Abstract
This thesis provides a model for understanding how social classes arose in the Gulf of Georgia area. This model distinguishes how social status in rank and class societies are manifested and maintained in non-state, kin-based societies, drawing mainly from ethnographic descriptions. The relationship between the living and the dead for making status claims in both rank and class societies makes the archaeological study of mortuary ritual important for investigating these relationships. I propose that burial mounds and cairns, which were prominent in the region from 1500 to 1000 years ago, reflect a time when status differentiation was defined mainly through social rank. Following this period, when all forms of below-ground burials cease and above-ground graves become the dominant form of mortuary practice, I propose that the historically recorded pattern of social class emerged. Archaeological investigations of the burial mounds and cairns at the Scowlitz site have provided the first fully reported instances of mound and cairn burials in this region. Using less well reported data from over 150 additional burial mounds and cairns reported from other sites in the region, evidence for the nature of status differentiation is sought out. Patterns in the burial record are investigated through discussing variation within classes of burials, demography and deposition, spatial patterning, grave goods, and temporal variation. These patterns and changes are then discussed within the context of the larger culture history of the region, suggesting that the late Marpole or Garrison sub-phase may be defined as ending around 1000 BP with the cessation of below-ground burial practices. The general patterns observed in mound and cairn burials and the changes in mortuary ritual subsequent to their being built generally support the idea of a shift from a rank to a class society. This thesis provides a basis for further investigation of questions of social status and inequality in the Gulf of Georgia region.
Acknowledgement
A number of people have helped me bring this work from an idea to a finished piece of writing. Firstly, my sincerest thanks goes to my supervisor, Dr. R.G. Matson. Dr. Matson has provided his knowledge, wisdom, encouragement and advice throughout this project. He has been patient and careful throughout the research and writing process. My second committee member Dr. Bruce Miller has been an excellent sounding board for my ideas about interpreting the ancient history of the aboriginal people living in the Gulf of Georgia region. Our many excellent conversations helped me gain a deeper appreciation for the ethnography of the Central Coast Salish people. Dr. Michael Blake's enthusiasm for the investigations of burial mounds and cairns peaked my initial interest in this project. Dr. Richard Pearson has given me guidance and advice throughout my academic career.
Sincere thanks must also be extended to the First Nations people who made investigating this research topic possible. Chief Clarence Pennier, and Sonny McHalsie of the Stó:lô Nation, and Diana Henry and the elders of the Saanich Native Heritage Society all supported the research and shared with me their respect for their ancestors. Gordon Mohs, archaeologist for the Stó:lô Nation also provided encouragement, enthusiasm and respect for research into mortuary ritual.
The work conducted at the American Museum of Natural History was made possible by a Collections Study Grant and the support of David Hurst Thomas and Anibal Rodriguez. Jerry Cybulski of the Canadian Museum of Civilization provided invaluable advice and a sharp eye for accuracy regarding the data presented in this work. Brian Chisholm aided me in selecting and submitting samples for radiocarbon dating.
The 1993 excavations of the burial mounds at the Scowlitz site were enthusiastically conducted by the University of British Columbia Archaeology Field School. My friends and colleagues Alison Baily, Jeff Brantingham, Doug Brown, Vicki Feddema, Peter Gron, Joyce Johnson, Terry Johnson, John Maxwell, Sandra Morrison, Heather Myles, Heather Pratt, Angela River, David Schneider, Lynn Vanderwekken and Sue Woods all volunteered their time and energy to help excavate the burial cairn at Scowlitz. Thank you very much all.
Tracey Mason and Chris West helped at various stages of preparing the manuscript. Deborah Tuyttens and Heather Pratt have always shared with me their support and wisdom. My parents have given my continuous and unwavering support. Finally, I would like to acknowledge Lynn Vanderwekken for the countless hours she spent proof-reading and talking with me about the project. Her loving care and concern for me and my work has helped build my abilities as a scholar and a person.
Introduction
Aboriginal people in Central Coast Salish society who occupy the watershed of the lower Fraser River, southeastern Vancouver Island and the interceding islands of the Gulf of Georgia (Figure 1; see also Suttles 1990:454), have been documented as having been divided into three distinct social classes (Suttles 1990:465). This thesis is an investigation into the development of social class among the ancestors of these people. I propose that a transition in how social status was expressed occurred from the Marpole phase when social rank was the main means of defining social status and inequality to the Late phase, where social class stratified the society. I develop a model, based on ethnographic observations, which distinguishes between rank and class societies. This model emphasizes the importance of achieved or acquired status for leaders in communities where social rank is the main expression of social inequality. Ascribed or inherited social status is also important in rank societies, but is not critically linked to the ownership of inherited resources, wealth and privileges. As social status competition becomes more intense, these things become essential for the creation and maintenance of social classes. Social class is distinguished mainly by ascribed status, where families control productive resources, wealth and privileges through inter-marriage of elite families over a wide region.
Many of these relations of inequality are exhibited in mortuary ritual. I examine the tradition of burial mounds and cairns, which occurred throughout the Gulf of Georgia region during the late Marpole (Garrison sub-phase) phase, from 1500 to 1000 BP. By examining patterns within this tradition and the changes in mortuary ritual which followed it, I suggest that social rank was the most important measure of social inequality during the Marpole phase. Increasing control over wealth during this time resulted in a shift to social class at the beginning of the Late phase, as exhibited in part by changes in mortuary ritual to above-ground interments.
I evaluate these ideas through analysis of the existing Gulf of Georgia burial mound and cairn data. A total of 155 individual burials from a sample of over 100 archaeological sites have been tabulated (Appendix II) in order to observe patterns and changes in mortuary ritual. Five lines of mortuary evidence are examined, including a classification of these burials; observation of patterns in demography and context of deposition; investigation of regional and local spatial patterns of mortuary ritual; trends in quantities and classes of grave goods, and changes in burial practices over time. The results of this analysis largely support the model presented for the development of social class at the end of the Marpole phase.
The first section of this thesis describes social rank and class and provides a model for how social class may develop in a kin-based society. Descriptions of social class are provided from ethnographically documented Central Coast Salish society. From this material, inferences are drawn about how archaeological evidence of mortuary ritual may provide insights into social rank and class. The following section provides a discussion of mound and cairn burials from the Gulf of Georgia region, through providing a detailed description of the burial mounds and cairns investigated at the Scowlitz site. Following this, the burial data is analyzed in order to evaluate how social status may have been exhibited and acquired through the practice of mound and cairn burial. Finally, interpretations of the archaeological record are made from the evidence of burial mounds and cairns and are placed in the context of the larger culture history of the area. Some suggestions are made to prompt future archaeological research into the questions of the development of social class in the Gulf of Georgia region.
Defining Social Rank & Class
The investigation of the development of social class is central to understanding the broad issue of social complexity in societies that do not practice agriculture and have little political centralization. In particular, this issue is an important aspect of the historical development of regional social structure among the Central Coast Salish people of the Northwest Coast of North America. There has been a long debate in the local ethnographic literature as to whether Northwest Coast societies had social inequalities divided along lines of rank or class (see Suttles and Joanites 1990 for a review). Suttles has provided the most convincing evidence of these societies as having class-based social stratification (Suttles 1958). Since the 1970's archaeologists have tried to document and explain the origins ascribed social inequality, which may be traced to at least 2,400 years ago (Matson and Coupland 1994; Moss 1993). However, these archaeological models have not clearly defined the nature of inequalities visible from archaeological evidence, often assuming that evidence for ascribed status adequately represents ethnographic descriptions of class. This thesis distinguishes issues of rank/class, and achieved/ascribed status by providing ethnographic examples of these different kinds of social inequalities. Out of this I have constructed a model for the development of social class which may be evaluated by examining patterns of and changes in the archaeological record. The ultimate goal of this thesis is to shed light onto the question of the origin of the social classes documented by ethnographers for the Central Coast Salish through an investigation of prehistoric mortuary practices.
The model I have developed here is largely a neo-Marxist one, where social class structures social and economic relations, beyond the bounds of kinship which govern most relations in non-state societies. This model builds on the early work of Freid (1967) and the more recent work of Collier (1994), Gailey (1987) and McGuire (1992). These scholars have suggested ways to understand how societies with inequalities based primarily on social rank may develop social classes. The model proposes that social classes form when inter-married elite families gain control over wealth through ownership of productive resources and access to inter-regional systems of exchange. The control of wealth by these elite families makes low-status families dependent on them for access to resources. Such class-based social organization perpetuates itself through the ideology of kinship as membership in the upper-class is mainly ascribed through marriage patterns and primogenitor. However, high social status may also be obtained by individuals through their successful efforts of achieving social status, thus ranking can occur within a social class. The upper-classes are maintained by their members being trained in "proper". Such behaviour includes the maintenance of private knowledge about ritual, resource locations and techniques, genealogies, and gossip about other families (Suttles 1951; Amoss 1978). Social classes in Central Coast Salish society are corporate groups composed across various affiliated nuclear families, houses, villages and satellite villages (Miller 1993:381).
To understand the development of social classes from a system of social rank, the nature of how these inequalities are manifested in a society must be defined. Freid has defined a rank society as "one in which positions of valued status are somehow limited so that not all those of sufficient talent to occupy such statuses actually achieve them" (Freid 1967:109). Rank societies differentiate status on the basis of a combination of many factors including age, gender, position in kin-group, and life-experiences (Gailey 1987). Having social rank does not simply equate with one's achieved status, but rather involves a complex set of social relations and obligations. The age and gender of an individual often provides cultural limitations to how much or what kind of social status can be achieved. Family or kin connections are a contributing factor in gaining social status in a rank society. However, there is a certain ambiguity of rank based on kinship alone, as contractions in descent-based rank (age and gender), shifting marriage patterns and an individuals ability to generate a following of labour makes social ranking somewhat ambiguous (Gailey 1987:47). In Gailey's words "[c]ontradictory ranking principles created a situation where claims to others' labor had to be justified through the establishment of higher rank in the eyes of the society." (Gailey 1987:48). Thus, although ascribed status can be a factor in determining over-all social rank, achieved status gained through "status rivalry" (Gailey 1987:48) is the most critical.
Collier (1994) has suggested several ways that leaders achieve and maintain status in rank societies. These leaders gain authority and legitimacy "by competing with other would-be leaders to demonstrate who is the most willing to sacrifice personal well-being for the common good..." playing down "differences in lifestyle from their followers while calling attention to the selfless service they give in leadership" (1994:120-121). Rank-based power is maintained through "serving their communities by acting as brokers with the outside world or as intermediaries to other groups" (Collier 1994:120) in relations of trade and disputes. However, some of the power gained in these situations is disseminated through fulfilling social obligations to share wealth at gift-giving occasions. Social obligations are owed to family members who assist in the labour of production or who provide exchange goods which aid in the production of wealth. Although differences between leaders and their followers is ideally minimized, real power differentiation does occur.
In contrast to rank-based stratification, social classes form as a result of families gaining more exclusive access to power and wealth (Freid 1967). Leaders in a class-based society are frequently "power holders [who] use political position to shepherd economic resources in a more exclusive manner, usually to better themselves at the expense of others, even if doing so means excluding others from access to wealth" (Collier 1994:121). In kin-based societies which have social classes, status differences often occur between inter-married, elite family groups and other families who do not have inherited connections to wealth. The families who inherit wealth can maintain their position of high-status and power through generations, by (1) limiting the ability of others to obtain productive surpluses and in turn, wealth; (2) controlling access to inter-regional trade and exchange networks and; (3) establishing and controlling the symbols which reinforce their class position.
What kind of social processes leads a society to change from rank-based to class-based stratification? There is not one simple answer for this question. The model I have developed follows Collier's observation that controlling wealth is the main means for elites to maintain power in a class-based society, as opposed to successfully maintaining social obligations, which is a major factor for gaining and maintaining status and prestige in rank societies (Collier 1994:120). High-status individuals will limit the number of families who can access wealth by making it socially unacceptable for high-status people to marry into low-status families. Gift-giving at public occasions initially disseminates wealth among all families, but as obligations to give gifts becomes limited to elite kin, access to wealth becomes constrained for low-status families. It becomes increasingly important for low-status families to trace their connections to high-status families in order to gain access to productive resources and wealth. High-ranking individuals have an additional responsibility to coordinate inter-regional interactions such as exchange and warfare. Individuals who take leadership roles in interacting with distant communities will have additional opportunities to arrange for beneficial marriages of their children into other elite families - thus extending the regional class system. Rare prestige items obtained in inter-regional exchange become important for symbolizing high status. Establishing kinship ties to elites over a wide region creates networks of high-status families that control access to wealth on a regional level. Productive inequalities are taken to extremes when high-status individuals who are successful in resolving disputes with other groups begin to make slaves out of their adversaries. Concentrated wealth and power can sustain such drastically unequal relations for generations.
