Logo weiterlesen.de
Light from Ancient Campfires

Light from Ancient Campfires



In memory of Dorothea Catherine Moritz (1905–1992),

my first and finest instructor concerning the value of the past,




for my parents, Donna and Gurdon Peck,

my first and finest instructors.






Scope of the Book

Archaeological Systematics

Theoretical Perspective


Prior to ca. 11,050 BP


The Sites

Pre-Clovis: Still Searching for the Evidence


ca. 11,050 to 8,600 BP

CLOVIS PHASE (CA. 11,050 TO 10,800 BP)

The Sites

Clovis: An Unprecedented “Homecoming”?

FOLSOM PHASE (CA. 10,900 TO 10,200 BP)

The Sites

Folsom: High Mobility with a Focus on Bison


The Sites

Sibbald: The Beginning of Regionalization


The Sites

Agate Basin/Hell Gap: Continuity from Folsom to Scottsbluff?

ALBERTA PHASE (CA. 9,600 TO 9,000 BP)

The Sites

Alberta: Earliest Point Style of the Cody Tradition


The Sites

Scottsbluff-Eden: The First Major Occupation?


ca. 8,600 to 7,500 BP


The Sites

Plains/Mountain: Transition from Spears to Darts in the Foothills/Mountains

LUSK COMPLEX (CA. 8,300 TO 7,500 BP)

The Sites

Lusk: Transition from Spears to Darts on the Plains


ca. 7,500 to 1,500 BP


The Sites

Country Hills: Barbed Darts in the Front Range and Foothills


The Sites

Mummy Cave: The Earliest Side-notched Point Assemblages


The Sites

Maple Leaf: Subsisting in the Foothills and Front Range


The Sites

Calderwood: Social and Material Culture Diversity Five Thousand Years Ago

GOWEN COMPLEX (CA. 5,900 TO 5,200 BP)

The Sites

Gowen: A Distinct Cultural Entity?

ESTEVAN PHASE (CA. 4,900 TO 4,500 BP)

The Sites

Estevan: Oxbow Beginnings on the Alberta Plains

OXBOW PHASE (CA. 4,500 TO 4,100 BP)

The Sites

Oxbow: Cemeteries, Boiling Pits, and More

MCKEAN COMPLEX (CA. 4,200 TO 3,500 BP)

The Sites

McKean: Migrants from the Big Horn Basin/Black Hills Area


The Sites

Pelican Lake: The Twilight Days of Bison Stalking


The sites

Outlook: The First Wave of Middle Missouri Invaders


The Sites

Sandy Creek: Besant Beginnings?

BRACKEN PHASE (CA. 2,800 TO 2,100 BP)

The Sites

Bracken: Industrializing Bison Pounding and Mobilizing a Workforce

BESANT PHASE (CA. 2,100 TO 1,500 BP)

The Sites

Besant: A Renewed Perspective


ca. 1,500 to 1,350 BP

SONOTA PHASE (CA. 1,500 TO 1,350 BP)

The Sites

Sonota: Reviving the Newman-Syms Perspective


ca. 1,350 to 250 BP

AVONLEA PHASE (CA. 1,350 TO 1,100 BP)

The Sites

Avonlea: Migrant Archers from the East


The Sites

Avonlea–Old Women’s Transition: End of Avonlea Within the Beginnings of Old Women’s


The Sites

Old Women’s: Archaeological Evidence for the Prehistoric Blackfoot (Nitsitapii)


The Sites

Highwood: Shoshonean (Snake) Invaders in Southern Alberta


Protohistoric Period, ca. 250 to 200 BP


The Sites

Protohistoric Old Women’s: Continuity and Change


The Sites

One Gun: More Migrants from the Middle Missouri Area





1. Clovis points listed in Eugene Gryba, An Inventory of Fluted Point Occurrences in Alberta (1988)

2. Folsom points listed in Eugene Gryba, An Inventory of Fluted Point Occurrences in Alberta (1988)

3. Radiocarbon dates for Sibbald sites

4. Radiocarbon dates for Agate Basin/Hell Gap sites

5. Radiocarbon dates for Alberta (phase) sites

6. Radiocarbon dates for Scottsbluff-Eden sites

7. Radiocarbon dates for Plains/Mountain sites

8. Radiocarbon dates for Lusk sites

9. Radiocarbon dates for Country Hills sites

10. Radiocarbon dates for Mummy Cave sites

11. Radiocarbon dates for Maple Leaf sites

12. Radiocarbon dates for Calderwood sites

13. Radiocarbon dates for Gowen sites

14. Radiocarbon dates for Estevan sites

15. Radiocarbon dates for Oxbow sites

16. Radiocarbon dates for McKean sites

17. Radiocarbon dates for Pelican Lake sites

18. Radiocarbon dates for Outlook sites

19. Radiocarbon dates for Sandy Creek sites

20. Radiocarbon dates for Bracken sites

21. Radiocarbon dates for Besant sites

22. Radiocarbon dates for Sonota sites

23. Radiocarbon dates for Avonlea sites

24. Radiocarbon dates for Avonlea–Old Women’s transition sites

25. Radiocarbon dates for Old Women’s sites

26. Radiocarbon dates for Protohistoric Old Women’s sites


1. Culture-historical models: Proposed model (left column) and current model (right column)

2. Pre-Clovis sites within Alberta

3. Clovis and Goshen sites within Alberta

4. Folsom sites within Alberta

5. Sibbald sites within Alberta

6. Agate Basin sites within Alberta

7. Alberta (phase) sites within Alberta

8. Scottsbluff-Eden sites within Alberta

9. Plains/Mountain sites within Alberta

10. Lusk sites within Alberta

11. Country Hills sites within Alberta

12. Mummy Cave sites within Alberta

13. Maple Leaf sites within Alberta

14. Calderwood sites within Alberta

15. Gowen sites within Alberta

16. Estevan sites within Alberta

17. Oxbow sites within Alberta

18. McKean sites within Alberta

19. Pelican Lake sites within Alberta

20. Outlook sites within Alberta

21. Sandy Creek sites within Alberta

22. Bracken sites within Alberta

23. Besant sites within Alberta

24. Sonota sites within Alberta

25. Avonlea sites within Alberta

26. Avonlea–Old Women’s transition sites within Alberta

27. Old Women’s sites within Alberta

28. Highwood sites within Alberta

29. Protohistoric Old Women’s sites within Alberta

30. One Gun sites within Alberta


1. Flaked tools from Varsity Estates site

2. Clovis points

3. Folsom points

4. Basally thinned points

5. Agate Basin/Hell Gap points

6. Alberta (phase) points

7. Scottsbluff points

8. Plains/Mountains points

9. Lusk points

10. Burmis barbed points

11. Mummy Cave points

12. Maple Leaf points

13. Calderwood points

14. Gowen points

15. Estevan points

16. Oxbow points

17. McKean points

18. Pelican Lake points

19. Outlook points

20. Sandy Creek points

21. Bracken points

22. Besant points

23. Sonota points

24. Avonlea points

25. Avonlea–Cayley Series transition points

26. Cayley Series points

27. Highwood points

28. Protohistoric Old Women’s points

29. One Gun points


I owe a particular debt of gratitude to my supervisors at the Heritage Division, Culture and Community Spirit. Dr. David Link, Assistant Deputy Minister (Acting), and Brian Ronaghan, Director of the Archaeological Survey, were exceedingly supportive of the project, allowing exceptional access to the Cultural Resource Management literature. As well, constructive critical comment was provided by Wendy Unfreed, Plains Archaeologist, Archaeological Survey, throughout the project.

