1961 The Agawa Pictographs. Canadian Ceographical Journal 69: 126-129.
Mallory, G. 1880 Cesture Signs and Signais if the North American Indians. Smithsonian Institution, National Anthropological Archives, Box 8: 2372, Folder 1.
1881 A Collection if Cesture-Signs and Signais if the North American Indians with Some Comparisons. Smithsonian Institution, Bureau of American Ethnology, First Annual Report for 1879-180.
Page 659
1894 Picture Writing qf the American lndians. Smithsonian Institution, Bureau of Ethnology, 10th Annual Report for 1893.
Matson, G. A., E. A. Koch, and P. Levine 1954 A Study of the Hereditary Blood Factors among the Chippewa lndians of J\1innesota. America Journal ifP~sical Anthropology 12: 413-426.
Maurer,]. D. and]. P. Whelan 1977 A Canadian Shield Pictographie Design. American Antiquiry 42: 196-202.
Mayer-Oakes, W. T. 1970 Archaeological Investigations in the Grand Rapids Manitoba Reservoir 1961-1962. Occasional Paper # 3, Department of Archaeology, University of Manitoba, Winnipeg.
Menu, M. and P. Walter 1991 Les Premiers Artistes Peintres: L'Analyse Physico-Chimique des Peintures Préhistorique pemet de J\1ieux Comprendre L'art des premiers artistes. La Recherché 22 (235): 1086-1089.
Meyer,M. L. 1990 Signatures and Thumbprints: Ethnicity among the White Earth Anishinnabeg, 1889-1920. Social Science History 14 (3): 305-345.
Meyer, D. and S. Hamilton 1994 Neighbours to the North. In Plains lndians AD 500-1500: The Archaeological Past if Historie Groups, edited by K. H. Schlesier, pp. 71-95. University of Oklahoma Press, Norman.
Meyer, D. and D. Russell 1987 The Selkirk Composite of Central Canada: A Reconsideration. Arctic Anthropology 24 (2): 1-31.
J\1inning, G. V., W. R. Cowan, D. R. Sharpe, and T. A. Warman 1994 Quaternary Geology and Drift Composition, Lake if the Woods Region, Northwestern Ontario., Geological Survey of Canada Memoir 436. J\1inster of Energy, J\1ines and Resources Canada ..
J\1inning, G. V. and D. R. Sharpe Rowan Lake, Ontario/1773A [map]. Canada 1:100,000. Surficial Geology Ottawa: Geological Survey of Canada, 1991.
Rat Portage Bqy, Ontario/I770A [map]. Canada 1:100,000. Surficial Geology Ottawa: Geological Survey of Canada, 1991.
Northwest Angle Ontan'o - United States/ 1771A [map] Canada 1:100,000. Surficial Geology. Ottawa: Geological Survey of Canada. 1991,
Page 660
Molyneaux, B. L. 197 4 Trent University Rock AI1 Research Prqject S ummary Repol1. 1974 Field S eason. Department of Anthropology, Trent University.
1975 CCI/TRAP Pictograph Condition Inspection SurtJry. Canadian Conservation Institute, Ottawa.
1977 Formalism and contextualism: An historiograpf?y if rock al1 research in the New World Master's thesis. Trent University, Peterborough, Ontario.
1978 The Lake if the Woods Petrogfyph Prqject. Report to the Ontario Heritage Foundation.
1979 The Whiteftsh Bcry Pictograph Prqject. Report to the Ontario Heritage Foundation. No licence.
1980a. Landscape Pictures. Rotunda 13 (3): 6-11.
1980b In the Wake of the Manitou Rqyal Ontario Museum Newsletler 185: New Series.
1980c The Stutfy if Prehistoric Sacred Places Evidence from Lower Manitou Lake. Archaeology Paper 2, Royal Ontario Museum, Toronto.
1981 Archaeological Fieldwork Repol1: Licence Number 80-F-0383. Department of New World Archaeology. Royal Ontario Museum, Toronto.
1982 Preliminary RepOI1: Licence 82-60. Department of New World Archaeology. Royal Ontario Museum, Toronto.
1985a Archaeological Fieldwork Repol1: Licence Number 84-36. Department of New World Archaeology. Royal Ontario Museum, Toronto.
1985b Archaeological Fieldwork Repol1: Licence N umber 85-32. Department of New W orld Archaeology. Royal Ontario Museum, Toronto.
1987 The Lake of the Painted-Cave. Archaeology 40 (4):18-25.
Morphy, H. 1989 Introduction. In Animais in Art, edited by H. Murphy, pp. 144-157. Volume 7. Unwin Hyman, London, Sydney and Wellington.
Myers, R. M. and]. M. Taylor 1974 An Investigation of the Natural Deterioration of Aboriginal Rock Paintings by Scanning Electron Microscopy and X-Ray Microanalysis. Microbeam Ana/ysis Society) T utorial and Proceedings if the Ninth Annual Conforence. Carlton University: Ottawa.: 16 A-16 C.
Page 661
Nelson, J. and A. Hinshelwood 1998 Archaeology and Native Spirit Beliefs in Quetico. The Boundary Waters Journal Summer: 12- 16.
Nelson, R. E. 1983 Inscribed Birch Bark Scroil and other Objects of the Midewiwin. In Actes de Quartor::;jéme Congrés des Aigonquinistes) edited by W. Cowan, pp. 219-236. Carleton University, Ottawa. 1984 Midewiwin Medicine Bags of the Ojibwa. In Papers if the Fifteenth Aigonquian Conference, edited by D. H. Pentland, pp. 397-408. Carleton University, Ottawa.
Noble, W. C. 1968 "Vision Pits", Cairns and Petroglyphs at Rock Lake, Algonquin Provincial Park, Ontario. Ontario Archaeology 11: 47-64.
Nurse, D. 1997 The Contributions of Linguistics to the History of Africa. Journal if Ajrican History 38: 395-391.