McGuire has suggested that in class societies "the elite are no longer embedded in the kin group, but form their own kin groups; this removes products (surplus) from the kin groups of direct producers and does away with a major limitation on the exercise of power" (McGuire 1992:186). However, in a kin-based society were social classes exist, members of the higher classes still have many of the social obligations to their broader community, thus limiting some of the centralization of power that might create new levels of political organization.
Social Inequality in the Gulf of Georgia Region of the Northwest Coast
Given the above understanding of how social inequalities are differently defined and maintained in rank-based and class-based societies, I now turn to the Northwest Coast archaeological and ethnographic records. Much of the archaeological research concerned with social organization has frequently focused on observing the first signs of ascribed social status. Burley and Knüsel (1989) have provided the most convincing arguments from the burial record, indicating that ascribed status was present at least by Marpole times (2400 to 1400 BP), as shown by the presence of sub-adults with elaborate grave goods. Matson and Coupland have summarized much of the current archaeological research in the area stating that during the Marpole period, "large winter villages with enormous, multifamily planked houses, an economy based on stored salmon supplemented with other seasonal resources [and] highly developed art" were present during this period (Matson & Coupland 1994:241). The available archaeological evidence clearly indicates that ascribed and achieved social inequalities were well established by at least 2400 BP. While archaeological analysis have clearly shown the antiquity of social inequality in the Gulf of Georgia region, Central Coast Salish ethnography (and more generally the ethnography of the Northwest Coast) has revealed a more detailed picture of social stratification during the late 18th and early 19th centuries.
Turn of the century ethnographers such as Hill-Tout (1905) and Boas (1894) saw Northwest Coast cultures as being class divided. From the 1930's to the 1960's, a "rank" model for pre-contact Northwest Coast culture was prominent in the literature (ie: Barnett 1938, Drucker 1939). Since Suttles' important publication "Private Knowledge, Morality, and Social Classes among the Coast Salish" (Suttles 1958), ethnographers have increasingly viewed these societies as being formed into three social classes (Suttles 1960, 1990; Elmendorf 1960; Donald 1985, 1990).
Social and economic relations occurred on a number of levels in Central Coast Salish society including family, household, local group, village and kin-group (Suttles 1990:464-465). The family was made up of a husband, wife (wives), minor children, dependent older relatives and slaves. Several families make up a household, which participates in many economic and ritual activities together, and is led by a man and woman who oversee activities. The local group gains its identity through the ideology of descent from a common ancestor and is composed of members of an established household, and one or more affiliated, dependent houses. Members of this local group may break off to form new groups or join other groups, depending on circumstances. Members of local groups do not regard themselves as having affinal or consanguineal relations - even though they are descended from a common ancestor - and are divided into upper- and lower- classes. The village is formed by a number of houses and local groups who are frequently tied through relations of close kinship. The kin group is traced within and between different residential communities and provides the basis for affiliation patterns including the sharing of resource rights, names, ritual activities and privileges. Kin groups are not necessarily co-residential and frequently form across household and local group and village boundaries.
The most critical difference between upper- and lower- class people in Central Coast Salish society is their fundamentally different rights access to important resources and wealth (Suttles 1987 [1960]:16-17):
Within most communities there seem to have been three distinct social classes - a majority identified as "high class" [sim], a somewhat smaller group identified as "low class," [sslyn or stšam] and still smaller groups of slaves [sk'wys]. The slaves lived in the households of the upper class; the lower class often occupied separate houses in its own section of the community or in a location sufficiently separate so that it might be regarded as a lower class community subservient to an upper class group. In Native theory the lower class consisted of people who "had lost their history," that is, people who had no claim to the most productive resources of the area and no claim to recognized inherited privileges, and who furthermore "had no advice," that is, they had no private knowledge and no moral training.
In Central Coast Salish society high, low and slave classes clearly have different access to productive resources, wealth and power. Such differences in actual power are clear where poor, low-classes "had fewer extra-village affines who had shared with them and enabled them to build their supplies" in times of stress (Suttles 1973:622).
These social classes comprised social networks that were formed mainly by following a pattern of elites maintaining marriage ties to high-class families in other communities. As Amoss has described, "the artful knitting of all the principles of affiliation", which included kinship and residence principles and supported "a social web which promoted the economically critical exchange system" (Amoss 1978:28). Such a social web integrated many of the same general principles of resource location ownership, control over inter-regional interaction, and accumulation of wealth. This network of economic and social relations integrated inherited privileges with obligations of reciprocal exchanges of food between kin members, and redistribution of food to the broader community.
High-class families had inherited privileges to particular resource locations such as productive fishing spots, clam beds, or camus-bulb plots. However, to substantiate a claim to such valuable inherited resource locations, individual leaders were obliged to give public feasts where wealth was distributed to other elites, extended family members and followers. Gaining enough wealth to host such feasts was not done by individual nuclear families alone, but required the host to utilize and organize the human resources of their extended family and non-kin followers who would participate in the food- and wealth-quest at sites owned by the host family. Suttles has also observed that an individual who was able to produce surplus food (because of his or her productive resource location) "could release some of the members of his household from food-producing activities and let them produce wealth" where in turn he or she could "attract more food-producing and wealth producing persons to his household" (Suttles 1987 [1960]:22). These important individuals "mobilized labor of house group or village to construct fish traps and smoke houses" which could be used for more efficient procurement and production of resources (Amoss 1978:10). As such, upper-class families controlled the resources which could be used to gain wealth and lower-class families provided the labour, receiving benefits or "thanks" at public events.
The production of surplus food and wealth cannot be thought of in a strictly local sense, but was tied up into a system of regional marriage and exchange networks. Central Coast Salish social classes were maintained by the inter-marrying of elite families throughout the region. Duff has commented that "such marriages brought prestige as well as the social and economic advantages of alliances with important families in distant villages" (Duff 1952:81). Social and economic advantages would include access to resource locations that were not locally available through affinal exchange (Suttles 1960). Amoss has summarized the importance of exchange for the mid-20th century economy of the Nooksack, a Central Coast Salish group who live in the Nooksack River watershed:
They [the Nooksack] travel extensively within the wider limits of the Coast Salish community and know people in all the various villages and reserves. Because of the long history of village and tribal exogamy, almost every Coast Salish Indian who has been able to find out who his relatives are has kin ties in every settlement. Not all these potential kin in a personal network are activated, but anyone who participates regularly in one or another of the pan-Salish ceremonial systems will actualize a high percentage of his potential kin ties (Amoss 1978:36).
Suttles has further described how only upper class individuals possessed an intricate knowledge of their extended family base, and that this kind of important knowledge was private. Halkomelem terms for low-class people [sslyn or stšam] are frequently translated as "people who have no history". Low-class people were not able to gain the surplus resources themselves to participate in extensive exchange networks, nor make claims to inherited privileges of resource locations which, in theory, they may have been able to (Jenness n.d.:58). Their economic role was thus somewhat limited to providing labour to the high-class individuals who were in turn, socially obliged to share some of the surplus resources with them. Slaves were an essential component of this labour and occupied the lowest levels of status, not being able to give or receive wealth at public gatherings.
The final mechanism for maintaining social class that Suttles has observed is the display and giving-away of wealth. This occurred at large public gatherings (potlatches) where "by giving wealth at a potlatch a man validates a claim to noble descent and inherited privilege and thus converts wealth into high status" (Suttles 1987 [1960]:22). The wealth received was in the form of "blankets, shell ornaments, fine baskets, hide shirts, bows and arrows, canoes, slaves - items of varying utility but all relatively imperishable" (Suttles 1987 [1960]:22). Having control over such wealth items materially, as well as ideologically, separated high-class and low-class individuals in Central Coast Salish society.
This class system operating in a kin-based society has a number of checks and balances to power, which likely contributed to the maintenance of relatively low levels of political centralization. McGuire has documented these patterns in other kin-based class societies: "In many incipient cases of class formation, kin relations still structure production, and surplus extraction requires that class relations be masked as kinship" (McGuire 1992:186). Donald has convincingly argued that this "masking" ideology is an important mechanism for elites to maintain their position of high-status in the Northwest Coast societies. He asserts that the ideology of a rank society, particularly the notion of "obligations" to kin, is used by high-class individuals to validate and legitimize their elite social status:
"...[the rank without class society] interpretation may be called the 'traditional' one in a more important sense: native informants often described their society in such terms to ethnographers. Given the origin within their societies of most informants (the upper or title-holding group), this is not surprising ... Given that class divides and rank unites, the advantage of emphasizing rank from a titleholder's perspective is obvious; it is part of the ideology that the ruling elite uses to perpetuate itself. Thus, the kind of title-holder statements about rank described in this paragraph are not strong evidence for the "rank without class" view of the Northwest Coast society" (Donald 1985:241).
This "masking" of social classes is a strategy for high-status individuals in class divided societies to reinforce their own position while acknowledging their obligations to kin. This relationship can also be seen with regards to the "generous" nature of Central Coast Salish leaders, where institutions like the potlatch and more common feasts require high-class individuals to give away large amounts of food and wealth, but at the same time re-enforce their own status and prestige.
Central Coast Salish Mortuary Ritual
One gains the impression that burial styles, including confining methods and position of the body, may have fluctuated or changed rapidly in aboriginal times. There is presumptive evidence... that canoe burial started as an upper-class trait and had spread throughout the society by 1850; flexed burial in boxes on the ground surface was by then a conservative method restricted to some poorer families, while box burial in trees, sometimes used for slaves, may have been a yet older custom (Elmendorf 1960:455).
Elmendorf's observations of the relative status and class of individuals buried in different types of burial markers suggests the importance of mortuary ritual for expressing the social status of the deceased. However, the seemingly arbitrary nature of the symbols (canoe burials, flexed box burials, tree burials, etc) used in association with the mortuary ritual, makes it difficult to interpret the relative status of individuals. Furthermore, the rapid changes over time illustrate how difficult it is to interpret these issues with any degree of certainty. The ideology associated with Central Coast Salish mortuary practices reveals the meaningful connection of beliefs associated with the dead to the structure of social inequalities among the living.
In 1899, Ellen Webber recorded a discussion she had with an elderly Kwantlen woman who mentioned the burial practices of former times. In this narrative, the association of the deceased with the spirit world was critical:
It was the custom of our villagers to bury their dead within an hour of death. They were in most cases placed in a tiny house raised on posts; but, if there was no house ready, or, if they were at a distance from the "dead houses," they were wrapped in skins and blankets and placed on pole platforms high above the reach of animals, or in trees. With the dead were placed pipes, bowls, hammers, or such things as he made or might require to start life in the next world. Before the burial-house was placed a stone or wooden figure to guard the dead from evil spirits (Webber 1899:313).
The symbols and ideology associated with burial emphasizes important connections to the spirit world that the dead had. Suttles has documented that connections to the spirit world were established in mortuary ritual through the inclusion of grave goods and elaborate carved symbols on graves. (Suttles 1987 [1983]:127) (see Figure 2). Disposing of food and wealth items with the deceased provided the dead with goods needed in the spirit world. The symbols displayed on or near the tombs re-affirmed the claims high-class families made to the private knowledge and training necessary to have these successful relationships with the spirit world. Those who did not have such successful relations (and by extension were not members of the upper-class), received much less elaborate graves, often consisting only of a simple box (Barnett 1955:216; Elmendorf 1960:453).
The symbols represented in these burials perpetuate the ideology of social class. Suttles has quoted Fetzer's informant from Matsqui who said that "high-born families had `totem poles' called `sxwayxwoi'... set up at their graves, while other families had `totem poles' with 'other things'" (Suttles 1987 [1983]:129). The sxwayxwoi were associated with important spiritual powers and were owned by high-class families, privileges to use this symbol being inherited through the mother's line. Thus, high-class people have hereditary rights to create and display the sxwayxwoi in ceremony and with mortuary remains, while those of low-born families had access to a different, less prestigious set of symbols. Slaves, as Elmendorf describes, were often given very little ritual treatment at all: "[b]ox burial with the box coffin lashed in a tree was occasionally used for slaves" (Elmendorf 1960:453).
It is clear from this brief discussion of ethnographically documented Central Coast Salish mortuary practices that the social status of the deceased and social claims of decedents are being expressed in traditional Central Coast Salish funerals. Variation in the symbols used to reflect this relationship appears to be important in making status claims - particularly with some symbols being inherited privileges.