A handful of people reviewed the document at various stages of completion. The sheer size of the volume would have frightened off most scholars. I am deeply indebted for the constructive criticism provided by four anonymous reviewers as well as by Caroline Hudecek-Cuffe, Brian Vivian, Alison Landals, and, again, Wendy Unfreed. The work was enhanced by the many, many, many suggested improvements.

Numerous colleagues generously shared copies of their theses and unpublished papers, including Brian (Barney) Reeves, Laureen Bryant, Charles (Chuck) Ramsay, Sean Webster, Alison Landals, Brian Scribe, Jennifer Tischer, Barbara Neal, Brian Kooyman, Allyson Ramsay, Benjamin Hjermstad, Grant Clarke, Karl Hutchings, Gerald Oetelaar, Jason Gillespie, Steven Kasstan, Irene (Rena) Varsakis, Mack Shortt, David Norris, Brad Himour, Dale Walde, Michael Klassen, Jean Prentice, Leslie (Butch) Amundson, Suzanne Zurburg, Riel Cloutier, Margaret Hanna, Thomas Head, Mary Malainey, and Caroline Hudecek-Cuffe.

The Royal Alberta Museum deserves special thanks for their support of this work. Jack Brink, Bob Dawe, Kristine Fedyniak, and Karen Geiring repeatedly obliged requests for access to collections. Similarly, Parks Canada, the Department of Archaeology at the University of Calgary, and Lifeways of Canada Ltd. also allowed access to collections for photography.

Permission to reproduce a number of photographs was granted by the following individuals and firms: Dale Walde (University of Calgary), Gerald Oetelaar (University of Calgary), Eugene Gryba, Lifeways of Canada Ltd., Laurie Milne, Jason Gillespie, John Brumley, Darryl Bereziuk, Michael Quigg, FMA Heritage Resources Consultants Inc., Bison Historical Services Ltd., Glenbow Museum, and ARESCO. Photographs taken on contract to the Archaeological Survey were skilfully produced by Kim Tymko, Kathy McCormack, and Jason Gillespie. Robin Woywitka plotted the sites on the maps.

Finally, I thank my family for their support for this project. My parents, Donna and Gurdon, have been supportive of my career in archaeology for over twenty years. Their enthusiasm for this project was no different despite its ambitious nature. My brother Chris and his family have provided verbal support from their home in Australia. My wife, Beth, expertly drafted all the plates with projectile points and her professional editing skills polished the draft document in its final stages. My appreciation cannot be expressed in words.


People have lived in what is now southern Alberta for about the past 13,000 years. Archaeological evidence for Native lifeways in Alberta has been documented by ordinary citizens who find artifacts on their land, by academics conducting research at universities, and by local archaeological societies who rally to map and preserve sites. The greatest source of archaeological information, however, is Alberta’s cultural resource management (CRM) community, which includes the Alberta Historic Resources Management Branch and the archaeological consultants. In accordance with the Alberta Historical Resources Act (1973), the Alberta Historic Resources Management Branch regulates all archaeological matters in the province. Among other things, the act influences when archaeological studies are required as part of land development projects. Consulting archaeologists are contracted by land developers to mitigate effects of development projects on Alberta’s archaeological resources. The joint efforts of all these groups — concerned citizens, avocational archaeologists, academics, archaeological consultants and government regulators — has led to the identification of 35,500 sites in Alberta. This wealth of information provides the basis for this book.

Scope of the Book

The aim of this text is to provide a synthesis and revision of the culture-historical sequences used to organize archaeological evidence of Alberta plains prehistory (see Figure 1). To accomplish this aim, the sequence and its cultural units (traditions, complexes, and phases) are examined in a three-fold process. First, background research previously conducted for each cultural unit is summarized. Second, the vast amount of data generated by the cultural resource management community as well as other relevant sources is presented as raw data for each cultural unit. Third, a new definition of the cultural unit is offered with a synthesis of its material culture, technology, social organization and spiritual practices. The defined cultural unit is compared to similar entities in nearby Saskatchewan, Manitoba, Montana, North Dakota, South Dakota and Wyoming to determine if a broader distribution of each archaeological culture can be recognized.

The presentation of so much raw data provides opportunity for readers to interpret the data for themselves rather than simply accept the author’s perspective. This is especially true given the restrained interpretation of the voluminous raw data presented in the text. Thus, the volume ends with a commentary rather than a summary to loosen the restraints on interpretation with the aim of further eliciting discussion within the discipline.

In recent years, archaeological excavations and studies in cultural resource management have dramatically increased. Many sites have been reported to the Historic Resources Management Branch in detail but the information is not widely available to the academic community or public. The first thorough synthesis of Alberta’s plains prehistory was done by Reeves (1969, 1983a), with more recent critiques of the model by Vickers (1986), Brumley and Dau (1988:20–77), and Brumley and Rennie (2005). A reassessment of the prehistoric record of archaeological material on the Alberta plains is long overdue.

While the production of a new sequence is important and much attention has been paid to timing of events, the relationships between the various cultural entities that comprise the sequence still require elucidation. As stated, this text’s focus is provincial, with heavy emphasis on the material culture recorded in Alberta. Additional effort has been made to pursue cultural entities into adjacent regions of the Northwestern Plains in order to provide a broader context for interpreting Alberta’s past.

The text is mainly organized chronologically, based on diagnostic projectile points that define archaeological phases and complexes. With a few exceptions, only sites with diagnostic projectile points and associated radiocarbon dates have been included in this review. Some sites with dubious associations have been included but the issues associated with these sites are included. A few sites lacking radiocarbon dates but containing diagnostic projectile points have been included, owing to their comparability to known-age specimens. As is demonstrated throughout the text, the morphology of a projectile point can be misleading with respect to its age. The best way to evaluate a site is through recovery of diagnostic materials in a well-dated context.

Culture-historical models: Proposed model (left column) and current model (right column)



Chronological organization of the culture-historical sequence starts with the earliest archaeological evidence for people in Alberta and proceeds to the present. As mentioned above, each part of the culture-historical sequence (i.e., phases or complexes) is discussed separately, addressing three distinct elements. First, the background research that has been previously conducted for each cultural unit is presented. Questions such as What defines the cultural entity? Has our notion of it changed with increased information? and What is the current state of thought concerning this issue? are presented in the introductory passages. Any interpretation of the archaeological record makes more sense couched in the context from which that understanding stemmed. Second, the vast amount of data generated by the cultural resource management community is rarely published in readily accessible sources; this work usually presents raw data for site type, location, amount of excavation conducted, artifacts collected, and radiocarbon dates, site by site. Lastly, a new or reviewed definition of the phase or complex is offered. A generalized synthesis of the culture’s place in the past with respect to material culture, technology, social organization, and spiritual practices is presented. The defined cultural units are compared with similar material in nearby Saskatchewan, Manitoba, Montana, North Dakota, South Dakota, and Wyoming, and a broader distribution of each is suggested where continuity in material culture with adjacent regions is demonstrated.