Nute, G. 1948 The Painted Rock of Crooked Lake. Minnesota History 29: 130-136.
Oberholtzer, C. and N. N. Smith 1995 l'm the Last One Who Does Do It: Birch Bark Biting, An Almost Lost Art. In Papers if the Fifteenth Aigonquian Cotiference, edited by D. H. Pentland, pp. 306-321. Carleton University, Ottawa.
OIson, S. F. 1961 Painted Rocks. National Parks Maga::;jne 135 (163): 4-7.
Ossenberg, N. S. 1974 Origins and Relationships of Woodland Peoples: The Evidence of Cranial Morphology" In: Aspects if Upper Great Lakes Anthropology Papers in Honour if Llqyd A. Wi!fOrd, edited by E. Johnson, pp. 15-39. Minnesota Prehistoric Archaeology Series II, St Paul: Minnesota Historical Society.
Owen, R. C. 1965 The Patrilocal Band: A Unguisticaily and Culturaily Hybrid Social Unit. American Anthropologist New Series 67 (3): 675-690.
Pastershank, G. 1989 An Archaeological Survry if Sabaskong Bery) Lake if the Woods. Ontario Conservation Archaeology Report #13, North-western Region. Ontario Ministry of Culture and Communications, Kenora.
Peers, L. 1994 The Ojibwa if Western Canada 1780-1870. Manitoba Studies in Native History VIII. University of Manitoba Press, Winnipeg.
Page 662
1999 'Many Tender Ties'; the Shifting Contexts and Meanings of the S Black Bag. World Archaeology 31 (2): 288-302.
Pelleck,]. A. 1981 An Archaeological Survey of the Upper English River. In Studies in West Patricia No.2: 1979-80, edited by C. S. Paddy Reid and W. A. Ross, pp. 115-130. Report # 16, Ontario Ministry of Culture and Communication, Toronto.
Pelshea, V. 1979 A Medium for Recording Pictographs Alternative or Complementary to Conté. Canadian Rock Art Research Associates Newsletter 2; 24-25.
1980 The West Patricia Project; First Year. In Studies in West Patricia No. 1: 1978-79, edited by C. S. Paddy Reid and W. A. Ross, pp. 49-69. Report #15, Ontario Ministry of Culture and Communication, Toronto.
Penny,D. W. 1986 -1987 The Origins of Indigenous Ontario Arts Tradition: Ontario Art from the Late Archaic through the Woodland Periods, 1500 B. c.-A. D. 600. Journal of Canadian Studies 21 (4): 37-55.
1991 Floral Decoration and Cultural Change: A Historical Interpretation of Motivation. American Indian Culture and Research Journal 1 5:53-77.
Pepe, c.,]. Clottes, M. Menu, and Ph. Walter 1991 Le liant des peintures préhistorique ariègeosis. Compte Rendu Academie des Sciences, Paris. 312 (II): 929-934.
Petch, V. 1991 An Archaeological Survey of the Sasaginnigak River System lncluding the Kautunigan Lake Area. Manitoba Archaeological Journal 1 (2): 23-44.
Pettipas, L. 1982 Naturalism and Stylization in Canadian Shield Rock Art; A Discussion. Canadian Rock Art Research Associates Newsletter 17 (May); 13-27.
1991a Discovery Rock Art. Insights 1 (2): 4.
1991 b Rock Pictography and Oral Literature. Manitoba Archaeological S ocie!) Journal 1 (2):45-49.
Phillips, R. B. 1984 Zigzag and Spiral: Geometrie Motifs in Great Lakes lndian Costume. In Papers if the Fifteenth Algonquian Conference, edited by W. Cowan, pp. 409-424. Carleton University, Ottawa.
1991 Comment: 'Catching Symbolism': Studying Style and Meaning ln Native Ameriean Art. Arctic Anthropology 28 (1):92-100.
Page 663
1999 Trading Identitt'es: The Souvenir in Native North American Artfrom the Northeast, 1700-1900. University of Washington Press, Seattle and London, McGill-Queen's University Press, Montreal and I<ingston.
Price, S. 1989 Primitive Art in Civilized Places. The University of Chicago Press, Chicago and London.
Pohorecky, Z. 1968 Rupestral Art in the Precambrian Shield of Canada. In: Proceedings 0/ the 28'h International Congress 0/ Americanists (1968), edited Sol Tax, pp. 103-1 08. Stuttgart-Munich.
1980 A Study of Pigmentation for Planning Conservation at the Pictograph Sites on Hickson and Maribelli Lakes, Saskatchewan. Canadian Rock Art Research Associates Newsletter (12 May): 26-39.
1981 Human Figures in Northwestern Shield Rock Paintings. Canadian Rock Art Research Associates Newsletter(16 November): 12-18.
Pohorecky, Z. S. and T. E. H. Jones 1966 Recording Pictographs Man 1: 104.
1967 Recording Pictographs Man 2: 305-306.
1968a Canada's Oldest Known Pictograph? Saskatchewan History XXI (1): 30-36.
1968b Aboriginal Pictographs on I<ipachigan Lake in the Precambrian Shield of Saskatchewan and Manitoba. The Musk-Ox (2): 3-36.
Porter, V. 1994 A Guide to the Description of Architectural Drawings. G. K. Hall, New York.
Pufahl, J. K. 1990 Conserving Aspects of Native Culture through Pictograph Documentation in Northwestern Ontario. In: First Annual Archaeological Report, Ontario, pp.: 15-16 Volume 1 (New Series). The Ontario Heritage Foundation, Toronto.
Rabb, T. K. and J. Brown 1988 The Evidence of Art: Images and the Meaning in History. In Art and History: Images and Their Meaning, edited by R. 1. Rotbberg and T. K. Rabb, pp. 1-6. Cambridge University Press, Cambridge.
Rackerby, F. 1975 Current Research: The Northwest. American Antiquiry 40 (4): 479-485.
Rainbird, P.