Moving from ethnographic observation to archaeological interpretation, I set out to evaluate the above model of the development of social class. I am concerned with the period from about 2000 to 1000 BP in the Gulf of Georgia area. Essentially, I am not convinced that the ascribed social inequality observed in the archaeological record by Burley and Knüsel (1989) for Marpole times was necessarily the same as the social classes which organized historic Central Coast Salish society. I am proposing here that the society in the Gulf of Georgia region during Marpole times had inequalities based primarily on a highly competitive system of social rank, which culminated with the use of burial mounds and cairns as grave markers in the period from 1500 to 1000 BP. Subsequent to that, the dramatic change in burial practices from below-ground burials to above-ground interments reflects a fundamental shift in the ideology of this society to one where social inequalities are based on social classes, similar to the inequalities known from the Central Coast Salish ethnographic record.
Archaeological Implications for the Development of Social Classes
It is one of the greatest challenges of archaeologists to move from discussions of ethnographically observable social phenomena to the material remains left in the archaeological record. Difficulties arise as archaeological interpretations are often built on scant and fragmentary patterns of material evidence which are constructed to support broader ethnographically observed patterns. I subscribe to the view held by Wylie and others (Wylie 1992) that although there are no "truths" which can be constructed from archaeological remains, our interpretations are limited or constrained by the observable patterns in the archaeological data. Using multiple lines of independent evidence to show support or critiques for a model, a certain amount of logical rigour is gained in order to back up claims made from the archaeological record. In the context of the present study, the model and suggestions made regarding the formation of social classes in the Gulf of Georgia region should certainly be subject to further examination and scrutiny as further lines of evidence are revealed and as ideas of the nature of social inequality develop and change.
Archaeological manifestations of mortuary ritual can provide useful information to aid in interpreting social organization and social inequality in the Gulf of Georgia region. Many of the social relations of gaining and maintaining social status and class are played out during mortuary rituals, and thus are a prime indicator of rank and class. Status claims made by the decedents of deceased elite individuals play an important role in how and in what form the burial ritual is carried out. These status claims are different in a rank and a class society. This relationship between the living and the dead has potentially meaningful archaeological ramifications.
The essential idea behind studying mortuary remains to interpret past social structures is that the ritual associated with burying the dead is meaningful to the living (Morris 1992). The variation that occurs between the mortuary rituals held for the dead reflect something of their social status during life. This relationship frequently manifests itself in a direct correlation between status and degree of elaboration in mortuary ritual (Tainter 1978). The mortuary ritual not only includes the actual interment of the body, but also includes ceremonies and events that occur before, during and after the deceased is buried.
Thus, the archaeological manifestations of mortuary ritual have often been used to interpret social inequality (Tainter 1978; Pearson et al 1989; Morris 1992; Kim 1994 among others). Although the material symbols used in this mortuary context may not have implicit, cross-culturally standardized meanings, examining them within a particular culture may often reveal how they are not arbitrary in relation to each other (Watson 1994:67). Binford (1971) has suggested on the basis of cross-cultural ethnographic observations that achieved and ascribed status may be inferred from differences in treatments of burials in a particular society. Achieved status may be interpreted from burials where people of the same social group are treated differently depending on age or gender. This inference is based on the assumption that older individuals have more opportunities to acquire status in their lives than do younger ones, and that this status will be accordingly reflected in burial. Gender plays into this interpretation as culturally defined gender roles define or limit how much status an individual may achieve. Ascribed or inherited status may be differentiated between social groups. In this case members of one social group, regardless of age or gender, receive similar treatment in burial which is distinct from that of another social group. The degree to which achieved and ascribed status is prevalent in a society is an important factor in determining the existence of social rank or social class.
Ian Morris has delineated several useful lines of evidence through which to interpret differential treatment of the dead in mortuary ritual (Morris 1992:24-26). The different lines of evidence - classification of graves; demography and deposition of individuals; spatial patterning of graves; patterns in grave goods; and temporal variation in mortuary practices - comprise a meaningful set of symbols. When these are taken in relation to each other, they provide insight into relations of social status.
Classification
One of the main distinctions in the treatment of the deceased in mortuary ritual is the class (not to be confused with "social class") of grave. Different classes of graves reflect variation in mortuary symbolism. Such variation may reflect different dimensions of social status, given that the variations are contemporaneous and occur within one social group. Individuals who gain social status through life achievements are likely to receive different degrees of elaboration in their graves, with the more elaborate graves being given to individuals of higher social status. Although there may be varying degrees of elaboration in mortuary symbols, the basic principles or structures of mortuary ritual is likely to be the same for all individuals in the society. This may also be associated with rank societies which generally have an ideology of minimizing perceived social differences between high- and low-ranking individuals (Collier 1994:121; Donald 1985). However, there are many things which contribute to an individuals "social persona", and this variation may reflect distinct social roles and not relative social status (Tainter 1978:331-332).
Demography and Deposition
Patterns in the demographic composition of a burial assemblage and the contexts of deposition of individuals in graves are important factors which help delineate social groups. In a rank society, the social position of individuals is often differentiated on the basis of gender, age, position in kin group and life experiences. These should be evident in a contemporaneous mortuary record of a rank society with men and women, as well as young and old people being treated differently in burials (Binford 1971; Morris 1992:26-27). Social status may be identified given the association of wealth symbols with individuals. If the individuals of all ages and both genders from a particular segment of a burial population have such items representing high social status, the status reflected may be largely inherited. Regular depositional patterns of individuals in graves may also provide evidence for the existence of individual social groups. If wealth or status symbols are regularly associated with particular burial assemblages, then some kind of social group with class-based inequalities may be inferred.
Spatial Patterning
Regional and local variation in spatial patterning of graves may also identify status differentiation between social groups and between individuals. Regional patterns of mortuary ritual reflects cultural beliefs about how the dead should be treated. Although we can never know the belief structures themselves, their patterning over a wide area may suggest how people interacted throughout the broad social network. Patterns of variation within the region may indicate different social groups. These groups may represent ethnic identity or may relate to the structure of social networks. The significance of these social groups is further definable through identifying local or site-specific spatial patterning. Variation of graves at one cemetery or burial place may reflect relative social status of individuals. Regular patterns in spatial patterning of graves can reinforce interpretations made from other levels of analysis, particularly when such patterning becomes redundant or repetitive (Watson 1994:101).
In a rank society, elaborate and unelaborate graves are likely to be found together at the same site, as the ideology of elites is to minimize social distance between themselves and their followers. However, a class society is more likely to separate the graves of the upper- and lower-classes, re-emphasizing the distinctiveness of these two social groups. The location of high and low status burials alone will also not accurately discriminate between the types of social inequality, as other cultural and ideological factors may be present.
Grave Goods
Pearson et al. (1989) have pointed out that patterns in both the quantities and classes of grave goods associated with the different burials often provide a measure for social stratification between individuals in contemporaneous interments. The frequency of occurrence of any individual class of grave good may reflect something of the relative social status of the individual buried within that grave good. However, raw numbers of goods may not provide a useful measure because different types of objects have different cultural values. For example, 7,000 stone beads can not be quantitatively valued against a single, rare copper ornament. Thus, examining the relative frequency of a particular class of object and the association of that class of object with other burial attributes (such as gender, age or method of burial), may provide further interpretive insights into the value of an object, and their significance for showing social status in a burial.
Temporal Variation
Finally, the change of any combination of these factors over time may indicate a change in the way social inequality was manifested in the society. As Morris has pointed out, "static patterns cannot be interpreted easily. We need to identify points at which patterns change, and the precise nature of the changes" (Morris 1992:25). Documenting the cultural context of the changes in mortuary ritual pattern will provide further insight into the social transformations that occurred.
These test implications provide the basis from which the analysis moves from the archaeological record to the social model being discussed. Detailed analysis of related contemporaneous aspects of the archaeological record are beyond the scope of this study, but must be mentioned to provide context to the mortuary evidence. Patterns of hostilities and violence, inter-regional interactions, and changes in economy will be briefly discussed in the context of the data collected to provide further independent lines of evidence from which to evaluate the proposed model.
The first step in evaluating the ideas I have presented is to describe the nature of the burials found in the Gulf of Georgia region during the Marpole phase. The excavations at the Scowlitz site provide the most recent description of burial mounds and cairns in this region. Moving from this detailed description of burial practices to those which are less adequately described, I present the first full description and analysis of mound and cairn burial data from the Gulf of Georgia region to date (Appendix I; Table A)
Burial Mounds and Cairns: The Scowlitz Site
The Scowlitz site (DhRl 16) is located at the confluence of the Harrison and Fraser Rivers (Figure 3). It is currently known for its important spring salmon runs (Albert McHalsie 1995 personal communication), and was historically the base of an important transportation route from the Fraser River valley to the Lillooet River valley. At the site itself, there are 42 visible burial features found along the back of a terrace, lying behind a number of house depressions (Figure 3). At the time of writing, the 1995 U.B.C. Archaeological Field School have documented a number of other burial mounds and cairns on a terrace above the main part of the site, to the north of the site behind DhRl 15, and to the south of the site on DhRl 25 (Michael Blake personal communication 1995).
The burial features date to between 1600 and 1100 BP, about 1,000 years more recent than the date of 2460 + 90 BP reported from the lowest deposits of one of the house depressions (Matson 1994). The cairns and mounds at the Scowlitz site vary in form and size, ranging from a very large mound, to a series of lower, but still prominent mounds, to a number of small, discrete cairns. Three of these burial features, the largest mound, a mid-sized mound and a rock cairn, have been investigated since 1992 by researchers from the University of British Columbia in conjunction with members of the Stó:lô Tribal Council (now called the Stó:lô Nation) (Morrison and Myles 1992; Morrison 1995; Blake, Coupland & Thom 1993b; Blake 1995).
Mound 1 (Figure 4a) is a large and elaborate burial feature. The bulk of the mound is made up of compact, silty sand extracted from a talus slope above the site. This earth is piled up in a roughly rectangular feature measuring 12 m on each side at the base. The mound reaches a maximum height of 3 meters from the lowest point of the surrounding slope. Figure 4b shows the internal structure of the mound as revealed by the excavation of five 2 x 2 m units. A square alignment of carefully arranged boulders surrounds the perimeter of the mound. A second concentric rock alignment is found about 1 m nearer the centre. A 1.5 m high pillar of boulders marks the southeast corner of the second rock alignment. At the centre of the mound is a solid mass of approximately 500 boulders. All of these rocks sit on a thin layer of grey clay which truncates at the edge of the mound fill.
Beneath this inner rock cairn is a shallow grave pit which contains the well preserved remains of an adult male, lying on his right side, flexed, head oriented south-southeast and facing east (Figure 4c). No pathologies were indicated on the bones, and the cranium was clearly deformed. The grave goods associated with this burial were impressive. Four cold-hammered, perforated copper disks with preserved knotted leather thongs were found near the head along with a copper ring, four abalone shell pendants and approximately 7000 sectioned dentalium shell beads. The human remains from this burial were re-buried the same day they were excavated, under the guidance of Stó:lô Elders and the artifacts remains in the U.B.C. Laboratory of Archaeology.
Several radio-carbon dates were obtained for this mound. The two most reliable are an AMS date from a fragment of human bone dating 1370 + 60 BP (Beta 56200), and a standard C14 date on charcoal from an area near the cranium, dating to 1410 + 80 BP (Beta 56199) (Table B). These dates are uncalibrated and if taken to only one standard deviation, overlap considerably. A less secure date of 620 + 110 (Beta-56201) was obtained from the burial pit on likely intrusive wood or root fibres which were found underlying the human remains.
Mound 23, a mid-sized burial feature (Figure 5a) is somewhat less elaborate than Mound 1. The base of this mound is roughly square in shape, approximately 10 m on each side and 2 m high. Four adjacent 2 x 2 m units were excavated from the edge through to the centre of the mound revealing similar but less complex internal architecture to Mound 1. A mixed fill of dark organic soil laden with fire-cracked rock and midden debris makes up the bulk of the earthen deposits. The mound is constructed over older occupation deposits, which were revealed in test excavations below the burial pit. The soil that makes the fill of the mound was undoubtedly taken from these midden deposits. A roughly rectangular formation of "rubble" consisting of compact pebbles and cobbles defines the perimeter of the mound. Beneath the mound fill is a thin layer of compact yellow clay which, like the grey clay in Mound 1, terminates at its edges.
At the centre of the mound lies an oval-shaped cairn made up of approximately forty boulders, seen in Figure 5b. These boulders cover a shallow burial pit where no traces of human remains or grave goods are preserved. This mound was also rebuilt upon completion of the excavation, with the assistance and guidance of Stó:lô community members and Elders.