Archaeological Systematics

In order to organize the great amount of archaeological evidence that has been gathered from the province, a consistent classification scheme is required. A number of different systems have been proposed for archaeological assemblages, with most relying on the nature of morphological change in projectile points over time. This approach is possible owing to the largely diachronic technological change in projectile points (i.e., spear to dart to arrow) known to have taken place across North America, coupled with more localized morphological variation during each technological stage. Such morphological change in the projectile points has been attributed to many phenomena, including function and stochastic variation, but ethnicity dominates most interpretations of Northern Plains cultures (e.g., Dyck 1983:132; Greaves 1982; Kehoe 1966b:839–840; Peck 1996:117–136).

Traditionally, the archaeological record in Alberta, and on the Northwestern Plains in general, has been divided into segments of time called periods. Each period is further subdivided into smaller segments of time called complexes, traditions, or phases (sometimes with subphases). As alluded to above, the periods have been differentiated based on changes in projectile point morphology, changes which have been inferred to reflect changes in hunting systems technology (e.g., Dyck 1983; Forbis 1970; Mulloy 1958:204–223; Peck 2004:2–3; Reeves 1969, 1983a:35–37; Vickers 1986:9–16; Wormington and Forbis 1965:183–198).

Mulloy (1958:204–223) proposed the first organizational scheme for archaeological materials from the Northwestern Plains. He considered archaeological material from his Early Prehistoric period, ca. 11,000–4,000 B.C., to be amongst the earliest produced by people in North America (Mulloy 1958:204–205). Interestingly, he suggested that the evidence from this time period indicates that large lanceolate points were used with spear-throwers rather than thrusting spears (Mulloy 1958:205, 208). He proposed a hiatus, ca. 4,000–1,500 B.C., between the Early Prehistoric period and the Early Middle Prehistoric period, owing possibly to climatic conditions. The Early Middle Prehistoric period, ca. 1,500 B.C.–A.D. 1, is represented by small lanceolate points and stemmed points (Mulloy 1958:209). The Late Middle Prehistoric period, (A.D. 1–500, produced corner-notched points (Mulloy 1958:209). In the Late Prehistoric period, A.D. 500–1800, small corner-notched points occurred, but small side-notched points are considered the best diagnostic along with ceramics (Mulloy 1958:211, 213). The Historic period, A.D. 1800 to the present day, differed from the previous period, as evidenced by the inclusion of European trade items in the archaeological record.

Wormington and Forbis (1965) developed the first systematic classification of archaeological material designed specifically for assemblages in Alberta. They rejected Mulloy’s system because it leads to confusing terminology such as “Late Early Prehistoric” and “Early Late Prehistoric” (Wormington and Forbis 1965:13). They did recognize that the three technological-cultural stages found in the United States are found in Alberta, but instead proposed the terms Palaeo-Indian, Meso-Indian, Neo-Indian, and Historic (Wormington and Forbis 1965:13). During meetings in 1966, Forbis (1968b:44) acknowledged that “the historical outline placed before us several years ago by Mulloy is by and large still valid for Alberta. I do not imagine that we will seriously alter his general scheme. What we can do is fill in some extremely interesting details.” In another proposed scheme, based on data attained in Alberta up to 1964, Forbis (1970) proposed slightly different terminology for the technological-cultural periods already established. He suggested “Early Lithic period,” “Middle Lithic period,” and “Late Lithic Period,” with the possibility of a pre-projectile point horizon at the beginning of the sequence (Forbis 1970).

It was not until the late 1960s that Mulloy’s model was accepted as the basis for the modern version of the classification scheme. Brian Reeves stated: “Following Mulloy (1958), the Post-Glacial cultural sequence may be divided into three periods: Early, Middle and Late. The following sequence is divided either into complexes when the relationship is unclear between the sequent assemblages or into phases when the relationships are discernible between the serial assemblages. The latter are linked by cultural traditions” (Reeves 1983a:19). Reeves (1969) presented the essence of the current classification with the Early Prehistoric period consisting of the Clovis, Folsom-Midland, Agate Basin/Hell Gap, Alberta-Cody, Lusk, and Frederick complexes. His Middle Prehistoric period included Mummy Cave, Oxbow, McKean-Duncan, Hanna, and Pelican Lake phases of the Tunaxa cultural tradition, and the Besant phase of the Napikwan cultural tradition. Reeves’ Late Prehistoric period included the Avonlea phase of the Tunaxa cultural tradition and the Old Women’s phase of the Napikwan cultural tradition. While these cultural traditions are not often recognized, and some of the complexes are now called phases, this arrangement strongly agrees with the modern classification scheme. Following Willey and Phillips (1958:72), Reeves (1983a:39) defined a phase as an archaeological unit that exhibits traits distinct enough to discern it from other units. A phase does not necessarily correlate to a locality or region; it can change through time, and may be found in two or more environmentally distinct regions. A phase must also demonstrate a discernable relationship between serial assemblages. When the relationship is unclear between the sequential assemblages, it is called a complex (Reeves 1969:19). If a sequence of phases is established based on discernable relationships between serial assemblages it is called a tradition (Reeves 1983a:40). In short, Reeves’ cultural traditions articulated phases that were inferred to be related along a space-time continuum and are labelled with a specific cultural tradition title, such as Tunaxa cultural tradition.

Of course, projectile points have been used as the principle diagnostic criteria for each of the phases and complexes on the Alberta plains. Almost always, the phase or complex is named for its diagnostic projectile point type. The traditions have been labelled based on inferred ethnic identification: for example the Napikwan cultural tradition being an archaeological unit ultimately affiliated with the Blackfoot, and the Tunaxa cultural tradition reflecting the archaeological past of the Kutenia.

In Alberta, it seems fair to say that most archaeologists recognize the terms phase, complex, and tradition as defined here. The same cannot be said for the rest of the Northwestern Plains. For example, Dyck (1983:69) and Syms (1977: 70–72) both discussed the Northern Plains in their syntheses but used different terms for archaeological entities than those used in Alberta. The diversity of meanings for the various terms can cause confusion when interregional discussions occur. The definitions delimited above will be retained for the purposes of this text.

Recently, there has been a trend to replace the term Prehistoric period with the term Precontact period in an attempt to use culturally sensitive language. This sentiment is misplaced. Precontact embeds the Eurocentric notion that contact with Europeans was the only significant contact event Native people have experienced (Jones 1997:64–65). This is not true, of course, as Native trade networks were known to have flourished and collapsed in many places at many times, bringing a variety of Native peoples into contact. In contrast, prehistory is used in archaeology to mean “prior to the keeping of written records,” a definition that can be found in most dictionaries. The term has been used appropriately for a long time in Northern Plains archaeology, in no way disparagingly, and it is correctly used here.

Other concerns over terms used in Northern Plains classification have been raised by Yellowhorn (2003). He proposed that the Northern Plains match terminology used in Europe’s Paleolithic period. This proposal elicited a reasonable rebuttal that the current classification system has solid historical precedent and demonstrated utility (Gillespie 2003). Again, for purposes of continuity with the Alberta literature, the terms and meanings currently in use, as described above, are retained.

Theoretical Perspective

This text explicitly pursues a culture-historical paradigm. For Alberta, culture-historical models were first sketched by Wormington and Forbis (1965) in the mid-1960s and solidified by Reeves (1969, 1983a) in the late 1960s, with updated versions in recent years (Reeves et al. 2000; Reeves 2003). This model has benefited from a number of other individuals who have slotted new sites into the appropriate archaeological cultures, such as Brumley and Dau (1988), Peck and Hudecek-Cuffe (2003), and Vickers (1986, 1994). Lyman and O’Brien (2006) revised the history and development of culture historical models by providing a novel understanding of their fundamental underpinnings. This text follows their groundbreaking effort.