Page 664
2002 Making Sense of Petroglyphs. In Inscribed Landscapes: Making and Marking Place, edited by M. Wilson and B. David, pp. 93-103. University of Hawai'i Press, Honolulu.
Rajnovich, G. 1980a Ballysadare (DkKp-l0): A Laurel Blackduck Site at the Source of the Winnipeg River. Ontario Archaeology 33: 37-58.
1980b The Rainy Lake Archaeological Survey: 1978-1979. Ontario Ministry CI! Culture al1d Recreatiol1, Archaeological Reports 011 File.
1980c Two Pictographs Near Ignace, Northwestern Ontario. Canadian Rock Art Research Associates Newsletter 14: 34-38
1980d Paired Morphs at Cuttle Lake in Northwestern Ontario. Cal1adial1 Rock Art Research Associates Newsletter 13: 46-50
1981a Four Pictograph Sites in the Rainy Lake Area of Northwestern Ontario Canadian Rock Art Research Associates Newsletter 1 5: 12-32.
1981b The Deer Lake Pictographs. In: Studies in West Patricia. No.2: 1979-80, edited by C. S. "Paddy" Reid and W. A. Ross, pp.280-307. Report # 16, Ontario Ministry of Culture and Communication, Toronto.
1981c Four Pictograph Sites in the Rainy Lake Area of Northwestern Ontario. Canadian Rock Art Research Associates Newsletter 15: 13-32.
1984 A Study of the Possible Wild Rice Gathering on the Lake of the Woods, Ontario. North Americal1 Archaeologist 5 (1): 187-215.
1989 Visions in the Quest for Medicine: An Interpretation of the Indian Paintings of the Canadian Shield. Midcontinental Journal if Archaeology 14 (2): 179-225.
1994 Readil1g Rock Art: Intepreting the Indian Paintings if the Canadian Shield. Natural Heritage/Natural History Inc., Toronto.
Rapaport, H.
Ray, A.
1997 Is There Truth in Art? Cornell University Press, Ithaca and London.
1974 lndians in the Fur Trade: Their Role as Trappers, Hunters and Middlemm in the Lands S outhwest if the Hudson Bqy 1660-1870. University of Toronto Press, Toronto.
Reagan, A. B. 1921 Sorne Chippewa Medicinal Receipts American Anthropologist N. S. 23 (2): 246-249.
1922 Medicine Songs of George Farmer American Anthropologist N. S. 24 (3): 332-369.
Page 665
1935 A Ritual Parchment and Certain Historical Charts of the Bois Fort Ojibwa of Minnesota. Americana 29: 228-244.
Redsky,]. 1972 Great Leader if the Ojibwqy: Mis-quona-queb, edited by James R Stevens McClelland and Stewart, Toronto.
Reid, C. S. "Paddy" 1974 Archaeological S urvry ifS andbar Lake and Arfjacent Drainages. Division of Parks and Historical Sites Branch, Northwestern Region. Ontario Ministry of Natural Resources.
1976 Environment, man and mqymqyqwqyshi: The Dimensions if Past Human Cultural Impact in the Lake if the Woods Area; Regional Archaeologist, Historical Planning and Research Branch, Northwestern Region, Lake of the Woods Planning Area.
1977 Environment, Man and Mqymqyqwqyshi: The Dimensions if Past Human Cultural Impact in the Lake if the Woods Area; Part II 1977. Data Box 312. Ontario Ministry of Culture and Recreation, Toronto.
1979 The Tranquil Channel Petroglyphs: Cultural Continuities in Lake of the Woods Rock Art. Papers from the Fourth Biennial Conference of the Canadian Rock Art Research Association, Victoria, British Columbia. Heritage Record 8: 245-53.
1988 Sorne Ideas Concerning the Formulation of Research Designs and Excavation Methodologies on Boreal Forest Habitations Sites. Midcontinental Journal if Archaeology 13 (2):187-221.
Reid, C. S. "Paddy" and G. Rajnovich 1991 Laurel: A Re-evaluation of the Spatial, Social and Temporal Paradigms. Canadian Journal if Archaeology 15: 193-234.
Reid, D. and]. Vastokas 1984 From the Four Corners: Native and European Art in Ontario 5000 BC to AD 1867. Art Gallery of Ontario, Toronto.
Rhodes, R. A. and E. M. Todd 1981 Subarctic Languages. In: Handbook if the North American Indians, edited by ]. Helm, pp. 52-66. Volume 6. Smithsonian Institution, Washington D. C.
Rich, E. E. 1949 James Isham's observations on Hudsons Bqy, 1743 and Notes and observations on a book entitled A vqyage to Hudsons Bqy in the Dobbs Gallry, 1749. Introduction by E.E. Rich, assisted by A. M. Johnson. Champlain Society, Toronto.
Rifkin, A.
Page 666
. ~.
1986 Art's Histories. In The NeJJ! Art Histm)' edited by A. L. Rees and F. Borzello, pp. 2-10. Camden Press, London.
Ritzenthaler, R. and P. Ritzenthaler 1970 The Woodland Indians 0/ the Western Great Lakes. The Natural History Press, Garden City, New York.
Ritzenthaler, R. 1972 Woodland Indian Art. In: American Indian Art: Form and Tradition. An Exhibition organized by Walker Art Center, edited by R. Ritzenthaler, pp. 49-55. Indian Art Tradition: The Minneapolis Institute of Arts, 22nd October - 31 st December 1972. E. P. Dutton and Co. Inc.: New York.
Rogers, D. P. 1962 Indian Rock Paintings. Ontan·o Archaeology Series B. 2: 47-48.
Rogers, E. S. 1963 Changing Settlement patterns of the Cree-Ojibwa of northern Ontario. Southwestern Journal 0/ Anthropology 19: 64-88.
1972 The Round Lake Oijiwa. Occasional Paper 5. Toronto: Royal Ontario Museum.
1973 Review of The Chippewa and their Neighbours: A Stutjy in Ethnohistory by Harold Hickerson. Michigan History 57: 83-86.