Several radiocarbon dates were assayed, including two samples of charcoal which come from just below the yellow clay floor, beneath the burial cairn stones. These dates are reported at 1130 + 80 BP (Beta 56204) and 1190 + 70 BP (Beta 56217) (Table B) and may represent some kind of ritual burning activity that occurred before or after the body was placed in the burial pit. However, there is no further corroborating evidence for this and it is also possible that the charcoal may represent an earlier occupation surface of the site (Blake, Coupland and Thom 1993a:8). An older date of 2200 + 160 BP (Beta 57218) was obtained on charcoal found well below the burial in a cultural layer likely associated with earlier residential activity. A date of 520 + 100 BP (Beta 56202) was obtained on a large sample of charcoal found under the compact rubble wall, but above the yellow clay floor. Although I have accepted the first two dates as the most likely to represent the date for the burial of Mound 23, it is possible that the later date more accurately reflects the age of this deposit. Given the nearness of the charcoal which dated to 520 + 100 BP to the outer surface of the mound, there remains a real possibility of this being intrusive organic material, although there is no clear stratigraphic evidence for this.
The burial cairn (with the field designation Mound 20) is made up of a circular pile of several hundred cobbles and boulders. The cairn measures 3 m in diameter and the cobbles and boulders are piled up to a meter thick in the deepest part of the cairn, near its centre (Figure 6a). The cairn boulders are aligned on top of midden deposits which form the older habitation component of the site. A very shallow burial pit was found underlying these rocks at the centre of the cairn (Figure 6b). Excavation of the burial pit revealed the very highly decayed enamel of seven deciduous teeth from a child estimated to be just under 6 years old. No radiocarbon dates are available.
The dating of the features requires some further discussion. The two most reliable dates - 1370 + 60 BP for Mound 1; and 1190 + 80 for Mound 23 - are not, in all likelihood, contemporaneous. If these dates are taken to two standard deviations it is possible that they are temporally separated by up to about 500 years. If the very late date of 520 + 100 BP for Mound 23 is correct, the two burials are separated by almost a millennium. This suggests that the site was used as a cemetery for several generations, with similar kinds of burials being made over this long span of time. Further excavation of other mounds and dating material from Mound 20 would provide more insights into this problem of chronology. The dates from other deposits at the site - particularly the 2200 + 160 BP for the cultural layers below Mound 23 and the date of 2460 + 80 for the house depression excavated by Matson (Matson 1994) suggests that the site was used as a settlement about one thousand years earlier than its use as a cemetery. The use of a former village site as a cemetery, and the midden deposits used as mound fill in at least two of the burials, may signify some important symbolic meaning. Cybulski has observed this for the Greenville site on the Nass River and suggests that midden is a culturally appropriate place for the dead (Cybulski 1992:166-169). Carlson has recently suggested that the association of burials with midden deposits shows some cultural continuity with the ideology expressed in the contemporary Central Coast Salish practice of "burnings", where prepared food is burnt in a ritual which feeds ancestor spirits (Carlson 1995; Duff 1952:94).
It is also notable that not all the burial features have necessarily been discovered. As I have described elsewhere (Thom 1994), the remains of a possible second cairn were found 70 cm below the surface, underlying Mound 20, which may indicate more burial cairns could lie below, undetected to surface inspection. Furthermore, only mounds and cairns (no unelaborate pit burials) have been found, although excavations have not been extensive enough to imply that the later do not exist.
Although the Scowlitz site provides an excellent model for understanding the nature and structure of burial mounds and cairns, a much broader and more significant sample of burial mounds and cairns from different sites in the Gulf of Georgia region is more effective for analysis. Appendix I provides a description of the history of mound and cairn excavation, from the earliest investigations in the 1850's to the most recent contract projects completed in 1994. Although understanding this history is not critical to this discussion, my main argument, it does provide some insights into the limits of the data set being analyzed, and gives a basis for future analysis of related or new research questions.
Analysis of Mortuary Ritual
In researching the literature on burial practices in the Gulf of Georgia region, I came across a total of 157 individual excavated burial mounds and cairns for which I am able to present data in some detail (Table A). A total of 108 separate sites where mounds and cairns were surveyed (Figure 7; Table C). Only a small percentage of these are completely described in archaeological reports and even fewer of them are dated (Table B). Although this data set is not perfect, it is an exhaustive compilation of the data from British Columbia, and provides significant information, if its limitations are kept in mind.
Classification
Every recent discussion of prehistoric burial practices of the Gulf of Georgia region has noted that simple midden interments dominate the archaeological record until the Late period. At this time, between 1500 and 1000 years ago, below-ground interments disappear from the archaeological record and it is assumed that the above-ground burials noted in the ethnographic record (Figure 2) become the dominant burial practice (Cybulski 1994; Moss 1993; Matson and Coupland 1994; Burley and Knüsel 1989). The later is very difficult to detect archaeologically as no human remains are placed in the ground. Cybulski has recently debunked the idea that the "scattered human remains" frequently found in shell middens are the remains of above ground burials, arguing that they are almost always disturbed elements from below-ground interments (Cybulski 1992:39-44). Burley and Knüsel have noted that for the vast majority of Marpole and earlier burials, there are "strong tendencies toward adult shallow pit interments without grave goods" (Burley & Knüsel 1989:4). This is certainly the case for 84% of the Marpole and earlier graves listed in their study. However the remaining 16% of the graves fall into a number of different categories - some of which include cairn burials, rock-lined pits, burial mounds, slab burials, and wooden boxes (Burley & Knüsel 1989; Hill 1992). This ratio is important, as it is likely that the unusual, non-midden pit burials represent a distinct set of beliefs - possibly reflecting social status.
For this analysis I initially only documented what had been described as "mounds" and "cairns". I quickly realized that the same criteria was not used by all investigators to describe these features. To make distinctions more clear, I have devised a series of classes based partially on descriptive elements and partially on a clustering of size attributes.
Below, I have divided these burials into two broad groups - mounds and cairns - based largely on the presence or absence of earth covering a stone cyst or cairn. A mound is defined broadly as an earth-covered cairn, while a cairn is a burial with significant rock associations. I have further divided the mounds into three rather intuitive groups of based largely on size. The cairns may be more discretely divided based on their size and quantity of rocks used to construct them. Figure 8 is a plot of width vs. length showing a small cluster of "minor cairns" with diameters below 1.5 m, and a larger cluster of "full cairns" all having diameters larger than 1.5 m. This loose clustering of sizes may not be convincing alone, but when taken with descriptions of the graves as "particularly well built" or "poorly constructed" (cf. Smith & Fowke 1901), many can be classified into the groups defined above. The other kinds of "cairns", including rock-lined pits, large rock associations (slab burials) and small rock associations, clearly exhibit different symbols and (as in the case of TW-D16-1989 or GB-1-1968) may have very different meanings given their relative lack of elaboration. However, these may be thought of as occurring within the larger "theme" of graves with significant rock associations.
Burial Mounds
Elaborate Mound (M+): This class includes Mound 1 from Scowlitz (SW-1-1992) and Class 4 and 5 mounds described by Hill-Tout (1895) (HZ-1, 2, 3-1894). These are larger than 6 m in diameter and have multiple concentric stone rings enclosing a central cairn. All the internal features are then covered with several meters of piled earth. These mounds are the most elaborate of any below-ground interment ever constructed by Aboriginal people in the Gulf of Georgia area.
Mound (M): Mound 23 from Scowlitz (SW-23-1992) is a good example of this class, being between 3 m and 6 m in diameter, having a central cairn, often with some kind of enclosing stone ring, and covered with earth. These mounds are similar to (M+), but are generally smaller with less elaborate internal structures.
Minor Mound (M-): This class takes into account single burials found beneath a cairn which were subsequently covered with earth. These are usually less than 3 m in diameter and have no elaborate stone enclosures associated with them. The burial described by Richardson (1871) at Little Qualicum River (LQ-1-1872), and the mounds at Cadboro Bay (CB-17-1898, CB-18-1898, CB-19-1898) described by Smith & Fowke (1901) provide examples of this class.
Burial Cairns
Full Cairn (C+): A large burial feature, often visible from the surface, with hundreds of large boulders and cobbles completely covering and surrounding a shallow pit which contains the burial. These are greater than 1.5 m in length or width. This type is occasionally associated with a number of stone enclosures, much like in an elaborate mound, but not covered with earth. Mound 20 from Scowlitz (Thom 1994), and many of the cairns excavated by Smith and Fowke (ie: NS-3-1898, NS-9-1898, NS-11-1898, and NS-12-1898 among others) are typical examples of this class.
Minor Cairn (C -): This class is composed of cairns made with numerous cobbles and small boulders surrounding and covering the burial. These are not as massive as a "Full Cairn", measuring 1.5 m or less in length or width. Smith and Fowke (1901) provide descriptions of cairns in this class including NS-1-1898, NS-2-1898, NS-4-1898 and NS-7-1898 among others.
Rock-Lined Pit (RLP): This class is defined by a pit which has been covered with cobbles and large pebbles, over which the body is placed. Occasionally, more such rocks are piled on the body. The size of these graves varies dramatically from less than 1 m to over 3 m in diameter. The burials excavated by Smith and Fowke (1901) at Point Roberts (PR-1-1898, PR-2-1898) as well as some of those excavated by Arcas (1991) at the Tsawwassen site (TW-D15a-1989, TW-D16-1989 and TW-D26-1989) provide examples of this class.
Large Rock Association (LRA): This class of cairn is defined by a grave pit with fewer than ten massive boulders or rock slabs covering all of or a portion of the burial, frequently only the head area. Mitchell (1971) has described burials of this class at Montague Harbour (MH-4-1963 and MH-9-1963).
Small Rock Association (SRA): Small concentrations of small cobbles covering only a portion of the burial defines this class. I have recorded only a few burials in this class, and given some of the variation in their antiquity, there is some question whether they represent the cairn tradition. Haggarty and Sendy (1976) describe a possible SRA at Georgeson Bay (GB-1-1968), and Monks (1977) describes one at Deep Bay (DB-9-1977).
It is not clear from these classifications alone what the differences may indicate and reflect about social status. It is clear, however, that there is a lot of variation in how mounds and cairns were constructed. This suggests that these grave classes do not form discrete groups, as may be expected in a society with social classes. This is particularly so if the burial mounds are considered as a kind of "extended cairn" with earth heaped over top of essentially an elaborate rock cairn feature. The variation within these grave types suggests that people burying their dead were making status distinctions with more elaborate graves reflecting higher status. This interpretation of the relative status of individuals by analysis of the type alone is not strictly convincing. This is due to the uncertain temporal association of many of these burials, and the varying patterns of burial in different areas of the Gulf of Georgia region. These issues are developed further below.
Demography & Deposition
The analysis of demography provided here is based on excavated mound and cairn burials where gender (Figure 9) and age (Figure 10) information was recorded. I have relied on the accuracy of reporting of the various authors for descriptions of the human remains and have not made physical observations myself. For purposes of analysis, I have summarized age categories into five groups: infant (0-2); child (2-15); sub-adult (15-25); adult (25-50); old (50+). Physical descriptions of the majority of the individuals is not published, and so the resulting set of data is rather small, between 63 and 73 individual cases. However, these numbers are not insignificant and do reveal interesting patterns if these limitations are kept in mind.
In the mound and cairns where both age and gender is reported, all age categories and both genders are found in cairns (Figure 9, n=73), while child, adult and old are found in mounds (see Table A). If my assertion that mound and cairns are the burials of high status individuals (being more elaborate than the majority of midden interments which occur at this time), this pattern would suggest that some measure of ascribed social status is being shown. This is not surprising or unique, as ascribed status has been suggested for the Gulf of Georgia region by many other authors for the Marpole phase (Mitchell 1971; Matson 1976, 1985; Burley 1980, 1988; and Burley and Knüsel 1989 among others).
The occurrence of males and females in cairns varies over different age categories (Figure 10, n=63). For three age groups (infant, sub-adult & old), the ratio of males to females is virtually 1:1. For children, no specific identifications of gender has been made. For adults this ratio increases to 3.3:1 in favour of men. This may indicate that adult males are involved in achieved status competition, or are honoured for their status in their adult lives almost twice as often as females. Upon reaching old age, both men and women again have equal ability to obtain this status in their communities and are represented equally in cairns. Taken with the ascribed status patterns shown in the data overall, this pattern is suggestive of my earlier model of rank society. In a rank society status is gained differentially on the basis of many attributes including age and gender and to a lesser degree, through family connections.
Although the data is too limited to make definitive conclusions about age representation in mounds and cairns, it should be noted that only adults and old individuals occur in elaborate mounds (M+), and that adults dominate the burial assemblage of the other mounds and cairns. This pattern lends further support to the earlier suggestion that although some status is certainly being ascribed at birth, status is also being achieved throughout a lifetime, with the greatest levels of status being obtained by adult males who are frequently represented in elaborate mounds and cairns.