Culture-historical perspectives are often viewed as being mainly interested in the chronology of archaeological phenomenon: time-space grids (Lyman and O’Brien 2006:6). Lyman and O’Brien (2006) argued that the culture-historical paradigm is evolutionary in its underlying principles. One of the oldest artifact-based chronometers used to place assemblages in order within the cultural-historic framework is the direct historical approach (Lyman and O’Brien 2006:12). The direct historical approach relies on “overlapping” to provide the measurement of time as a continuous variable between two points: “The principle of overlapping … concerns the occurrence of a cultural trait or artifact type in multiple cultural complexes or in artifact assemblages potentially of different age, and it is these shared, or overlapping, traits or types that serve as the basis for placing those complexes or assemblages adjacent to another in an ordering thought to comprise a sequence” (Lyman and O’Brien 2006:104). The assumption implicit in the direct historical approach is that the more traits shared between an historic assemblage and a protohistoric or prehistoric assemblage, the closer they are in time. “The concept of overlapping is critically important because it reveals the underpinning (and typically implicit) view of culture change as an evolutionary process minimally involving cultural transmission and inheritance, or what during the first half of the twentieth century was referred to commonsensically as persistence (Rouse 1939) or tradition (Willey 1945)” (Lyman and O’Brien 2006). Heritable/historical continuity through genetics, or cultural transmission between people, is what biologists call lineages and archaeologists call traditions. Thus temporally sequent phenomena share at least some kind of affinity. Transmission is not always precise despite cultural constraints ensuring a relatively high degree of fidelity; “descent with modification” can occur as a result of cultural innovation (i.e., “mutations”) and problems in the fidelity of replication (Lyman and O’Brien 2006:22–23). A caveat to this paradigm is that the use of assemblages with the direct historical approach, or even assemblages excavated from stratigraphic sequences, should not be considered real or discrete entities, but merely chunks of the cultural continuum: there is no steady state within a culture.

Descent with modification relies on the principle of heritable continuity, which requires fidelity of replication between the “ancestor” and the “descendent.” Transmission, along with its innovations and errors, and natural selection together ensure that change is constantly occurring. The direction of change cannot be determined because it is historically contingent. “What is available for transmission depends on the random — with respects to what is or might be needed among descendents — generation of innovative variants; what actually is transmitted depends on the transmission mechanisms and their operation; what is replicated depends on the size of the transmitting population and the particular sorting filters in operation at the time of transmission” (Lyman and O’Brien 2006:169).

To provide a practical application of this paradigm, I address the problem of heterogeneity and homogeneity in material culture. A major “sorting filter” affecting transmission in human life is kinship, a social structure through which much of “the way to make things” is transmitted. In a general sense, generic Plains social structure or kinship is generational or horizontal; that is, it exhibits little time depth with an extended family often centred on a sibling group. In the past, band organization was usually composite, consisting of extended family groups centered on a chief and his close relatives and followers. The composition of a band changed with the circumstances of politics and economics. Leadership was established through status and rarely through heredity. People were free to leave and seek other bands and other leaders. This configuration of social structure allowed for fluidity (e.g., Eggan 1955a, 1955b). In contrast, many groups coming on to the plains in the Historic period, especially from the east, did not exhibit this social structure (Eggan 1955b). Those exhibiting cross-cousin marriage, for example, created multiple bonds between a limited array of relatives and maintained those bonds from generation to generation. This is different than a generational kinship system that ties non-relatives together in a horizontally expanding system.

With regards to the production of material culture, the fluid nature of groups with generational systems brings together people with highly varied backgrounds in terms of heritable continuity. That is to say, transmission from parents to their children may have been produced with high fidelity, but others who have joined the group are less likely to fit within the “lineage” owing to historical contingency. So, material culture will look generally similar within the cultural group, with an overall appearance of heterogeneity. In contrast, for some of the groups coming on to the Plains, with cross-cousin marriage systems for example, transmission with high fidelity should occur, increasing the likelihood of heritable continuity. In this scenario, fewer outside social forces influence the transmission mechanisms, further promoting conformity. In short, transmission within groups with constrained filtering mechanisms (e.g., cross-cousin marriage, craft specialization, sodalities) should produce items of material culture that exhibit a strong similarity owing to the fidelity of transmission.

The following provides an example of this theoretical position from Alberta’s archaeological record. In the Late Prehistoric period, the arrival of the Avonlea phase and its characteristic projectile point, the Avonlea point, is striking. Avonlea pottery is also limited in its diversity, exhibiting parallel-grooved, fabric/net-impressed, and plain surface finishes. The regularity and homogeneity of Avonlea points and pottery is well known in Northern Plains archaeology. Interestingly, the most recent research suggests the Avonlea phase has its origin on the eastern periphery of the Plains. The current model would suggest it exhibited a structured kinship system that generated bonds through generations, thus passing down the “way to make things.” The Old Women’s phase has produced the most heterogeneous diagnostic projectile points and pottery in Alberta’s archaeological record. Importantly, there is strong evidence to link the end of the Avonlea phase to the beginning of the Old Women’s phase. As well, the Old Women’s phase has been linked with the ethnologically known Blackfoot through a series of convincing arguments. The Blackfoot are a classic band-oriented society. Using the model, the Avonlea phase, a group of people on the eastern periphery of the Plains, migrated across the Plains to southern Alberta, southwestern Saskatchewan, and northern Montana. There, their rigid social structure was slowly reconsidered in the face of the harsh Plains environment in favour of a more fluid band structure. Through this process, the rigid social structure that produced homogeneous Avonlea points and pottery was replaced with a more fluid structure. The result was the Old Women’ phase, which exhibits heterogeneous point and pottery forms. The correlation of Old Women’s phase to Blackfoot culture strongly hints at a classic band organization for the archaeological culture.

The renewed culture-historical paradigm has many profound implications for Alberta’s culture-historical sequence. First, the entire cultural sequence should be revisited given the evolutionary nature of the cultural-historic models. It should be kept in mind that cultural transmission and descent with modification are the underlying principles of this perspective. Second, the sequence itself need not change but the idea of the static “phase” should now be re-interpreted as points within a continuum of culture change. Third, evidence of “overlapping” should be sought out as it indicates culture change as an evolutionary process minimally involving cultural transmission and inheritance. Lastly, social structures such as kinship should be examined as possible filtering mechanisms for transmission. As the example above suggests, it seems possible that generational kinship structures, more varied in their historical contingencies, will transmit/inherit more-varied transmissions. In contrast, cross-cousin kinship structures, very cohesive in their historical contingencies, will transmit/inherit more-succinct transmissions.

1 Earliest Inhabitants
PRIOR TO CA. 11,050 BP

The glacial maximum in Alberta occurred about 20,000 BP (before present, where present = A.D. 1950). Ice covered most of the province with the exception of areas in and around the Cypress Hills and a few isolated locations along the foothills of the Rocky Mountains. At this time the Cordilleran, or Montane, ice coalesced with the Laurentide ice sheet in the area of the foothills. There is some debate whether deglaciation began shortly after this as a slow process lasting until 12,000 BP or whether coalescence lasted until about 12,000 BP with a rapid deglaciation. Regardless, meltwater played an important role in shaping the Alberta landscape.

People living in such an environment would have experienced a relatively unstable and rapidly changing periglacial world. Archaeological evidence of these people would likely be quickly destroyed or buried in unusual localities and/or deep sediments (Beaudoin and Oetelaar 2003:187).