1975 Review of the Sacred Scroils of the Southern Ojibwa by Selwyn Dewdney. Ethnohistory 22: 81-85.
1983 Cultural Adaptations: The Northern Ojibway of the Boreal Forest 1670-1980. In Boreal Forest Adaptations: the Northern Algonkians, edited by A.T. Steegman, pp. 85-141. Plenum Press, New York .
Rogers, E. S. and J. G. E. Smith 1981 Environment and Culture in the Shield and Mackenzie Borderlands. In Handbook 0/ North American Indians, Volume 6, The Subarctic, edited by J. Helm, pp. 141-144. Washington, Smithsonian Institution, Washington D. C.
Rogers, E. S. & Taylor, J. G. 1981 Northern Ojibwa. In: Handbook o/the North American Indians, Volume 6, The Subarctic, edited by J. Helm, pp. 213-243, Smithsonian Institution, Washington D. C.
Rogers, E. S. and M. B. Black 1986 Subsistence Strategy in the Fish and the Hare Period, Northern Ontario: the Weagomow Ojibwa 1880-1920. Journal 0/ Anthropological Research 31: 1-43.
Rousseau, J. 1956 L'origine du motif de la double-curve dans l'art algonkin. Anthropologica 2:218-221.
Page 667
Rose, G. 2001 Visual Methodologies: an Introduction to the Intepretation ifVisual Materials. Thousand Oaks, London; Sage, California.
Rowe,]. S. 1972 Forest Regions if Canada. Department of the Environment, Canadian Forestry Service, Publication # 130.
Rusak,]. 1989 An Archaeological Survry of a Portion of the Winnipeg River, North-western Ontario. Ontario Conservation Archaeology Report #14, North-western Region, Ontario Ministry of Culture and Communications, Kenora.
1991 Two pictographs from White Gtter Lake. Ontan"o Rock Art Conservation Association Newsletter. December: 4-5.
1992a The White Otter Lake Pictograph Prqject 1991 Results. Papers in Ontario Rock Art #1. Kenora: Ontario Rock Art Conservation Association.
1992b "Images on Stone" -The Second Year Field Activities for ORACA Crew. Ontario Rock Art Conservation Association Newsletter. Summer/ Autumn: 2-7.
Sackett, J. R. 1977 The meaning of style in archaeology: a general model. Amencan Antiqui!} 42 (3): 369-380.
1982 Approaches to style in lithic archaeology. Journal if Anthropological Archaeology 1: 59-112.
1985. Style and ethnicity in the Kalahari: a reply to Wiessner. American Antiqui!} 50: 151-159.
Sager, D. 1994 The Rose Collection of Moccasins in the Canadian Museum of Civilization: Transitional Woodland/Grassland Footwear. Canadian Journal of Native Studies 14 (2): 273-304.
1995 The possible origins of the Blackfeet crooked nose moccasin design. American lndian Art Maga=?fne 20 (4): 36-43.
1996 An unusual eastern grasslands Ojibway srurt type. American Indian Art Maga:dne 22 (1) 36-47.
2000 The dual side seam: an overlooked moccasin. American lndian Art Maga=?fne 25 (2) 68-75.
2002 Tracking some Characteristics of Assiniboin Quillwork and Beadwork. Amencan lndian Art Maga=?fne 27 (2) 74-81.
Page 668
Sapir, E. 1916 Time Perspective in Aboriginal American Culture: A Stucfy in Method. Memoir 90, Geological Survey of Canada, Ottawa.
Saouter, C. 1998 Le languge visuel. XVZ Montréal.
Schenk, T. 1994 Identifying the Ojibwa. In: Actes du Vingt-Cinquième Congrès des Algonquinistes, edited by W. Cowan, pp. 395-405. Carleton University, Ottawa.
Schoolcraft, H. R. 1851 Personal Memoirs if a Residence ifThirry Years with the Indian Tribes on the American Frontiers: With Briif Notices if Passing Events, Facts and Opinions, A.D. 1812 to A.D. 1842. Lippincott and Grambo, Philadelphia:
1983 [1860] Information Respecting the History Conditions and Prospects if the Indian T ribes if the United States. Volume 6,J. B. Lippincott and Co., Philadelphia.
Schultz, B. A., R. S. Malhi, and D. S. Smith. n.d. Examining the Proto-Algonquian Migration: Analysis of mtDNA. Paper presented at the 32nd Algonquian Conference October 27-29 2000, Montreal, Quebec, Canada.
Schurr T. G., S. W. Ballinger, Y. Y. Gan,]. A. Hodge, D. A. Merriwether, D. N. Lawrence, W. C. Knowler, K. W. Weiss, and D. C. Wallace
1990 Amerindian Mitochondrial DNAs Have Rare Asian Mutations At High Frequencies, Suggesting They Derived From 4 Primary MaternaI Lineages. American Journal ifHuman Genetics 46: 613-623.
Scott, E. 1980 Pictographs on the Clearwater River. Saskatchewan Archaeological Society Newsletter (New Series). l (3): 6-7.
Scozzari, R, F. Cruciani, P. Santolamazza, D. Sellitto, D. E. C. Cole, L. A. Rubin, D. Labuda, E. Marini, V. Succa, G. Vona, and A. Torroni
1997 MtDNA and Y chromosome-specifie polymorphisms in modern Ojibwa: Implications about the origin of their gene pool. American Journal ifHuman Genetics 60: 241-244.
Shanks, M. and C. Tilley 1987 Social Theory and Archaeology. Polity Press, Cambridge.
1994 Re-Constructing Archaeology: Theory and Practice. Second Edition. Routledge, London and N ew York.
Shennan, S.].
Page 669
1989 Introduction: archaeological approaches to cultural identity. In Approaches to Cultural Identiry) edited by S. J. Shennan, pp. 1-32. One World Archaeology. Volume 10. Unwin Hyman, London.