The occurrence of cranial deformation is another trait in burials which may reveal information about inherited status. The assumption behind linking cranial deformation with status is that the practice would be restricted to elite families with privileges to deform an infant's head. Cranial deformation is done by binding the malleable head of an infant with boards and cedar bark pads (Cybulski 1990:52-53). The data obtained was on the whole not detailed enough to indicate the kind of cranial deformation, only its presence or absence. Of the total number of cases where data was obtained for both gender and cranial type (n=41) 50% of the males, and 53% of the females exhibited this trait. This is very similar to the general pattern of the Gulf of Georgia burial assemblage where 51% of the burials are found with cranial deformation (Hill 1992:36). Thus, overall cranial deformation does not occur in any clear pattern associated with elaborate burials and is not exclusive to any one burial type. If mound and cairn burials reflect high status individuals, cranial deformation does not appear to be a useful indicator of status during Marpole times. Jenness has noted ethnographically that during the historic period, the heads of the children of nobles, commoners and slaves were all flattened (Jenness n.d.:59).
The context of deposition in burial mounds and cairns does not appear to follow a particular set of rules. As Pinart observed (1876), there seems to be no particular direction which the bodies face. In the data presented in Table A, 10 face north, 16 face west, 19 face south, 1 faces east, 4 face southeast, 5 face southwest, 3 face northeast, 2 face up and 6 face down. The right side is preferred almost over the left with 31 of the burials on their right; 16 on their left. One of the bodies in a cairn was burnt or cremated, as were three individuals in mounds. There were also three cairns where two people were buried together. In two instances, this was a female and child, and in the other it was two adults, one male and the other unidentified. This summary shows that there is a great deal of variation between individual interments, and that, like the variety of types of cairns and mounds, there is much flexibility in how the bodies are placed within the graves. These contexts of deposition do not appear to co-vary with other burial elements. Such variation reaffirms my previous suggestion that although these burials follow a general "theme" of being elaborate. They are not standardized to the extent that they might be expected in a society with rigid social classes.
Spatial Patterning
The spatial patterning of burial mounds and cairns on both a regional and local level provides further evidence for the importance of achieved status being represented in these graves. This patterning is also suggestive of a number of other issues, including ethnic identity and possible claims to resource locations. There are, however, a number of limitations to these interpretations, most of which are due to the quality of the data set.
To look at the regional pattern of mound and cairn burials, I have compiled site survey and excavation data for burial mounds and cairns in the Canadian portion of the Gulf of Georgia region (Table C; Figure 7). Much of this data comes from site surveys conducted since the 1970's and is now recorded on the Canadian Heritage Information Network (CHIN 1994), although some information is found in early reports and from my own personal observations. This data was not systematically collected, nor is it by any means complete. In fact, despite their notoriety in the archaeological literature at the turn of the century (see Appendix I), archaeologists conducting surveys in the past 20 years have often overlooked mounds. Mounds are easy to overlook in the forested landscape of much of the Gulf of Georgia region and have been some of the first archaeological victims of urbanization. The map in Figure 7 shows the distribution of these mounds and cairns over southwestern British Columbia. This search did not include records for the state of Washington, and thus this area is virtually unrepresented. However, as Reagan (1917) has described, these features have also been found in some abundance south of the 49th parallel. Given these limitations, some general patterns can be observed on the regional level.
Generally speaking, cairns are more common in the contemporary Straits Salish speaking areas on the Gulf Islands and Saanich Peninsula (Suttles 1990:454). The converse is true for mounds which are most common in the present-day Halkomelem speaking areas, although neither area is mutually exclusive of either class of burial. Point Roberts, which has been used variously by a number of different groups in the region (Suttles 1951) has a notable abundance of rock-lined pits. It is tempting to suggest that these general patterns may reflect something of the antiquity of the ethnicity of the contemporary First Nations groups in the region, however the evidence is not conclusive.
On a more local level, the location of mounds and cairns tends to be on promontories and peninsulas overlooking important bodies of water. Grant Keddie (1984) has previously suggested an association between burial cairns and "defensive earthworks". Certainly the dates provided by Moss and Erlandson (1992) for defensive earthworks in the Gulf of Georgia region establish their contemporaneity with burial mounds and cairns - around 1200 years old. However, Cybulski has observed that during Marpole times there are very few incidents of death due to violence - only 6.2% of all bodies observed (Cybulski 1994:83). This is not a sign of a society under seige. This evidence of the association of mounds and cairns with defensive features and low incidences of violent death would not be inconsistent with slave raiding reported historically in this region (Suttles 1990:465). Both Suttles (1958) and Donald (1985) have argued that the presence of slaves is a strong indicator of social class being present in that society. If slave raiding did occur at this time, and if slaves did have the same productive importance then as they did in historic times, then this may be evidence for the presence of social classes begin during the time when burial mounds and cairns were the usual burial of social elites. However, further evidence challenges this suggestion. Cybulski notes that during the Late period, incidences of violence increased to 27.6% on the Northwest Coast as a whole. If these defensive works were used as protection against slave raids (and Buxton (1969) has questioned whether defensive works provided a military function at all), and the incidence of violence was so much lower 2500 to 1500 years ago, then it may be reasonable to suggest that slaves (if there were slaves) had much less economic importance during these times.
At the site-specific level, a number of observations can be made about the data from well described sites such as Scowlitz. Mound 1 (M+) is located on the highest point of the terrace, somewhat removed from and overlooking the other mounds. The apparent lack of burial features between this and the other interments may be distorted by the fact that an orchard and a small shack existed in the early 20th century in this area of the site, which may have disturbed any burials that were there (John Williams 1992, personal communication). The remaining somewhat smaller burial mounds (M) are surrounded by smaller cairns (C and C+)and low, earth covered mounds (M-). Only having excavated three of the 42 mound features, it is difficult to discern any particular demographic or temporal patterning associated with the layout of these features. It is, however, interesting that a child was found in the burial cairn (Mound 20) located adjacent to Mound 23. If the individual in Mound 23 was an adult, there may have been some kind of family connection to the infant in Mound 20.
The occurrence of these many different grave types together at the same location leads to a number of different possible interpretations. If the burials are all contemporaneous, it is possible that they reflect different social rankings of individuals according to the degree of elaboration of each grave. Thus an individual in a larger, more elaborate grave would have been more highly recognized than that on an individual buried in a less elaborate grave. This interpretation is complicated by the fact that the burials at one site are often not contemporaneous (such as North Saanich and Scowlitz). As Cannon (1989) has pointed out, the meaning of the relative amount of elaboration of the grave often changes over time, but relative status differences in the society may remain the same. The occurrence of different grave types at the same site may also reflect different age or gender distinctions. Further testing of these burial features may reveal more clear patterns which would provide stronger evidence to support an interpretation of relative social rank being shown in graves. This may have been the case at Scowlitz where an adult male was found in an elaborate mound (Mound 1) and a child in the cairn (Mound 20). However, adult males and females are found in cairns at North Saanich, similar to the child's grave at Scowlitz. Thus, although relative status may be reflected at one site, regional patterns suggest that the situation was somewhat more complex.
Data from the Tsawwassen, Hill and Deep Bay sites show that contemporary cairn and pit burials are found together in the same cemetery (Arcas 1991; Hall & Haggarty 1981; Monks 1977). Since archaeological data is poor on the existence of separate cemeteries for high-status individuals and low-status individuals (ie: cemeteries with either exclusively mounds and cairns or exclusively simple midden pit burials), it is difficult to come to any firm conclusions about the existence of social classes through this line of evidence. The preliminary evidence from these sites indicates that social class was not established, or at least not reflected in spatial patterning of cemeteries. Both high- and low-status people could be buried at the same site. The spatial evidence from these sites thus generally supports the interpretation of social rank being exhibited in mortuary remains.
Grave Goods
Grave goods are likely grossly under-represented due to the problem of preservation of wooden objects in the acidic soils of the Gulf of Georgia region. The presence of some grave goods, however does reveal a number of interesting patterns. Table 1 summarizes the kinds of grave goods found in the mounds and cairns listed in Table A, the overwhelming majority being items associated ethnographically with social status and wealth (Suttles 1987 [1960]:22). Non-utilitarian objects which were often not locally available such as dentalium shells, abalone shell pendants and various types of copper objects reinforce the importance of traded items for obtaining wealth. Objects such as the composite harpoon valve may represent the importance of non-staple (non-salmon/ shellfish) food items such as sea mammals. Sharing these kinds of foods at feasts was important for gaining prestige ethnographically (Suttles 1951). Whole, large clam shells which were found in several graves may have been used as a bowl, possibly associated with feeding the dead (Carlson 1995). Patterns of association of these different types of grave goods with other mortuary dimensions again reveals a pattern consistent with rank societies.
The number of occurrences of grave goods in burial reveal that 17% of males and females in mounds and cairns received these burial inclusions (n=27). The graph shown in Figure 11 breaks this presence/absence data down further by age group for cairns (n=65 individuals). Of the different age categories, children and sub-adults are found more often with grave goods than individuals who are younger or older. This pattern is suggestive of the importance of ascribed status among some individuals. This presence/absence data may be deceptive, however, as the problem of preservation has likely skewed the overall patterns.
Pearson et al (1989) have suggested that looking for patterns in the kinds of burial goods may be more revealing than presence/absence analysis. Using Suttles' (1987 [1960]:22) criteria for wealth items (Table 1), I have graphed the occurrence of different classes of goods in mounds and cairns for different age groups. The pattern depicts that wealth items are important in the graves of all age groups except children. The occurrence of these goods in graves of sub-adults and infants is further suggestive of the use of wealth items to reflect inherited or ascribed status. However, that children in mounds and cairns do not have grave goods associated with them indicates the variability in exhibiting inherited status in through grave wealth. Recalling the description of rank and class presented earlier, ascribed status is found in both instances. Ascribed status is less important in societies where social status is manifested by social ranking. Thus, the overall importance of inherited status in the mound and cairn burials should not be over-emphasized on the basis of grave goods alone, particularly as so few of these elaborate graves are found with inclusions.
Grave goods are more revealing for interpreting gender inequalities, if the limitations of the data are kept in mind. Although many of the individuals with burial goods have not had their genders determined, women appear to be uniquely associated with utilitarian grave goods in both mounds and cairns (Figure 12). Men, on the other hand, are exclusively found with wealth items. This evidence is suggestive of the more prominent role elite women may have had in domestic production than their male partners. The presence of wealth items reinforces the notion that women were able to obtain and acquire high social status on their own (Thom 1993). Further identification of the genders of individuals buried in mounds and cairns would shed more light on this issue.
The similarities and importance of copper objects in mounds and cairns is striking. Almost identical copper disks were found in Mound 1 at Scowlitz, the mound excavated by Hill-Tout and Lazenby at Hatzic (Hill-Tout 1895; HZ-1-1894; Figure 13), and in a cairn at North Saanich (NS-17-1898; Smith & Fowke 1901:65). These objects were all cold-hammered from yet undetermined native copper sources (local copper mines are currently located on the east coast of Vancouver Island, and along the lower reaches of the Thompson River) . Similarities in such wealth items over a broad area highlights the regional nature of interactions between elites and the importance of wealth items in trade.
The presence of wealth items in graves is not restricted to mound and cairn burials. One example of this has been described by Burley (1988:59) where a midden burial with very elaborate grave goods at the False Narrows site was found. Midden burials with elaborate grave goods, as well as mound and cairn burials having no wealth items, provides further support to the suggestion that wealth was not restricted to a certain class defined in one burial type. Rather, many individuals clearly competed for status and exhibited this in different kinds of graves. It is difficult to say what the social relationships may have been between an individual buried in a cairn with no grave goods and a contemporaneous individual buried in a simple pit with elaborate grave goods. Overall, the mix of utilitarian and wealth items in graves suggests that status of an individual at death resulted from both ascribed and achieved status - a feature of rank societies.
Temporal Variation
Finally, patterns in change over time must be examined. Table B lists the mound and cairn burials which have been radiocarbon-dated or have been associated with a particular time period on the basis of stratigraphic association. The issue of dating is somewhat difficult as only 15 radiocarbon dates are associated with mounds and cairns (Figure 14). A further 28 burials have been given a temporal association from their general stratigraphic context (Figure 15). The burials which have reliable radiocarbon dates associated with them generally fall between 1500 and 1000 years BP. The ages of the burials that have been assigned a date on the basis of stratigraphic associations are somewhat problematic.