Most scholars agree that the peopling of the Americas took place via the northern latitudes, across the Bering Land Bridge or along its shores. Despite the numerous lines of evidence that support this perspective, a precise route and arrival time of people to the Americas has yet to be documented. Alberta is intimately intertwined in this debate since one possible entry route would have been an inland penetration along the front range of the Rocky Mountains; alternatively, a coastal route would have taken people along the shores of British Columbia.

The Coastal entry route would require access to now submerged palaeoshorelines along the west coast of Alaska and British Columbia. These shorelines began to submerge as ice on land melted, returning water to the oceans and causing sea levels to rise. There is evidence arguing against the coastal route. First, the earliest known sites in the coastal region date between ca. 10,500 to 10,000 BP, consisting of lanceolate points and lithic assemblages, and located between Alaska and Vancouver Island (e.g., Carlson and Magne 2008). Second, no obvious predecessor in western Beringia has been found with a marine-based culture of an appropriate age (West 1996). And third, there does not appear to be any cultural descendants of a coastal-adapted culture. The Clovis complex has possible antecedents in the technology and hunting adaptation of the Nenana complex or other large-mammal hunting people of the interior of Beringia or Alaska-Yukon. There does not appear to be any evidence of an early populating entity exhibiting any signs of a remnant coastal background. Still, the debate over the entry route taken by the first people has become progressively interesting with the increased acceptance of the Monte Verde site in south-central Chile. The Monte Verde site, dated to 12,500 BP, may provide evidence for the initial occupation of the Americas. The site consists of huts with log foundations, clay-lined pits for cooking, food remains including organics (e.g., seeds, nuts, fruits, berries, and tubers), and evidence of mastodon. Eleven consistent radiocarbon dates average about 12,500 BP (Adovasio and Pedler 1997:578). More astonishing is an older layer with chipped stone and possible fire pits dated to 33,000 BP. Many authorities accept the authenticity of Monte Verde (e.g., Adovasio and Pedler 1997; Fiedel 2000:85) while others do not (e.g., West 1996:540).

While the acceptance of Monte Verde as the oldest New World human occupation (i.e., pre-Clovis) would have profound implications for archaeology in the Americas, three important points must be acknowledged. First, internally consistent radiocarbon sequences in apparent firm association with actual artifacts can be a product of contamination. Second, regardless of Monte Verde’s status, it has been thirty years since the discovery of the site but few if any comparable sites have come to light. When the first Folsom fluted point recoveries were made in the late 1920s, tens of additional similar finds were recorded within a decade of the original find. Thus, the site of Monte Verde may represent the initial occupation of the Americas, but it likely does not represent evidence for a peopling event. Thus, the view of Clovis technology spreading between existing groups of people (e.g. Stanford 1978b; Waters and Stafford 2007) seems improbable since there is no evidence to indicate that a substantial pre-Clovis population existed. Third, recently there have been acknowledged differences between the structural form of the radiocarbon calibration curves from the northern and southern hemispheres (McCormac et al. 2004; Landals 2008). Dates in the southern hemisphere, such as those from Monte Verde, have the potential to be incorrect, as the southern hemisphere needs to develop dendrochronologically dated wood and radiocarbon calibration curves independent of the northern hemisphere (McCormac et al. 2004; Landals 2008).

Alternatively, to enter the Americas via the inland route across the Bering Land Bridge would require the Ice-free Corridor to connect the northern land bridge or Beringia to the lands south of the mountain and continental ice sheets. The Ice-free Corridor consisted of deglaciated land along the eastern front range of the Cordilleran mountains in the Yukon, British Columbia, and Alberta. It is expected that the earliest known archaeological sites south of the ice sheets should coincide with the timing of the opening of the Ice-free Corridor, as the Cordilleran and Laurentide ice sheets separated. At present, the route is inferred to have opened by ca. 12,000 BP (e.g., Osborn et al. 2000).

The Clovis complex consists of the oldest sites of undisputed authenticity in the Americas south of the ice sheets. There are, however, sites that are as old as, if not older than, Clovis sites in the Americas of undisputed authenticity. These sites are all located north of the former North American ice sheets, in the Yukon and Alaska. Bluefish Caves is located 54 km southwest of the Old Crow settlement in the northwestern Yukon Territory. Bluefish Cave I is a multicomponent site with cultural material in levels III through V. Late Pleistocene and early Holocene-age fauna such as bison and horse were recovered from these levels. The lowest clear cultural level, level V, produced three flakes and thirty-seven micro-chips in a context expected to date about 13,000 BP (Cinq-Mars 1979:28). A microblade may occur in the overlying level and micro-chips may occur in underlying levels (Cinq-Mars 1979:28). The assemblages contained so few artifacts, however, that it is difficult to compare the material to other assemblages.

Perhaps more informative are the sites of the Nenana complex in the Alaska Range in central Alaska. Within the Nenana Valley is a series of sites — Dry Creek 1 (11,120 +/- 85 BP), Owl Ridge (11,340 +/- 150 BP), Moose Creek (11,730 +/- 250 BP), and Walker Road (11,010 +/- 230 BP; 11,170 +/- 180 BP; 11,300 +/- 120 BP; and 11,820 +/- 200 BP) — that are coeval with or predate Clovis sites (Hoffecker, Powers, and Bigelow 1996; Hoffecker, Powers, and Phippen 1996; Goebel et al. 1996; Hoffecker 1996). Goebel, Powers, and Bigelow (1991:74) noted that Nenana toolkits are virtually identical to Clovis tools, except the Nenana toolkits lack lanceolate projectile points. They suggest Nenana and Clovis may represent the northern and southern remnants of a peopling event, respectively, but they remain open to whether the peopling event was a late or early entry. A late entry about 13,000–12,000 BP would have occurred as the ice melted and allowed dispersal down the Ice-free Corridor. A much earlier entry could have had people into the Americas about 22,000–25,000 BP, before the Ice-free Corridor closed (Goebel et al. 1991:75). Recent excavations at the Yana RHS site, Siberia, have produced bifacial lithic industries and bone foreshafts that are similar to Clovis materials, dating to 27,000 BP (Pitulko et al. 2004). The recovery of materials dating to such an early time period so far north in Siberia verifies northern-adapted people were positioned for an early or late dispersal into the Americas.

Importantly, there are sites in Alaska that predate the Nenana complex. Materials from the Broken Mammoth, Swan Point, and Mead sites date to 12,000–11,500 BP. The lowest level at Swan Point produced dihedral burins, microblades, and a core tablet, and may be related to the Dyuktai culture of northeast Asia. While the context of the microblades has been questioned, Hoffecker (2001:149) noted that microblades had also been recovered from Bluefish Cave. Hoffecker (2001:149) suggested there was an early Beringian Dyuktai industry in Alaska no earlier than 11,500 BP. It was succeeded by the Nenana complex ca. 11,500–10,800 BP, which would provide a plausible source for Palaeoindian complexes entering the remainder of the Americas via the Ice-free Corridor (Hoffecker 2001:150).

The Sites

Because of the potential peopling route through the Ice-free Corridor, Alberta has enjoyed being in the spotlight for peopling studies (Beaudoin et al. 1996; Carlson 1991; Haynes 1987). A number of Alberta sites have become well known as potential early peopling sites (see Plate 1 and Figure 2).

PLATE 1 Flaked tools from Varsity Estates site. Photo credit: Jason Gillespie.