Shook, B. A. S. 2004 Detecting Relationships in the Great Lake.r region using ancient MtDNA. Paper presented at American Association for Physical Anthropologists meeting Tampa, FL, April 14-17.
Siebert, F. T. Jr. 1967 The Original Home of the Proto-Algonquian People. Contribution to Anthropology: Unguistics 1 (Algonquian). Anthropological Series 78. Bulletin 214, National Museum of Canada, Ottawa. Pp: 13-47.
Skinner, A. B. 1911 Notes on the Eastern Cree and Northern Saulteaux. American Museum ifNatural History. Anthropological Papers 9 Part 1.
1915 The Cultural Position of the Plains Ojibway. American Anthropologist New Series 16 (2): 314 - 318.
Smith, D. G., R. S. Malhi, J. Eshleman, J. G. Lorenz, and F. A. Kaestle 1999 Distribution of mtDNA haplogroup X among Native North Americans. American Journal ifPf?ysicalAnthropology 110: 271-284.
Smith, S. A. 1981 The Archaeology of Red Lake. In: Studies in West Patricia. No.2: 1979-80. Report # 16, edited by C. S. "Paddy" Reid and W. A. Ross, pp. 25-114. Ontario Ministry of Culture and Communication, Toronto.
Smith, T. S. 1995 The Island if the Anishnaabeg. University of Idaho Press, Moscow, Idaho.
Speck, F. G. 1914 The Double Curve in Birch Bark: A Circumpolar Trait. Museum of the America Inclian. Heye Foundation. Indian Notes and Monographs 11 (2): 45-125.
1915 Myths and Folklore of the Timiskaming Algonquin and Timagami Ojibwa. Geological Survey of Canada Bureau of Mines, Memoir 71, no. 9, Anthropological Series, Ottawa Government Printing Bureau pp. 87.
1937 Montagnais Art in birch bark: A Circumpolar Trait. Museum if the American Indian) Hrye Foundation) Indian Notes and Monographs 11 (2): 45-125.
Spier, L. 1915 Review of Speck's 'Double-curve Motive'. American Anthropologist N.S. 17: 334-346.
Page 670
. r-..
Soil Research InstÏtute Soil Map - Reconnaissance SUf7Jf!) of South-eastern Manitoba [map]. 1 Inch to 2 miles. Soil Research Institute, Research Branch, Canada Department of Agriculture, Ottawa. 1964.
Stark, M. T. 1998 Technical Choices and Social Boundaries in Material Culture Patterning: An Introduction. In: The Archaeology of Social BoundariesJ edited by M. T. Stark, pp. 1-11. Smithsonian Institution Press, Washington and London.
Steinbring, J. H. 1975 Taxonomie and Associational Considerations of Copper Technology during the Archaic Traditions. University of Minnesota, PhD thesis. Anthropology .
1976 Lunar Cultism in the Ancient Great Lakes. Bolletino dei Centro Camuno di Studi Prehistori 13 -14: 153-166.
1977 The Mud Portage Petroglyph Site in Northwestern Ontario: A Summary. Rock An PapersJ San Diego Museum Papers 21 (4): 139-143.
1979a 1979 University of Winnipeg Rock Art Studies, Canadian Rock An Research Associates Newsletter 12: 13-18.
1979b The Caribou Lake Gorget: Cross-Media form Correspondences in the Prehistoric Rock Art of the Interior of North America. Canadian Rock Art Research Association 1977. Papers from the Fourth Biennial Conference of the Canadian Rock Art Research Associates, Heritage Record 8: 99-125.
1981. Saulteaux of Lake Winnipeg. In: Handbook of Nonh American Indians, Volume 6, edited by J. Helm, pp. 244-255. Smithsonian Institution, Washington D. C.
1982 Shamanistic Manipulations and the AIgonkian ldiom in the Archaeoethnology of the Rock Art. American Indian Rock VII and VIII: 212-220.
1990 Post-Archaic Rock Art in the Western Great Lakes. In The Woodland Tradition in the Western Great Lakes: Papers Presented to Eldon Johnson, edited by G. E. Gibbon, pp. 167-194. Publications in Anthropology # 4. University of Minnesota ,Minnneapolis, Minnesota.
1992 Phenomenal Attributes: Site Selection Factors in Rock Art. American Indian Rock An: 17: 102-113.
1993 Dating Rock Art in the Northern Midcontinent. American Indian Rock An. XII: 15-31.
1995 Rock An Studies in the Americas. Papers presented to Symposium b of the First AURA Congress, Darwin 1988. Oxbow Monograph 45. Oxbow Books, Oxford.
Page 671
1998 Aboriginal Rock Paintings Sites in Manitoba. Manitoba Archae%gica/ Jouma/8 (1 and 2).
1999 Dating and the Earliest Known Rock Art. In Papers presented in Symposia 1-3 0/ the SL4RB Congress, Cochabamba, Bolivia, edited by M. Strecker and P. Bahn, pp. Oxbow Books, Oxford.
Steinbring, J. and R. Callaghan 1985 Comments on the Preparation and Weathering of Petroglyphs at the Mud Portage Site (DkKr-4) in Northwestern Ontario. Manitoba Archae%gica/ Quarterfy 10: 4-34.
Steinbring, J. E. Danziger and R. Callaghan 1987 North Archaic Petroglyphs in Northern North America. Rock Art Research 4 (1): 3 -16, 150-161.
Steinbring,J. and D. Elias 1967 Animal Designs in Rock Paintings Zo%g 8 (4): 10-12.
1968a Technical Refinements in the Study of Canadian Shield Rock Paintings. Manitoba Archae%gica/ News/etter IV (3): 3-8.
1968b A Key Pictograph from the Bloodvein River, Manitoba. American Antiquity 33: 499-501.