The data in Figure 14, show that the majority (10 out of 15) of the directly dated mounds and cairns were created between 1500 and 1000 BP. The first early exception was the mound Capes excavated at Mission Hill in Comox (Capes 1964; MHM-3-1959). The charcoal that was dated to 1980 + 70 BP was taken from mound fill, was likely re-deposited from earlier cultural deposits during mound construction. The other early exception is a small rock-lined pit dating to 3800 + 60 BP found at Tsawwassen (Arcas 1991; TW-D16-1989) which, given the distinct nature of rock-lined pits, may be a part of an earlier, unrelated mortuary tradition. Two of the three dates which occur after 1000 BP (HZ-2-1894, DB-1-1977, HH-1-1994) also may not reflect the true age of the burials they are associated with. The Hatzic mound (HZ-1-1894) has an AMS date of 840 + 60 BP from wood fibres preserved on a copper disk grave good (Figure 13). This disk was collected by Hill-Tout and Lazenby in 1894 (Hill-Tout 1895) and has experienced 100 years of various means of storage. It is not certain if these fibres were from a wooden or bark object in the grave itself, or if they were from later root intrusion, like those suspected in Mound 1 at Scowlitz. The cairn at Deep Bay (DB-1-1977) was dated from a piece of charcoal found directly behind the cranium. This charcoal may not date the burial feature itself (Monks 1977:63). The date of 770 + 60 BP (corrected to 955 + 60 BP - see Appendix I) on the burial from the Harbour House site (HH-1-1994) is an AMS date obtained from human bone collagen and has been accepted with confidence (Arcas 1994:78).
On the basis of the many cairn and mound burials firmly dated to 1500 - 1000 BP, I suggest that the Marpole (2500-1500 BP) ages associated with the mounds and cairns shown in Figure 15 should be reconsidered. Although the Marpole phase is generally defined as extending from 2500 to 1500 BP (Fladmark 1982:113), the ending point has been very unclear (Matson and Coupland 1994:218, Burley 1980). Matson has suggested that the Marpole phase be divided into three sub-phases which he has distinguished through a statistical multi-dimensional analysis of artifact assemblages (Matson et al 1980). The latest sub-phase, named "Garrison", was interpreted to be the transitional time period between the Marpole and Late phases. Such a transitional phase has also been suggested (although in a different context) by Borden (1970), and more recently by Arcas in their study of the Tsawwassen site (Arcas 1991). I propose that the Garrison sub-phase, extends from 1500 to 1000 BP, may be tentatively confirmed on the basis of the occurrence of mound and cairn burials dating to this time period.
The transition from the Garrison sub-phase to the Late phase may reflect the change from a society with social status based on rank to a one based on class. By the beginning of the Late period (1000 BP to 200 BP), below-ground midden interments no longer occur in the archaeological record (Cybulski 1992, 1994; Burley and Knüsel 1989). The change of mortuary ritual from midden interments to above-ground interments may have been the final stage in establishing social classes, through the control of symbols used in mortuary ritual.
Ian Morris (1987) (building on D. Miller's (1985) initial suggestion) has argued that elite people gain control of and manipulate symbols which are used to reinforce their status and prestige. When too many people have access to the symbols that represent high-status, elites may change the nature of the symbol used to affirm their position. This is illustrated in Figure 16, with a series of "stick-people" that represent members of different social status groups. On the right are chubby elites who initially have a particular symbol associated exclusively with them. As time goes on, a wider range of people gain access to this symbol. When people of much lower social status (skinny sad stick people) use the symbol, they being to challenge the ideology of the elites in a more material sense. Elite leaders, conscious of these social threats, have the ability to change what symbols can be used to represent high status. Thus in the Gulf of Georgia region around 1000 BP, as more and more people competed for high social rank, the display of certain prestigious symbols became wide-spread. Elaborate, permanent and visible burial markers such as cairns and mounds were initially important to promote visible kin connections to resource rights. As these rights became more exclusive, particularly through inter-marriage of elites, the symbols used to express kin connections and social status would be changed. When elaborate mortuary rituals were changed from mounds and cairns to above-ground grave, only those people with the ability (for instance) to hire artisans to produce new symbols to connect people to the spirit world, were able to make high status claims.
Clearly, many more cairns and mounds are seen in Garrison times than earlier in the Marpole phase. When certain powerful heads of families attempted to solidify their control of access to wealth, high-status burial symbols were changed. This change occurs at the beginning of the Late period where above ground burials in elaborate boxes (Figure 2) are constructed for entire high-class families, while individual, below-ground interments were abandoned. Special people with particularly skilled artistic abilities were required to create these elaborate boxes. The icons depicted on these new mortuary symbols emphasize the connection with the spirit world. These connections, as Suttles has described (Suttles 1958), are often restricted by private knowledge of high-class families. Those families who do not have access to wealth are not able to create these kinds of symbols and have to use a lower-class grave marker. This is in contrast to the large groups of followers and supporters which would have been required to construct elaborate burial mounds and cairns. Restricted private knowledge and powers may not have been necessary to create these mounds and cairns, while demonstrated leadership ability would have been. Of course, the manipulation of symbols by a new upper-class does not occur in mortuary ritual alone. Rather it is dependent on the control of trade and ownership of resources by a regional network of related elites.
In sum, the changes in mortuary ritual over time tend to support the idea that social classes emerged when mound and cairn mortuary practices were abandoned. Most of the patterns observed in the data - including classes of burials, demography and context of deposition of individuals in the graves, spatial patterning of interments, and variation in grave goods - support the notion of a rank society, where achieved status is the most important, but not the exclusive factor in relations of inequality. The proposition that social class emerged at the end of the Garrison sub-phase is supported by the evidence of the manipulation of mortuary symbols by elite families establishing new symbols to demonstrate their social position.
The patterns of mortuary ritual discussed above do not conclusively demonstrate a shift from social rank to social class. However, they do not reject the model completely, and where evidence is good enough, tend to support it. Given a tentative acceptance of the development of social classes in Garrison and early Late times around 1000 BP, I now proceed to put these ideas and observations into the context of the larger culture history of the region.
Implications for Regional Culture History
Along with the changes in mortuary ritual described above, there were a number of other cultural transformations that occurred between the Marpole and Late phases. The evidence of these other changes is consistent with the idea that social classes emerged during the period from 1500 to 1000 BP. Non-toggling harpoons were the most common form of sea-mammal hunting technology in Marpole times, while composite-toggling harpoons were prevalent during the Late phase (Mitchell & Donald 1988:318). The Late phase is noted for the profusion of single- and bi-pointed bone objects, thought to be associated with fishing technologies such as trolling, bentwood hooks and herring rakes (Mitchell & Donald 1988:318). This shift in sea-mammal hunting and fishing tool types likely reflects a change in productive relations, with increasing specialization of non-staple or prestige hunting techniques and more elaborate organization of labour in fishing. These changes in productive relations may have concentrated wealth in the hands of fewer families than in previous ages and contributed to the development of social classes.
The notions of migration and diffusion have been suggested at various times to account for the changes between Marpole and Late phases. Borden at one time thought that a great population movement marked the onset of the Late (or Stselax) phase, based partially on the above noted changes of toggling harpoons and a general increase in bone tools (cf. Thompson 1978:15-16). As suggested above, these changes may be better explained by the changes in social and productive relations associated with establishment of social classes in the Gulf of Georgia region. Burley (1980:39) and Charlton (1977:192) have noted the occurrence of some "interior-like" and other "outer-coast-like" artifacts in the Gulf of Georgia region during the Late phase. Burley has attributed this to "differing diffusion streams" into the Gulf of Georgia area (Burley 1980:39). It seems likely that these similarities in artifact styles over broad areas may be due to the emergent networks of elite, inter-married families that formed the upper-class in Central Coast Salish societies. Such networks were historically connected through the idiom of kinship over a broad geographic region. Thus, the ideas of migration and diffusion as the main impetus for culture changes between Marpole and Late phases may be more subtly interpreted as a change in social relations during the transitional Garrison sub-phase.
Finally, Matson has suggested that the development of the Late from Marpole phase was a largely adaptive process, with technological and subsistence changes being far less drastic than those seen in previous periods (Matson 1992:389). He notes that the Garrison "subphase shows many similarities with the Late period, supporting an evolutionary development" in the Gulf of Georgia region (Matson 1992:389). This view does not contradict the mortuary evidence presented here, as Matson has argued mainly from an economic standpoint. Although the broad ecological adaptations in technology appear to be somewhat minor, those changes in subsistence production that do occur are largely consistent with those that could be expected of social transformation from a rank to a class society.
Conclusions and Suggestions for Further Research
This analysis of burial mounds and cairns in the Gulf of Georgia region generally supports the hypothesis that Marpole phase society had ranked status differences, while late phase society was differentiated by social classes. The time period from 1500 to 1000 BP was one where elites completed for social status which was gained mainly through life achievements and somewhat less by inheritance. Adult men were particularly involved in this status competition, although women, elderly and young people also took part. As access to owned resources became more critical to gaining the wealth to achieve prestige, networks of inter-married families formed over a wider region. These families gained control of wealth and formed the historically known Central Coast Salish upper-class. Lower class individuals provided labour in complex fishing operations. Ritual specialists (such as carvers and sea-mammal hunters) gained social and economic status through using their skills and inherited privileges. Slavery likely began to increase the wealth and status of upper-class individuals.
This model for the transformation of social life from the Marpole to the Late phase begs for further investigation and archaeological testing. Future research should be conducted to evaluate these conclusions through more independent lines of evidence. Observations in changes of settlement patterns would provide further insights into issues of resource ownership, productive relations and community structure. Social status differences may be observed through archaeological examinations of residential sites. The relationship of these sites to grave years may also provide further information on the nature of social groups and networks. An investigation of the amount, source and occurrence of traded items (such as shells, copper, jade and obsidian) should be conducted to further evaluate when and where inter-regional interaction became important. Further evidence to the timing of the emergence of social classes could be found by investigating archaeological indicators of slavery, possibly through a study of prehistoric violence or trauma on skeletal remains. A detailed study of all artifact assemblages from the 1500 to 1000 BP period may also provide insight into the nature of socio-economic changes.
Additional investigations along the lines conducted in this thesis would be particularly useful in addressing specific concerns raised in this study. Such investigations should include a re-evaluation of the dating of all classes of mound and cairn burial. Emphasis should be placed on obtaining reliable dates for burial features to clarify changes in mortuary practices. Such future research will be invaluable to further our understanding of the changes in social organization and inequalities in the Gulf of Georgia region over the past 2000 years.
Bibliography
Amoss, Pamela (1978) Coast Salish Spirit Dancing: The Survival of an Ancestral Religion. University of Washington Press, Seattle.
Arcas Consulting (1991) Archaeological Investigations at Tsawwassen, B.C. Arcas Consulting Ltd, Burnaby.
- (1994) Archaeological Investigations at the Harbour House Site (DfRu 3), Saltspring Island, B.C. Arcas Consulting Ltd., Burnaby.
Bancroft, Hubert Howe (1875) The Native Races of The Pacific States of North America. Volume IV. Antiquities. D. Appleton and Company, New York.
Barnett, Homer G. (1938) The Coast Salish of Canada. American Anthropologist. 40:118-141.
- (1955) The Coast Salish of British Columbia. University of Oregon Press, Eugene.
Binford, Lewis (1971) Mortuary Practices: Their Study and Their Potential. In Approaches to the Social Dimensions of Mortuary Practices. J. A. Brown, ed. Pp. 6-29. Memoirs of the Society for American Archaeology, No. 25. New York.
Blake, Michael (1995) Interim Report on the 1992 Excavations at the Scowlitz Site (DhRl 16). Interim Report Prepared for the Archaeology Branch, Victoria and the Stó:lô Nation, Sardis.
Blake, Michael, Gary Coupland and Brian Thom (1993a) Dating the Scowlitz Site. The Midden. 25(1):7-9.
- (1993b) Burial Mound Excavations at the Scowlitz Site, British Columbia. Paper presented at the Society for American Archaeology Meetings, April 14-18.
Boas, Franz (1894) Indian Tribes of the Lower Fraser River. In 64th Report of the British Association for the Advancement of Science for 1890. Pp. 454-463.
Borden, Charles (1970) Culture History of the Fraser-Delta Region: An Outline. Archaeology in British Columbia, New Discoveries. Roy Carlson, ed. BC Studies 6-7:95-112.
Burley, David V. (1980) Marpole: Anthropological Reconstructions of a Prehistoric Northwest Coast Culture Type. Simon Fraser Univerisy Archaeology Press, Burnaby.
- (1988) Senewelets: Culture History of the Nanaimo Coast Salish and the False Narrows Midden. Memoir No. 2. Royal British Columbia Museum, Victoria.