Taber Child (DlPa 4). The Taber Child site consists of fragments of an immature human skeleton that were found eroded from a steep coulee wall above the Oldman River just north of Taber. The site was discovered during a Geological Survey of Canada study conducted by Dr. A.M. Stalker in 1961. Wormington and Forbis (1965:117) and Wilson, Harvey, and Forbis (1983:180), amongst others, have called it the Stalker site and noted common confusion with the Bayrock site, which has been referred to as the Taber Early Man site (Bayrock is an Alberta phase kill site; see below).

The Taber Child skeleton is mainly represented by skull fragments, two vertebrae, a clavicle and part of a tibia (Moffat and Wainwright 1983:224). The child was estimated to be between nine months and four years of age at the time of death (Moffat and Wainwright 1983:224). Wormington and Forbis (1965:117) were the first to publicly propose great antiquity for the site based on the assessment of its geological position beneath till (Wilson et al. 1983:179).

Pre-Clovis sites within Alberta


In 1977, excavations at the site set out to provide stronger evidence for the age of the skeletal material (Wilson et al. 1983:185). Evidence of Holocene mudflow deposits was observed near the Taber Child. This discovery suggested that the Taber Child bones had been recovered from orange sand deposits that were determined to date as old as 20,000 BP. X-ray diffraction analysis of matrix from the Taber Child specimen did not show evidence of the orange sand. The matrix most resembled the Holocene slope deposits when compared to baseline samples from across the slope (Wilson et al. 1983:199). The authors concluded that the skeletal material likely derived from mudflow deposits inset within the orange sand unit (Wilson et al. 1983:203–204). An infrared spectrophotometry analysis was undertaken to assess the loss of collagen. The protein content suggested that the Taber Child, in comparison with reference specimens, was consistent with those 10,000 BP or less (Moffat and Wainwright 1983:229). Similarly, accelerator 14C dating of the Taber Child produced an age of 3,550 +/- 500 BP (Brown et al. 1983).

Varsity Estates (EgPn 413) and Silver Springs (EgPn 414). The Varsity Estates and Silver Spring sites are located on the north side of the Bow River on a roughly 50-metre-long escarpment of the valley in northwest Calgary. The Varsity Estates site consists of lithic artifacts eroding out of the gravel upper part of a layer of till that immediately underlies Glacial Lake Calgary sediments (Chlachula 1994a:104). The Silver Springs site consists of redeposited lithic artifacts in the upper part of some fluvial gravels of the Bow Valley till (Chlachula 1994a:104). The steep slope prohibited a large excavation but 15 m2 was excavated in 1992–1993 at the Varsity Estates site and a series of 0.5-×-1.0-m units were excavated at the Silver Springs site (Chlachula 1996:298–299).

The assemblage from the Varsity Estates sites consisted of an upper and lower series. The lower series, presumably derived from an older cultural context, produced sixteen flakes “removed from the original laterally exposed face of the Bow Valley till (Unit 1) over a distance of 70–100 m in the steepest middle part of the slope” (Chlachula 1996:298). The second, or upper, series consisted of forty artifacts including cores (n = 2), side scrapers (n = 2), an end scraper, choppers (n = 2), a biface, and pieces of debitage (n = 17, with five flakes from the biface tool) recovered in roughly a 15-m2 area on the till surface immediately beneath the lake (Chlachula 1994a:105, 1996:298). The Silver Springs site material (n = 31) was recovered from an eroded context. Eighteen artifacts were recovered from the till (Chlachula 2006:300–301). In short, a simple core and flake industry is said to be represented. A pollen sample from the site suggests sedge and pine were in the area (Chlachula 1994a:125). No datable macro-organic material was recovered from the site. The material was interpreted as a pre-Palaeoindian occupation in the foothills of Alberta that dates between roughly 25,000 and 21,000 BP based on chronostratigraphic correlation of the culture-bearing deposits to existing Late Quaternary temporal frameworks (Chlachula 1994a:126, 1994b, 1996:306).

Bryan and Gruhn (2007) argued that only four items recovered from the till at the Varsity Estates site were artifacts, but that the artifacts’ strati-graphic position suggest a terminal Pleistocene age rather than the previously hypothesised pre-Late Wisconsinan age. The authors suggested that the artifacts were left on the surface of ice still in the area and later incorporated into the till. They argued that quarrying activity would have taken place at the site, not habitation (Bryan and Gruhn 2007:98).

Grimshaw. The Grimshaw site is located in the Grimshaw Gravel Pit, in the middle Peace River area, in gently rolling terrain about 5 km north of the town of Grimshaw in northwestern Alberta. The site consists of quartzite cobble artifacts recovered from the base of a Laurentide till exposed in a gravel operation (Chlachula and Leslie 2001).

Seventeen cobble artifacts were recovered: two observed in the lab, twelve in situ in the field, and three eroded from an exposed face (Chlachula and Leslie 2001:873). The tools include unifacial choppers (n = 7), bifacial choppers (n = 2), side scrapers (n = 2), an end scraper, a retouched flake, hammerstones (n = 3), and a cobble core (Chlachula and Leslie 2001:876). All materials were recovered from the lower till, immediately above its contact with the underlying gravels, over a distance of 30 m. The material is comparable to the Varsity Estates and Silver Springs sites (Chlachula and Leslie 2001:876). An age prior to the last glaciation is assigned to this material, presumed to be the Middle Wisconsinan (Chlachula and Leslie 2001:883).

Pre-Clovis: Still Searching for the Evidence

Unfortunately, we are not yet able to say when the first people entered the Americas, or Alberta for that matter. Such an “event” continues to be a contentious issue. Many archaeologists advocate an early peopling in order to account for sites such as Monte Verde. Others strongly support a later dispersal with the Clovis complex as the material representation of the dispersal phenomenon. The route that people traveled or populated is also controversial, with the Ice-free Corridor and a coastal migration most commonly touted.

Research within Alberta has added fuel to this debate. The announcement of the Pleistocene-age human infant skeleton near Taber produced substantial controversy at the time. Finds of exceptional antiquity are usually confronted with caution and scepticism in the archaeological community. The 1983 paper presentations on “Dating the Taber Child” in the Canadian Journal of Archaeology (vol. 7, no. 2) demonstrated a strong consensus that the skeleton was an Indigenous infant of modest antiquity, perhaps a few thousand years old.

The recovery of lithic artifacts in close association with glacial till has been proposed at three sites in Alberta: Varsity Estates, Silver Springs, and Grimshaw. Also, the Varsity Estates and Silver Springs sites have received substantial criticism concerning both the nature of the artifacts and the purported age of the deposits from which they came. The reconstruction of the palaeoenvironment at the Varsity Estates and Silver Springs sites has been strongly questioned. Young, Rains, and Osborn (1998) indicated that the literature suggests that a Late Quaternary ice sheet coalesced in the Calgary area, followed by glacial lake formation rather than an interval for human occupation. This assertion is further supported in subsequent research detailing the NWSE oriented landforms that could have only been created by coalescing ice, and cosmogenic 36Cl dates of 12–17 ka on foothills erratics together indicating Late Wisconsinan coalescence (Osborn et al. 2000:209–215). Furthermore, concerning the artifacts themselves, Gillespie, Tupakka, and Cluney (2004) questioned the cultural origin of the stone tools. Sixteen lithic attributes were shown to exhibit significant differences between known archaeological samples and known natural samples from a river bed. When the Varsity Estates and Silver Springs sites were evaluated using these attributes, the results indicated that the sites were most like the geological specimens or geofacts (Gillespie et al. 2004:630–631).