Steinbring, J., G. Granzberg and M. Lanteigne 1995 Elemental Analysis in Pursuit of Function in Rock Art. In: Rock Art in the Americas. Rock Art Studies in the Americas. Papers presented to Symposium b of the AURA Congress, Darwin 1988, edited by J. Steinbring, pp. 55-65. Oxbow Monograph 45. Oxbow Books: Oxford.
Steinbring, ]. and H. Iwacha 1982 Long Lake Rock Painting Site, Manitoba. News/etter 0/ the Canadian Rock Art Research Associates 18: 7-23.
Steinbring, J. and D. Simpson 1986 The Mud Portage Site (DrKr-4). A preceramic Context for Petroglyphs ln
Northwest Ontario. The Amencan lndian Rock Art 10: 145-156.
Steinbring,]. and S. Steinbring 1978 Machin Point Petroglyph Site, Shoal Lake, Ontario Rubbings and photographs. J uly 13-15, 1978. News/etter 0/ the Canadian Rock Art Research Associates 10: 11-15.
Steinbring, J. and Nielson 1986 Reinterpretations of the Rush Bay road Site in Northwestern Ontario. Manitoba Archaeological Quarterly, 10 (1): 4-35.
Page 672
Stevens, J. R. 1971 Sacred Legends if the Sanc/y Lake Cree. McCleIland and Stewart, Toronto.
Stone A. C and M. Stoneking 1993 Ancient DNA from a Pre-Columbian Amerindian Population. American Journal ifPf?ysical Anthropology 92: 463-471.
Stoltman, J. B. 1973 The Laurel Culture in Minnesota. Minnesota Prehistoric Archaeology, Series No 8. Minnesota Historical Society, St. Paul.
Sweetman, P. W. 1955 A Preliminary Report on the Peterborough Petroglyphs. Ontario History. XlVII (3): 101-121.
Syms, E. L. 1977 Cultural Ecology and Ecological Dynamics of the Ceramic Period ln
Southwestern Manitoba. Plains Anthropologist Memoir 12, 22 (76) Part 2:1-160.
1980 The Co-Influence Sphere Model: A New Paradigm of Plains Development and Plains-Parkland Woodland Processual Interrelationships. In Directions in Manitoba Archaeology, edited by L. F. Pettipas, pp. 111-140. Association of Manitoba Archaeologists and the Manitoba Archaeological Society, Winnipeg.
1982 Identifying prehistoric western Algonquians: A Holistic Approach. In Approaches to Aigonquian Archaeology: Proceedings if the Thirteenth Annual Confirence, edited by M. G. Hanna and B. Kooyman, pp. 1-34. The Archaeological Association of the University of Calgary, Calgary.
Szathmary, E. J. E. and R. E. Reed 1972 Caucasian Admixture in Two Ojibwa Indian Communities in Ontario. Human Biology 44 (4): 655-571.
Szathmary, E. J. E., D. W. Cox, H. Gershowitz, D. L. Rucknagel, and M. Schanfield 1974 The Northern and Southern Ojibwa: Serum proteins and red ceIl enzyme systems. American Journal rfPhysicalAnthropology 40: 49-66.
Szathmary, E. J. E., J. F. Mohn, H. Gershowitz, R. M. Lambert, and T.E. Reed 1975 The Northern and Southern Ojibwa: Blood group systems and causes of genetic divergence. Human Biology 47 (3): 351-368.
Szathmary, E. J. E. and F. Auger 1983 Biological Relationships and Genetic Relationships within Algonkians. In Boreal Forest Adaptations: the Northern Aigonkians, edited by A.T. Steegman, pp. 289-315. Plenum Press, New York.
Tanton, T. L. Kenora Sheet, Ontario Map 266A. [map]. 1 inch to 8 miles, Geologïcal Survey and Ontario Department of Mines, 1939.
Page 673
Tassé, G. 1977 a Premieres Reconnaissances. In: Relevés et travaux récents sur l'art rupestre amérindien, edite pars G. Tassé et S. Dewdney. Laboratoire d'archéologie de l'Université du Québec à Montréal, PaléoQuébec 8: 37-69.
1977b Les Peintures Rupestres du Lac Wapizagonke. In: Relevés et travaux récents sur l'art rupestre amérindien, edite pars edited by G. Tassé et S. Dewdney Laboratoire d'archéologie de l'Université du Québec à Montréal, PaléoQuébec 8: 71-112.
1977c Études sur la conservation et la datation. In Relevés et travaux récents sur l'art rupestre amérindien, edite pars G. Tassé et S. Dewdney. Laboratoire d'archéologie de l'Université du Québec à Montréal, PaléoQuébec 8: 113-119.
1979 Newly Recorded Quebec Rock Paintings and their Conservation Problems. Canadian Rock Art Research Association 1977. Papers from the Fourth Biennial Conference of the Canadian Rock Art Research Associates, Heritage Record 8: 147-152.
Taylor,]. M., W. Bokman and 1. N. M. Wainwright 1979 Rock Art Conservation: Sorne realities and practical considerations. Canadian Rock Art Research Association 1977. Papers from the Fourth Biennial Conference of the Canadian Rock Art Research Associates, Heritage Record 8: 293-324.
Taylor,]. M., R. M. Myers, and 1. N. Wainwright 1974 Scientific Studies of Indian Rock Paintings in Canada. Bulletin if the American InstituteforConservation 14 (2): 28-43.
1975 An investigation of the natural deterioration of rock paintings in Canada. In: Conservation in Archaeology and the Applied Arts. Preprints if the Contributions to the Stockholm Congress) June 2-6 1975) pp. 87-91.: The International Institute for Conservation of Historie and Artistic Works: London.
Terasmae, J. 1970 Post-glacial muskeg development in northern Ontario. In: Proceedings in the 1]tb
Muskeg Research Cotifèrence) 7 and 8 Mqy 1970, edited by]. Butler, pp.73-90. Associate Committee on Geotechnical Research, Technical Memorandum No. 99. National Research Council of Canada.