Burley, David and C. Knüsel (1989) Burial Patterns and Archaeological Interpretation: Problems in the Recognition of Ranked Society in the Coast Salish Region. In Development of Hunting-Fishing-Gathering Maritime Societies Along the West Coast of North America. A. Blukis-Onat,ed. Reprint Proceedings, the Circum-Pacific Prehistory Conference, Seattle.
Buxton, Judith (1969) Earthworks of Southern British Columbia. Master's Thesis, Department of Archaeology, University of Calgary, Calgary.
Caldwell, Warren (1955) Two Cyst Burials From San Juan Island, Washington. American Antiquity. 20(4):383-384.
Cannon, Aubrey (1989) The Historical Dimension in Mortuary Expressions of Status and Sentiment. Current Anthropology. 30(4):437-458.
Carlson, Roy (1995) Reading the Burial Record at the Pender Canal Site. Paper presented at the 28th Annual Canadian Archaeological Association Meetings, Kelowna B.C.
Capes, Katherine H. (1964) Contributions to the Prehistory of Vancouver Island. Occasional Papers of the Idaho State University Museum,, Number 15. Pocatello. (1993) Letter to the Editor. The Midden. 24(5):1.
Charlton, Arthur (1977) The Belcarra Park Site. Department of Archaeology, Simon Fraser University Archaeology Press, Number 9. Burnaby.
Canadian Heritage Information Network (CHIN) (1994) Canadian Heritage Information Network Records for Burial Mounds and Cairns known in the "DR" Borden Grid of British Columbia. Unpublished computer printout in possession of author.
Collier, George (1994) Basta! Land and the Zapatista Rebellion is Chiapas. Institute for Food and Development Policy. Monroe, Oregon.
Condrashoff, Nancy (1984) DcRu 52: A Salvaged Burial Cairn in Songhees Territory. Ms. on file, Archaeology Branch, Victoria.
Cybulski, Jerome (1990) Human Biology. In Handbook of North American Indians, Volume 7, Northwest Coast. Wayne Suttles, ed. Pp. 52-59. Smithsonian Institution Press, Washington.
- (1992) A Greenville Burial Ground: Human Remains and Mortuary Elements in British Columbia Coast Prehistory. Anthropological Survey of Canada Mercury Series Paper No. 146. Canadian Museum of Civilization, Hull.
- (1994) Culture Change, Demographic History, and Health and Disease on the Northwest Coast. In In the Wake of Contact: Biological Responses to Conquest. Pp. 75-85.
Deans, J. (1891) Burial Mounds of Vancouver Island. American Antiquarian and Oriental Journal. 13:171-172.
- (1892) Antiquities of British Columbia. American Antiquarian and Oriental Journal. 14:41-44.
Donald, Leland (1985) On the Possibility of Social Class in Societies Based on Extractive Subsistence. In Status, Structure and Stratification: Current Archaeological Reconstructions. M. Thompson et al., eds. Pp. 237-244. University of Calgary Archaeological Association, Calgary.
- (1990) Liberty, equality, fraternity: Was the Indian really egalitarian? In The Invented Indian: Cultural Fictions and Government Policies. James Clifton, ed. Pp. 145-167. Transaction Publishers, New Brunswick, N.J.
Drucker, Philip (1939) Rank, Wealth, and Kinship in Northwest Coast Society. American Anthropologist. 41(1):55-65.
- (1983) Ecology and Political Organization on the Northwest Coast of America. In The Development of Political Organization in Native North America, Proceedings of the American Ethnological Society, 1979. Elisabeth Tooker, ed. Pp. 86-96. J.J. Augustin, New York.
Duff, Wilson (1952) The Upper Stalo Indians of the Fraser Valley, British Columbia. Anthropology in British Columbia. Memoir No. 1. British Columbia Provincial Museum, Victoria.
Elmendorf, William (1960) The Structure of Twana Culture. Washington State University Press, Pullman.
Fladmark, Knut (1982) An Introduction to the prehistory of British Columbia. Canadian Journal of Archaeology. 6:95-156.
Freid, Morton (1967) The Evolution of Political Society: An Essay in Political Anthropology. Random House, New York.
Gailey, Christine (1987) Kinship to Kingship: Gender Hierarchy and State Formation in the Tongan Islands. University of Texas Press, Austin.
Haggarty, James and John Sendy (1976) Test Excavation at the Georgeson Bay Site, Gulf of Georgia Region, British Columbia. Occasional Papers of the British Columbia Provincial Museum, Number 19. Victoria.
Hall, R. L. and J. C. Haggarty (1981) Human Skeletal Remains and Associated Cultural Material from the Hill Site, DfRu 4, Saltspring Island, British Columbia. In Contributions to Physical Anthropology 1979-1980. J. Cybulski, ed. Pp. 64-106. Mercury Series, No. 106, National Museums of Canada, Ottawa.
Hill, Warren (1992) Analytical Approaches to Identifying Social Complexity from Northwest Coast Burials. Ms. on file. University of British Columbia Laboratory of Archaeology, Vancouver.
Hill-Tout, Charles (1895) Later Prehistoric Man in British Columbia. Transactions of the Royal Society of Canada. Series 2:103-122.
- (1899) Notes of the Prehistoric Races of British Columbia and Their Monuments. British Columbia Mining Record. Pp. 7-23.
- (1905) The Salish Tribes of the Coast and Lower Fraser Delta. Ontario Provincial Museum Annual Archaeological Report. 12:225-235.
- (1930) Prehistoric Burial Mounds in British Columbia. Museum and Art Notes. 5(4), Art, Historical and Scientific Association of Vancouver, B.C.
- (1933) Monuments of the Past in British Columbia. Vancouver City Museum, Vancouver, B.C.
Jenness, Diamond (n.d.) The Saanich Indians of Vancouver Island. Ms. No. VII-G-8M in Canadian Ethnology Service Archives, Canadian Museum of Civilization, Ottawa.
Keddie, Grant (1984) Fortified Defensive Sites and Burial Cairns of the Songhees Indians. The Midden. 14(4):7-9.
Kim, Seung-Og (1994) Burials, Pigs, and Political Prestige in Neolithic China. Current Anthropology. 35(2):119-141.
Lazenby, Frederick T. (1898) Notes on the Form and Contents of Four Circular Barrows Opened in the Summer of the Year 1894. Ms. on file. American Museum of Natural History, New York.
McGuire, Randall (1992) A Marxist Archaeology. Academic Press, New York.
McMurdo, J. (1974) The Archaeology of Helen Point, Mayne Island. Master's Thesis, Simon Fraser University, Burnaby.
Matson, R.G. (1976) The Glenrose Cannery Site. Mercury Series Archaeological Survey Paper No. 52. National Museum of Man,Ottawa.
- (1983) Intensification and the Development of Cultural Complexity: The Northwest vs. the Northeast Coast. In Evolution of Maritime Cultures on the Northeast and the Northwest Coasts of America. Pp. 125-148. R. J. Nash, ed. Simon Fraser University, Department of Archaeology Publication 11, Burnaby.
- (1985) The Relationship Between Sedentism and Status Inequalities among Hunters and Gatherers. In Status, Structure and Stratification: Current Archaeological Reconstructions. M. Thompson et al., eds. University of Calgary Archaeological Association, Calgary.
- (1992) The Evolution of Northwest Coast Subsistence. In Research in Economic Anthropology, Supplement 6. B. Isaac, ed. Pp. 367-428. JAI Press Inc, Greenwich, Connecticut.
- (1994) Report on the 1993 Excavations at Scowlitz. Report on File, Provincial Archaeology Branch, Victoria.
Matson, R.G., Deanna Ludowicz and William Boyd (1980) Excavations at Beach Grove in 1980. Report on File, Provincial Archaeology Branch, Victoria.
Matson, R.G. and Gary Coupland (1994) The Prehistory of the Northwest Coast. Academic Press Inc, San Diego.
Menzies, T.P.O. (1933) Curator's Report. Vancouver City Museum and Art Gallery. May 1933. Ms Typescript, Vancouver Museum Archives.
- (1938) Northwest Coast Middens. American Antiquity. 3(4):359-61.
- (1949) The Great Fraser Midden. Publication of the Vancouver Art and Historical Association, Vancouver.
Miller, Bruce G. (1993) Women and Politics: Comparative Evidence from the Northwest Coast. Ethnology. 31(4):367-383.
Miller, Daniel (1985) Artefacts as Categories: A Study of Ceramic Variability in Central India. Cambridge University Press, Cambridge.
Milliken, A.C. (1961) Excavation of Stone Mound: Yale Creek Bridge. Ms. Typescript on file, Archaeology Branch, Victoria.
Mitchell, Donald (1971) Archaeology of the Gulf of Georgia Area, a Natural Region and its Culture Types. Syesis. Volume 4, Supplement 1.
Mitchell, Donald and Leland Donald (1988) Archaeology and the Study of Northwest Coast Economies. In Research in Economic Anthropology, Supplement 3, B. Isaac, ed. JAI Press Inc, Greenwich, Connecticut.
Monks, Greg ( 1977) An Examination of Relationships between Artifact Classes and Food Resource Remains at Deep Bay, DiSe 7. Doctoral Dissertation, Department of Anthropology and Sociology, University of British Columbia, Vancouver.
Morris, Ian (1987) Burial and Ancient Society: The Rise of the Greek City-State. Cambridge University Press, New York.
- (1992) Death-Ritual and Social Structure in Classical Antiquity. Cambridge University Press, New York.
Morrison, Sandra and Heather Myles (1992) The Sacred Mounds of Scowlitz. The Midden. 24(4):2-4.
Morrison, Sandra (1995) Scowlitz Dig Holds More Than Archaeological Remains. Sqwélqwel te Stó:lô, Stó:lô Nation Newsletter. 1(2):9.
Moss, Madonna (1993) Gender, Social Inequality, and Cultural Complexity: Northwest Coast Women in Prehistory. Paper presented at the 26th Annual Chacmool Conference, University of Calgary, November 11-14, 1993.
Moss, Madonna and Jon Erlandson (1992) Forts, Refuge Rocks, and Defensive Sites: The Antiquity of Warfare along the North Pacific Coast of North America. Arctic Anthropology. 29(2):73-90.
Oetteking, Bruno (1915) Preliminary Remarks on the Skeletal Material Collected by the Jesup Expedition. In Proceedings of the 19th International Congress of Americanists. Pp. 621-624. Washington.
- (1922) On Morphological Changes in Artificially Deformed Skulls from the North Pacific Coast. In Proceedings of the 20th International Congress of Americanists. 2:25-95. Rio de Janerio.
- (1926) Craniology of the Northwest Coast of North America. Proceedings of the 22nd International Congress of Americanists. 1:412-425. Rome.
- (1930) Craniology of the North Pacific Coast. Publications of the Jesup North Pacific Expedition, 11(1); Memoirs of the American Museum of Natural History 15(1). New York.
Pearson, R., J. Lee, W. Koh, A. Underhill (1989) Social Ranking in the Kingdom on Old Silla, Korea: Analysis of Burials. Journal of Anthropological Archaeology. 8(1):1-50.
Percy, R. (1975) The Prehistoric Cultural Sequence At Crescent Beach. Master's Thesis, Department of Archaeology, Simon Fraser University, Burnaby.
Pickford, A. E. (1944) Prehistoric Cairns and Mounds in British Columbia with a Report on the Duncan Burial Mound. British Columbia Historical Quarterly. 11(4):237- 264.
Pinart, Alphonse (1876) A French Scientist Explores the Indian Mounds of the Pacific Coast. Victoria Colonist. Sept 1, 1876. Pp. 3.
Reagan, A. B. (1917) Archaeological Notes on Western Washington and Adjacent British Columbia. Proceedings of the California Academy of Sciences, 4th Series. 7(1):1-31.
Richardson, James ( 1871) Field Notes. Unpublished Field Notes from the Geological Survey of Canada. National Archives of Canada, RG 45, Vol 171, File 1515 and 1518.
Smith, Harlan I. (1899) Archaeological Investigations on the North Pacific Coast of America. Science. 9:(224)535-539.
- (1900) The Cairns of British Columbia and Washington. Proceedings of the American Association for the Advancement of Science. 4(1):12-14.
- (1903) Shell-Heaps of the Lower Fraser River, British Columbia. The Jesup North Pacific Expedition: Memoir of the American Museum of Natural History. Edited by Franz Boaz. Vol. II - Part IV. New York.
- (1904) The Cairns or Stone Sepulchres of British Columbia, and Washington. Records of the Past. 3(8):243-254.
- (1907) Archaeology of the Gulf of Georgia and Puget Sound. The Jesup North Pacific Expedition: Memoir of the American Museum of Natural History. Edited By Franz Boaz. Vol. II - Part VI. New York.