The Grimshaw site has received similar criticism. Driver (2001) focused his criticism on the unequivocal assignment of the lithics at the site as culturally fractured rather than naturally fractured. He felt the criteria used to distinguish humanly flaked stone were never proven to be limited to artifacts. Control specimens were never established from analogous sediments to compare to the Grimshaw artifacts. Because of this shortcoming in the analysis, “it is just as likely that the fractured Grimshaw cobbles were produced naturally as by humans, and better evidence is required to support an argument for preglacial humans in western Canada” (Driver 2001:873).

It is fair to state that there are no unequivocal pre-Clovis sites known in Alberta. That is not to say that such sites do not exist, only that they have not yet been recognized or confirmed. The task in front of researchers rests in developing reliable methods of identifying where these sites could be located and being able to distinguish their assemblages from naturally occurring materials. Both the Varsity Estates and Silver Springs sites are buried deep beneath metres of glacial and postglacial sediments. Even early Palaeoindian sites, dating later in time, are likely to be buried under substantial amounts of sediment given the highly active geomorphology at the glacial/postglacial boundary. For example, the Indian Creek Folsom occupation in the front range of the Rocky Mountains in west-central Montana is approximately 7.5 m below the surface. Its discovery was facilitated by cutbank erosion (Davis and Greiser 1992).

2 Early Prehistoric Period
CA. 11,050 BP TO 8,600 BP

Many researchers consider the coalescence of the Cordilleran and Lauren-tide ice sheets to have ceased around 12,000 BP (e.g., Osborn et al. 2000). Following this, the exposed land between the retreating ice sheets produced the so-called Ice-free Corridor. This new landscape was relatively unstable, exhibiting rapid change resulting from deglaciation. Meltwater created extensive proglacial lakes and erosional landforms. Revegetation likely occurred from plant communities to the south. Data gathered mainly from pollen records in the foothills suggests shrubby open vegetation in the mountains, open vegetation along a broad band at the foot of the mountain range, and spruce-dominated vegetation in the northeast (Beaudoin and Oetelaar 2003:196). By 11,000 BP, the ice had largely retreated from the province with the possible exception of the very northeast corner. Aeolian activity in dune fields, common in parts of Alberta, may have preserved or destroyed evidence of human activity at this time (Wolfe et al. 2006). Periglacial processes lessened as the glaciers retreated. By 10,000 BP, landforms were more stable and rivers began to incise into the terrain (Beaudoin and Oetelaar 2003:199). The dynamic environment that predominated during this time period likely played a large role in limiting the number of sites preserved and the archaeological visibility of sites from the Early Prehistoric period. The question remains whether the dynamic environment largely destroyed or deeply buried evidence of human activity.

CLOVIS PHASE (CA. 11,050 TO 10,800 BP)

The earliest widely accepted archaeological culture in the Americas is the Clovis phase of the Llano tradition. The most characteristic artifact of the phase is a large, fluted projectile point: the Clovis point. Flaked stone technology exhibiting bifacial and flake tool manufacturing is also common. Flake tools include knives, gravers, end scrapers (in some cases spurred), and large (macro-) blades (Haynes 1993). The fluted projectile points have sharp tips for initial penetration, sharp blade edges for cutting a hole to allow further penetration of the point, flutes for easy attachment to a notched foreshaft, and a narrower basal edge area than distal area for application of the binding so it does not inhibit penetration (Frison 1993b:241). As well, Clovis toolkits often contain what appear to be cylindrical, bevelled bone points, bone shaft wrenches, and rib segments with rounded and polished ends (Haynes 1993). Based on recoveries from the Anzick Burial site, ceremonial items apparently include exceptionally large and well-made Clovis points along with large bifaces, cylindrical bone rods and points, and red ochre (Haynes 1993:219). In terms of the range of sites encountered in the Clovis phase, most excavated sites have been interpreted as either kill sites or meat-processing sites, with the remainder being interpreted as burials and caches (Bradley 1993:253).

The subsistence strategy of the Clovis phase is most commonly thought of as big-game hunting. Evidence from fauna recovered at Clovis sites indicates a more complex subsistence than simply hunting of large mammals. Grayson and Meltzer (2002) reviewed the faunal assemblages from seventy-six sites and determined that only fourteen provided evidence of Clovis predation on megafauna, with twelve sites containing mammoth and two containing mastodon. Likewise, Clovis people were not preying on the full array of then-extant fauna that are now extinct, which suggests that Clovis people had little role in their extinction (Grayson and Meltzer 2002).

Delimiting the geographic extent of Clovis phase is complicated by the difficulty of differentiating Clovis points from other fluted points (Howard 1990). Clovis points are “comparatively large and heavy bifacially flaked fluted lanceolate points, lenticular to near oval in cross-section with parallel to moderately convex lateral edges” (Howard 1990:259), which are “normally fluted on both faces. Flutes are most often produced by multiple flake removals. Length and quality of flutes are greatly variable, with length usually 30 to 50 percent of overall point length” (Howard 1990:259). Importantly, the eastern fluted points do not match this description as they tend to exhibit more advanced technological refinements such as large single channel flakes produced using the Folsom technique, deep basal concavities, constricted waists, extensive basal retouch, and comparatively fine, often sequentially positioned overall flaking (Howard 1990:259). A recent continent-wide study of Early Palaeoindian points supports this classification. Buchanan and Collard (2007) used cladistic analysis on morphological attributes of Palaeoindian projectile point assemblages. The analysis was conducted on the grounds that the migration route into the Americas by a group of people could be inferred by the relationships amongst the various populations left along the route; they argued that cladistics, because it takes into account projectile point morphology, could infer the historical relationships between the groups of people (Buchanan and Collard 2007). Their data indicate that the Clovis phase was a rapidly migrating population having its origin in the Ice-free Corridor, or possibly the Northwest Coast, but not the Isthmus of Panama or an ice bridge across the mid-Atlantic (Buchanan and Collard 2007; Hamilton and Buchanan 2007). Thus, the earliest Clovis points and Clovis variant points are found on the northern and western plains; the eastern fluted material is related to Clovis but is a later derivative, more properly “Folsom-like.”

The extreme rapidity with which the Clovis phase spread across the Americas from the northwest appears to be demonstrated in the aforementioned analyses by Buchanan and Collard (2007) and Hamilton and Buchanan (2007). Kelly and Todd (1988) suggested that Clovis hunters were not analogous to any modern hunter-gatherers. The apparent continual range shift would have brought these hunters into new and different environments but a primary dependence on large fauna would not have required a niche shift (Kelly and Todd 1988:234–235). As Clovis people were likely entering territory uninhabited by humans, they would have had few if any neighbouring groups to rely upon for assistance. They would have had no detailed knowledge of the terrain, thus, they would have used the landscape in a short-term and redundant fashion (Kelly and Todd 1988:235). Large Clovis bifaces of high-quality raw materials provided a transportable technology that was usable in unknown terrain for hunting-specific tasks (Kelly and Todd 1988:235). Long-term storage at successful kills would not be expected as stores reduced mobility; rather, renewed resource procurement was a less risky strategy under conditions of regionally abundant but locally unpredictable resources (Kelly and Todd 1988). An example supporting this model is Blackmar’s (2001) examination of the distribution pattern of Clovis points in Kansas, Oklahoma, and Texas; it suggests that the distribution of Clovis materials represents a homogeneous occupation across the study area, possibly reflecting a subsistence strategy towards large mammals.