Thwaites, R. G. (Editor) 1896-1901 The Jesuit Relations and Allied Documents: Travels and Explorations if the Jesuit Missionaries in New France) 1610-1791. 73 volumes. Burrows Brothers Cleveland.
Tobenson, M., A. Aufderheide, and E. Johnson 1992 Punctured Human Bones of the Laurel Culture from Smith Mound Four, :tYfinnesota. American Antiquiry. 57 (3): 506-514.
Tobenson, M., O. Langsjoen and A. Aufderheide 1994 Laurel Culture Remains from Smith Mounds Three and Four. Plains Anthropologist 39 (150): 429-444.
Page 674
Torroni A., T. G. Schurr T. G., C. C. Yang, E.J. E. Szathmary, R. C. Williams, M. S. Schanfield, G. A. Troup, W. C. Knowler, D. N. Lawrence, K. W. Weiss and D. C. Wallace
1992 Native American Mitochondrial DNA Analysis Indicates that the Amerind and the Nadene Populations Were Founded By 2 Independent Migrations. Genetics 130: 153-162.
Torroni, A. and D. C. Wallace 1995 MtDNA Haplogroups in Native Americans. American Journal 0/ Human Genetics 56: 1234-1236.
Trazer, N. H. C. Whitifish Bqy Area Lake 0/ the Woods [map].1 Inch to 1 mile. Toronto: Ontario Department of Mines. In: Trazer, N. H. C. Ontario Department of Mines Annual Report. Toronto. 1943.
Trevelyan, A. M. 1989 Continuity of Form and Function in the Art of the Eastern Woodlands. The Canadian Journal o/Native Studies IX (2): 187-203.
Trigger, B. G. 1995 Expanding Middle Range Theory. Antiquiry 69: 449-458.
Turner, B. N. 1979 Rock Paintings in Quetico Provincial Park, Ontario. Newsletter 0/ the Canadian Rock Art Research Associate. Manitoba Chapter 2 (1): 3-4.
Valentine, L. P. 1995 Severn Ojibwe Communicative Practices. University of Toronto Press, Toronto, Buffalo and London.
Vaillancourt. P. 2003. Étude archéologique et ethnohistorique d'un site rupestre (BiGf2) à tracés digitaux au lac Némiscau, territoire cri, Jamésie. Faculté des Lettres, Université Laval.
Vasto kas,] and R. Vasto kas 1973 S acred Art 0/ the Aigonkians: A S tu4J 0/ the Peterborough Petrogjyphs. Mansard Press, Peterborough.
Vastokas,]. 1984 Interpreting Birchbark Scrolls. In Papers 0/ the Fifteenth Aigonquian Coriference, edited by W. Cowan, pp. 425-444. Carlton University, Ottawa.
1986-87 Native Art as Art History: Meaning and Time from Unwritten Sources. Journal o/Canadian Studies 21 (4): 7-36.
Page 675
1996 History without Writing: Pictorial Narratives in Native North America. In Gin Das Winan: Documenting Aboriginal History in Ontario, edited by D. Standen and D. McNab, pp. 48-64. The Champlain Society, Toronto.
Vecsey, C. 1976 Review of the Sacred Scroils of the Southern Ojibwa by Selwyn Dewdney. American Anthropologist 78: 162.
1983 Traditional Ojibwa Religion and ifs Historical Changes. American Philosophical Society, Philadelphia.
1984 Midewiwin Myths of Origin. In Papers if the Fifteenth Algonquian Conference., edited by W. Cowan, 446-467. Carleton University, Ottawa.
Vennum, T. 1978 Ojibwa Origin Midewiwin Songs of the Midewiwin. Journal if American Folklore 91: 753-793.
Vickery,]. D. 1991 Studies in Red Ochre. Ontario Rock Art Research Association Newsletter, December: 22-27. Originaily published in Canoe Magazine February /March 1982.
Voegelin, E. W. 1942 Notes of Ojibwa-Ottawa Pictography. Proceedings if the Indiana Academy if Science 51: 44-57.
Voorhis, P. H. 1978 Some glottochronologic dates for Algonquian and Siouan Peoples. Archae-Facts 5 (4): 3-18.
Wainwright,1. N. M. 1990 Rock Painting and Petroglyph Recording Projects in Canada. APT Bulletin XXII (1-2): 55-81.
1997 Art, Archaeology and the Analytical Research Laboratory of the Canadian Conservation Institute. Techne: la science au service de l'histoire de l'art et des civilisations, 5: 80-88.
Wainwright,1. N. M., H. Sears and S. Michalski 1997 Design of a Rock Art Protective Structure at Petroglyphs Provincial Park, Ontario, Canada. Journal if the Canadian Association for Conservation 22: 53-76.
Wainwright,1. N. M. and]. M. Taylor 1977 On the Occurrence of Parailel Pigment Layer Phenomenon in the Crosssectional Structure of Samples from Two Rock Painting Sites in Canada. In Conservation if Rock Art: Proceedings if the International Workshop on the Conservation if Rock Art, Perth September, 1977, edited by C. Pearson, pp. 29-31. Institute for the Conservation of Cultural Material, Sydney.
Page 676
Wallace, D. C. and A. Torroni 1992 Arnerican lndian Prehistory as Written in the Mitochondrial DNA - A Review. Human Biology 64: 403-416.
Waller, S. 1993 Sound Reflection as an Explanation for the Content and Context of Rock Art. Rock An Research 10 (2): 91-101.
Ward R. H., A. Redd, D. Valencia, B. Frazier and S. Piiiibo 1993 Genetic and Linguistic Differentiation in the Americas. Proceedings 0/ the National Academy 0/ Sciences 0/ the United States 0/ America 90: 10663-10667.
Warkentin, G. 1999 In Search of "The Word of The Other". Aboriginal Sign Systems and the Book in Canada. Book History 2: 1-27.
Warren, W. W. 1984 [1885] History 0/ the Ojibwqy People. First published in 1885 by the Minnesota Historical Society as Volume 5 of the Collections of the Minnesota Historical Society. Minnesota Historical Society Press, St. Paul.
Watson, P.]., S. A. LeBlanc and C. L. Redman 1971 Explanation in Archaeology: An ExplicitlY Scientific Approach. Columbia University Press, New York and London.
Wellmann, K. F. 1979a North American Rock Art: Medical Connotations. New York State Journal 0/ Medicine 79: 1094-1105.
Wheeler, C. n.d. The Oxford House Pictograph or The May May Quah Sao are Alive and Weil at Oxford House. Unpublished paper.
1977 a The Historie Assiniboine: A territorial dispute in the ethnohistoric literature. In Actes du Huitième Congrès des AlgonquinistesJ edited by W. Cowan, pp. 115-123. Carleton University, Ottawa.
1977b The Oxford House Pictograph or The May May Quah Sao are Alive and Weil at Oxford House. In Proceedings 0/ the Second Congress, edited by: J. Freedman and J.H. Barkow, pp. 701-713. Canadian Ethnology Service, National Museum of Man, Mercury Series. National Museums of Canada, Ottawa.
1979 The Oxford House Pictograph. Canadian Rock An Rtsearch Association Newsletter 1 (4):2-9.
Whelan, J. P. 1983 Context and Association in Canadian Shield Rock Art Systematics. Canadian Journal 0/ Archaeology 7(1):77-84.
Page 677
/
Whelan,M. K 1990. Late Woodland Subsistence Systems and Settlement Size in the Mille Lacs Area. In: The Woodland Tradition in the Western Great Lakes. Papm presented to Eldon Johnson, edited by G. E. Gibbon, pp. 82-112. Publications in Anthropology # 4. University of Minnesota. Minneapolis, Minnesota.
Whitley, D. S. 2001 Handbook if Rock Art Research. Altamira Press, Walnut Creek, Lanham, New York and Oxford.
White, A. V. and A. F. Meyer 1916 Report if the International Joint Commission Official Rejèrence to the Lake if the Woods Lake Levels. Government Printing Office, Washington D. C. and Ottawa.
Whiteford, A. H. 1977 Fiber Bags of the Great Lakes Indians. Part 1 American Indan Art Maga;dne 2(3): 52-36.
1986 The Origins of the Great Lakes Beaded Bandolier Bags. American Indian Art Maga;dne 11 (3):32-34.
1991 Mystic and Decorative Art of the Aninishabe (Ojibwa/ Chippewa) Arctic Anthropology 28(1):74-83.
1997 Floral Beadwork of the Western Great Lakes. American Indian Art Maga;dne 22(3):68-79.
Wiessner, P. 1983 Style and social information in Kalahari San projectile points. American Antiquity 48:253-276.
1984 Reconsidering the behavioural basis of style. Journal if Anthropological Archaeology 3:190-253.
1985 Style or isochrestic variation? A reply to Sackett. American Antiquity 50:221-224.
Wilford, L. A. 1941 A tentative classification of the prehistoric cultures of Minnesota. American Antiquity 6(3):231-249.
1955 A revised classification of the Prehistoric cultures of Minnesota. American Antiquity 21(2):130-142.
Williams, G. 1969 Andrew Graham's observations on Hudson's Bcry, 1767-91. Edited by Glyndwr Williams with an introduction by Richard Glover. Hudson's Bay Record Society, London.
Page 678
Winterhalder, B. 1983 History and Ecology of the Boreal Zone in Ontario. In Boreal Forest Adaptations: the Northem Algonquians} edited by A. T. Steegman Jr., pp. 9-54. Plenum Press, New York and London.
Wissler, C. 1919 Indian Beadwork - A Help Jor Students if Design. American Museum of Natural History. Guide Leaflet No. 50. American Museum Press, New York.
Wissler, C. 1931 Indian Beadwork - A Help Jor Students if Design. American Museum of Natural History. Guide Leaflet No, 50. Second Edition. American Museum Press, New York.
Wobst, M. 1977 Stylistic behavior and information exchange. In For the Director; research essqys in honor if] ames B. Grijjin, edited by C. E. Clelland, pp. 317-342. Anthropological Papers 61, Museum of Anthropology, University of Michigan, Ann Arbor.
Wright, H. E. Jr. 1976 The dymanic nature of the Holocene vegetation. Quatemary Research 6:581-596.
Wright, J. V. 1963 An archaeological survry along the north shore if Lake S uperior. Anthropological Paper 3, National Museums of Canada, Ottawa.
1965 A Regional Examination of Ojibwa Culture History Anthropologica VII (2) 189-227.
1967 The Laurel Tradition and the Middle Woodland Period. Bulletin 217. National Museum of Canada, Ottawa.
1968a The Michipicoten Site Ontario. Contributions to Anthropology VI, Bulletin 224, Archaeology. Ottawa: National Museum of Canada. 85 pp.
1968b Cree culture history in the southem Indian Lake region. Contributions to Anthropology 7, Bulletin 232, Archaeology. National Museums of Canada, Ottawa. Pp.: 1-31.
1968c Application of the Direct Historical Approach to the Iroquois and the Ojibwa. Ethnohistory 15(1):96-112.
1972 The Shield Archaic. Publications in Archaeology No. 3. National Museum of Man, Ottawa.
1976 The Grant Lake Site} Keewatin District) N. WT.. Mercury Series Archaeology Paper No. 47 National Museum of Man, Ottawa.
Page 679
1981 Prehistory of the Canadian Shield. In Handbook 0/ the Norlh Amelican Indians, edited by J. Helm, pp. 86-96. Volume 6, Smithsonian Institution, Washington D. C.
1987 Cultural Sequences, AD SOO-European Contact [map] In: R. C. Harris (Ed.) HistolicalAtias rfCanada.From the Beginning to 1800. Volume 1. Toronto, Buffalo, and London: University of Toronto Press, Plate 9.