Traditionally the Clovis phase has been dated between ca. 11,500 BP and 10,900 BP (Frison 1991a:25). More recently, Waters and Stafford (2007) have determined a more accurate time span for the Clovis phase by obtaining and analyzing high-precision accelerator mass spectrometry 14C ages from previously dated sites. Their research suggested that the Clovis phase existed between 11,050 BP and 10,800 BP. Thus, the Clovis phase would have been present over a roughly 250-year period, regardless of whether one examines radiocarbon or calendar years (Waters and Stafford 2007:1124).

In Alberta, there are no Clovis points from excavated contexts in demonstrable association with other Clovis material. The vast majority of the evidence for the Clovis occupation of Alberta has been recovered as isolated projectile points from cultivated fields and eroded areas. Comparison of the formal morphology of these projectile points to specimens of known age from adjacent regions of the United States has provided the evidence for their chronological placement. A brief historical outline of the recovery of Clovis points in Alberta follows, documenting the development of research in this area over the past few decades.

In the late 1930s, an eroded area north of Chinook produced a complete Clovis point and a point base along with other Palaeoindian artifacts. This site is called Johnston Locality 4 or EkOr 1 (Wormington and Forbis 1965:75–80, especially 77, fig. 20b; Gryba 1988:A3-33 to 35). In 1955, when William Mulloy of the Glenbow Foundation investigated the location, he noted that the area had likely been destroyed by erosion (Wormington and Forbis 1965:75). In the early 1950s, Armin Dyck found a Clovis point in the bottom of a trench while serving on a city road construction crew in Lethbridge (Wormington and Forbis 1965:135, fig. 55). Wormington and Forbis (1965:177, fig. 77a, b) mentioned three additional locations of lanceolate points that resemble possible Clovis sites. One locale produced two lanceolate points near Clear Hill, northwest of Peace River (Wormington and Forbis 1965:176; 177, fig. 77a). A second locale, located north of Peace River, produced a cache of up to 150 flakes along with two lanceolate points (Wormington and Forbis 1965:183). E. Mott Davis tested the location, labelling it HbQi 1, with a cross-shaped trench covering 400 square feet (~37 m2), but he recovered only a few bone fragments (Wormington and Forbis 1965:183). The third locale was a cache located southeast of North Star. The cache consisted of bifaces as large as nine inches long (~23 cm), flakes, and “arrowheads” (Wormington and Forbis 1965:180). Similar caches of bifacially flaked specimens of black stone were reported for at least two other locations in the area (Wormington and Forbis 1965:180). The large bifacial tools and lanceolate points are hallmarks of Clovis cache strategy (e.g., Gillespie 2007).

Gryba (1988) conducted an inventory of fluted point occurrences in Alberta, with the objective of bringing together, in a single source, as many occurrences of fluted and Early Prehistoric period points as possible. He recorded 150 fluted point specimens during his fieldwork, including Clovis points (n = 46), multiple-fluted points (n = 13), fluted points (n = 21), Folsom points (n = 18), Midland points (n = 15), and basally thinned points (n = 37), as well as Palaeoindian points (n = 29) and one large biface fragment. (Gryba 1988). Gryba (1988:16) noted the heavy reliance on locally available lithic raw material in point manufacture, although some exotic materials were used. In terms of the geographic distribution of the Clovis points, Gryba (1988:17) described their almost exclusive presence in southern Alberta south of Cold Lake, as well as near Peace River, with a strong association to areas with agricultural and developed landscapes. Almost all subsequent research on fluted points has built on this significant collection of raw data.

Gillespie (2002; Gillespie et al. 2002) built upon Gryba’s (1988) database, from which he produced a typology. Attributes used in the typology were restricted to those related to basal morphology as these were seen as less controlled by function than the blade; a total of eight types were created (Gillespie 2002:65–80). In comparing the Alberta fluted-point database to an American fluted-point database (Tompkins 1993), Gillespie (2002:81–97) repeatedly found the strongest similarities between Alberta fluted material, Saskatchewan fluted material, and western U.S. fluted material, rather than to eastern U.S. or South American materials. The raw materials utilized in manufacturing the Alberta fluted points were dominated by locally available quartzite, siltstone, and mudstone (64%). In total, 71 percent of the raw materials were local, 16 percent were exotic, and 13 percent were undetermined (Gillespie 2002:108–109). This extensive use of local materials is in contrast to both modelled (Kelly and Todd 1988) and observed (Lahren and Bonnichsen 1974) Clovis lithic activity to the south in the United States.

Concerning the origin of the fluted material in Alberta, Gillespie (2002:130–135) argued that entry into Alberta could not have been from the north through the Ice-free Corridor, as palaeoenvironmental and faunal evidence suggests that people could not have been supported within the corridor before 12,000 BP. Instead he proposed a northward migration into the province, based on the current lack of known human occupations in Alberta that predate fluted point sites in the United States (Gillespie 2002:138). Further, he compared the similarly undated Alaskan fluted points to the Alberta sample, as both exhibit triple fluting and heavy pressure flaking. Again he saw similarities that he interpreted as evidence of a northward migration.

The Sites

As stated above, surface finds of Clovis points are not common. It is difficult to assess the exact number of Clovis points that have been recovered from surface finds, owing to their lack of stratigraphic context, which causes them to lack key information about associated material culture. That is, there are many fluted points in Alberta that likely postdate the Clovis phase (see the Sibbald phase below) or are regional variants of the classic form (e.g., Gillespie 2002:80). Regardless, there are a number of Clovis points recovered from the surface and they are discussed below (see Plate 2 and Figure 3).

Gryba’s Inventory. Gryba (1988) provided the first exhaustive inventory of fluted points in Alberta. Many of the Clovis points known today are a product of his extensive and tireless research on Palaeoindian surface finds in the province. Table 1 summarizes keys aspects of his report for Clovis point recoveries.

EfPl 93. EfPl 93 is a buried campsite at the prairie level on the east side of the Bow River about 0.8 km south of its confluence with Fish Creek in south Calgary (McIntyre 1975). Ten 2-×-2-m units were excavated at the site. Two occupations were recorded. The lower occupation produced an ash-filled elliptical hearth associated with fire-broken rock (FBR) (McIntyre 1975:15–16). The upper occupation consisted of two rock-lined hearths, a surface hearth, and concentrations of FBR with bone and a few artifacts. A Besant point was collected from the surface while the base of a Clovis point was recovered from 20 cm below surface (cm BS) (McIntyre 1975:15–17; Gryba 1988). The Clovis specimen exhibited an arced base and straight lateral margins. The occupation was interpreted as a one-time, short-term campsite likely during the Besant phase. The Clovis point base was considered to be out of context and likely redeposited by later users of the area (McIntyre 1975).

Clovis points. Illustrated are specimens from the Reed collection (a); Lougheed collection (b); Haug collection (c); DhPg 8 (d, i, n, and o); Bull collection (e); Cameron collection (g and l); Dyck collection (h); Johnston collection (j); Blumhagen collection (k); and Sawyer collection (m). Photo credit: Eugene Gryba (a–c, e, g–h, j–m); Shayne Tolman (d, f, i, n, and o).


Clovis and Goshen sites within Alberta


TABLE 1 Clovis points listed in Eugene Gryba, An Inventory of Fluted Point Occurrences in Alberta (1988)