Smith, Harlan I. and Gerald Fowke (1901) Cairns of British Columbia and Washington. The Jesup North Pacific Expedition: Memoir of the American Museum of Natural History. Vol. II - Part II. New York.
Suttles, Wayne (1951) The Economic Life of the Coast Salish of Haro and Rosario Straits. Doctoral Dissertation, Department of Anthropology, University of Washington.
- (1958) Private Knowledge, Morality, and Social Classes among the Coast Salish. American Anthropologist. 60:497-507.
- (1960) Affinal Ties, Subsistence, and Prestige among the Coast Salish. American Anthropologist. 62:296-305.
- (1973) CA* comment on Ruyle. Current Anthropology. 15:622-623.
- (1983) Productivity and Its Constraints: A Coast Salish Case. In Indian Art Traditions of the Northwest Coast. R. Carlson, ed. Pp. 67-87. Simon Fraser University Archaeology Press, Burnaby.
- (1987) Coast Salish Essays. Talonbooks and University of Washington Press. Vancouver.
- (1990) Central Coast Salish. In Handbook of North American Indians, Volume 7, The Northwest Coast. W. Suttles, ed. Pp. 485-502. Smithsonian Institution Press, Washington.
Suttles, Wayne and Aldona Joanites (1990) History of Research in Ethnology. In Handbook of North American Indians, Volume 7, The Northwest Coast. W. Suttles, ed. Pp. 73-87. Smithsonian Institution Press, Washington.
Tainter, Joseph (1978) Mortuary Practices and the Study of Prehistoric Social Systems. In Advances in Archaeological Method and Theory. Vol. 1. Edited by M.B. Schiffer, Pp. 105-141. Academic Press, New York.
Thom, Brian (1993) The Political and Economic Position of Chinook and Salish Women: The Ethnohistoric and Anthropological Records. Paper presented at the 46th Northwest Anthropology Conference, Bellingham.
- (1994) Excavations of Mound 20 - Burial Cairn. In Report on the 1993 Excavations at Scowlitz, DhRl 16. R.G. Matson, ed. Final Report prepared for Archaeology Branch, Victoria and the Stó:lô Tribal Council, Sardis.
Thompson, Gail (1978) Prehistoric Settlement Changes in the Southern Northwest Coast: A Functional Approach. University of Washington Department of Anthropology Reports in Archaeology Number 5. Seattle, Washington.
Watson, Paul (1994) The Archaeology of Rank. Cambridge University Press, Cambridge.
Webber, Ellen (1899) An Old Kwanthum Village - Its People and Its Fall. American Antiquarian and Oriental Journal. 21:309-314.
Wilson, Ian R. (1987) Excavation of a Burial Cairn DcRu 17, Piers Island, B.C. Permit 1987-17. Ms. on file B.C. Heritage Conservation Branch, Victoria.
Wylie, Allison (1992) The Interplay of Evidential Constraints and Political Interests: Recent Archaeological Research on Gender. American Antiquity. 57(1):15-35.
Appendix I - History and Description of Cairn and Mound Excavation
The description provided is intended to show the nature and extent of these kinds of burials, as well as to indicate the limits of this particular set of data. The first section deals with burials which are described but not dated and the second section with dated remains.
Early History of Mound and Cairn Excavation: Descriptions
In the winter of 1871-72, the Victoria Colonist published a series of articles by Mr. James Deans, who had been interested in exploring burial mounds and cairns since 1853 (also discussed in Bancroft 1875). He explored a total of five mounds and an unreported number of cairns, all located near Victoria on Vancouver Island between the years 1859 and 1871. Deans' account of mound excavation is more detailed than that of cairns. He described the remains in these features as being "cremated" (Deans 1892:43). Three of the burial mounds excavated contained the remains of one adult male and two adult females, one in each mound. The remaining two, much smaller mounds, were thought to be those of children. Burial goods accompanied only the women's burials; a wooden ornament in one and a piece of quartz in the other. The three mounds containing the remains of adults had a single stone border ring at the perimeter base of the mound, and a stone cyst at the centre. These mounds also had a fine sand layer located immediately above a shallow burial pit, and several layers of ash adjacent to the body. It is uncertain where the collections made from this first mound excavation currently reside, and dating information is as yet unavailable.
Three other local amateurs commented on their investigations into burial mounds between 1859 and 1876. Mr. Robert Homfray is reported by the Victoria Colonist to have opened a cairn at Florence Lake near Esquimalt (Pickford 1944:244) in 1859, but no description of the remains could be located. In 1862, Dr. Charles Forbes, a Royal Navy medical doctor, wrote an essay describing the natural features of Vancouver Island (Deans 1892:41). In it he commented on the cairns which were "scattered in irregular groups...crowning the rounded promontories, over all the southeastern end of Vancouver Island" (Deans 1892:41), and draws a parallel between them and the burial cairns of ancient England. No further description is given.
The French naturalist Alphonse Pinart described his excavations of some "sepulchral cairns" on Vancouver Island in Victoria Colonist drawing a connection between these "and the celebrated cairns of Scotland, Wales, &tc." (Pinart 1876). Pinart, however, acknowledged that there was in important connection between these cairns and the present day native populations: "...the skulls and bones found in the cairns opened by Mr. James Deans & myself belong undoubtedly to the Indians," and goes on to compare the very similar osteological morphology of the people buried in the cairns with the "Cowitchan Indians" (Pinart 1876). He estimates the ages of the cairns at "six or seven or more centuries", which is, surprisingly, not too far off the actual dates (Pinart 1876)(see Table B). He was a careful observer, noting:
I paid formerly a good deal of attention to the way in which the bodies were laid by the Indians; but I have found that there is no particular rule in which they do it; & I have been obliged to give it up altogether as an ethnological sign. (Pinart 1876)
Finally, Pinart notes the location of these burials on Vancouver Island as being "generally ... [in] large ditches enclosing in their area quite an extent of ground, cutting off sometimes from the mainland - a small promontory and forming in that was a fortified enclosure, where the natives could defend themselves from the incursions of war-like tribes...".
In 1871, the Geological Survey of Canada sent Mr. Alfred Selwyn and Mr. James Richardson to conduct preliminary explorations of the geological resources of British Columbia. During these explorations Richardson, with the help of Deans explored some mounds and cairns on Vancouver Island Richardson's unpublished field notebooks (Richardson 1871) record nine burial mounds, the widths of eighty-five "stone piles" or cairns at a site "in the Vicinity of Victoria", and the relative location of eight cairns on Beacon Hill. Richardson records having excavated the mounds at Hornby Island "by the means of a pointed stick using my hands as a shovel" (Richardson 1871), yet was able to provide a general cross-section of the largest. It appears that all the mounds share the same essential quality of having stones piled on top of shallow grave pits and then covered with earth. No grave goods were found in any, and the bones were burnt in the mounds at Comox.
Charles Hill-Tout provided the first widely published descriptions of archaeological explorations of burial mounds or cairns (Hill-Tout 1985). With the assistance of Frederick T. Lazenby (of Retford, England), Hill-Tout explored at least three burial mounds near the town of Hatzic in the Fraser Valley and one at Port Hammond. From these explorations, a sequence of five different types of mounds are classified by Hill-Tout.
Series 1 mounds are a simple, circular mound of dirt covering a shallow burial pit, with no grave goods or stone architecture. Series 2 are similar to series 1, but have dirt covering a cairn of boulders over the burial pit. Series 3 mounds have the same essential features as series 2, but also have a stratum of charcoal "extending over the whole area of the mounds between the boulders and the outer covering of clay" (Hill-Tout 1895:118). The mounds in Series 4 are typified by large boulders lined up in the form of a rectangle around the base of the mound. At the centre of the mound a cairn covers the burial. Many different layers are found between the cairn and the surface of the mound, most notably a yellow "quicksand" layer just above the burial. Series 5 mounds are the largest in size and contain three evenly spaced, concentric squares of boulders lined up around a central cairn.
By far the most extensive investigations of burial mounds and cairns was done by Harlan I. Smith and Gerald Fowke for the Jesup Expedition, under the direction of Franz Boas (Smith and Fowke 1901). A total of six burial mounds and eighty-eight cairns were fully reported by Smith and Fowke, with several other cairns having been excavated but only briefly mentioned in the reports. Three of these burial mounds reported were the ones excavated by Hill-Tout at Hatzic and two others were excavated some 60 miles down-river at Port Hammond. Neither of the later mounds contained any artifacts or stone architecture. Both, however, did contain what was described as a layer of "vegetable mould" at the bottom centre of the mound (Smith and Fowke 1901:60), which may have been decaying wooden material from a grave box. Charcoal, ash and burned clay were found at the bottom centre of one of the Port Hammond mounds. A complete skeleton, found lying fully extended on its back was found in the other. Fully extended burials are rare in this region, this being one of the very few examples found. The skull exhibited cranial deformation.
The sixth mound described by Smith and Fowke was excavated at Cadboro Bay. It was the only mound found among a great many cairns. This mound had a central cairn, under which was found a fully flexed skeleton, lying on its left side. No grave goods were associated with it.
The majority of the burial remains described by Smith and Fowke are cairns. The following summary is taken from Smith's Archaeology of the Gulf of Georgia and Puget Sound (1907):
In general, the cairns consist of irregular piles of bowlders, from three to twenty feet in diameter. One cairn is found over each body, which in some cases is surrounded by a more or less rectangular cyst, formed by placing the straight sides of several bowlders towards it, and covering the opening thus formed with one, two, or more slab-shaped rocks...The skeletons [when found] are sometimes burned, but the evidences of fire in the cairns do not seem sufficient to conclude that the bodies were burned in the vaults. The skulls show a variety of types and deformations...The cairns are always near [and sometimes in] coast shell-heaps...
As seen in Table A, few of these cairns contained grave goods, while many of the skeletons were preserved and are now housed in the American Museum of Natural History.
In a series of publications coming out of the last stages of research for the Jesup North Pacific Expedition, Bruno Oetteking reported on the osteological remains of all the skeletons in the collections of the American Museum of Natural History (Oetteking 1915; 1922; 1926; esp. 1930). Particularly useful were his determinations of sex and age and cranial deformation for some of the individuals collected by Smith in 1897-1899 (Oetteking 1930, Table of Measurements: 18-28, 64-74). This data has been integrated into Table A.
Albert Reagan published a paper in 1917 recording his 1905-1907 explorations of mounds along the Quillayute River, at Neah Bay, at Lummi, in the Fraser Delta, on Georgian Bay, in southeastern Vancouver Island, and on the mountain slopes overlooking Sumas Lake (Reagan 1917). These mounds are not included in Table C because their locations were not adequately described. Reagan did not find any mortuary remains in his excavations. His findings were limited to charcoal, ash, and loose piles of stones inside the earthen mounds. This led him to the interpretation that they were "oven mounds" or cooking features rather than burials. In the light of previous and present excavations of burial mounds in this region, I would suspect this interpretation is unfounded and that any burial remains that were left in the mounds had decayed. The ash and charcoal layers likely resulted from the same (possibly ritual) processes which created similar layers in the other mounds described here.
In 1930 T.P.O. Menzies, the curator of the Vancouver Museum, undertook excavations at the Marpole site in Vancouver. In two identical albeit fanciful reports (Menzies 1938; 1949), he describes the finding of a cairn buried four feet deep in the Marpole midden. At the top of this cairn:
...the head of a large stone image appeared...The image was seated on the apex of a pyramid so built that its four corners faced the cardinal points of the compass. The four sides of the pyramid were faced with rounded mountain-stream stones about the size of polo balls, and the inside was packed with bright orange-coloured sand, in which fragments of human bones were found. (Menzies 1949:16-17; 1938:360-361)
However, the stratigraphic association of the seated human figure bowl (illustrated in the 1949 report) is questionable given its context within the midden.
Menzies and Hill-Tout excavated a burial mound at the Pretty site at Harrison Mills in 1933, providing a report which is highly speculative and lacks thoroughness (Hill-Tout 1933, Menzies 1933). The accuracy of the cross-section of this mound provided in Hill-Tout (1933) is highly questionable, particularly given Menzies' taste for embellishment.
A large burial mound, located on a low hill at Cowichan Bay, was excavated by A. E. Pickford in the summer of 1944 (Pickford 1944). This mound has a rectangular stone ring at the bottom perimeter of the mound, and a stone cairn at the centre. No human remains were found, but a layer of charcoal and ash was detected at the surface level of the mound. A layer of fine earth extended under the cairn. In the spring of 1994, Doug Brown and Michael Blake of the University of British Columbia recovered some dateable paleo-botanical remains from the layers directly below the outer set of boulders of this mound. No report or analysis of this material has been prepared to date.
Warren Caldwell published a very short report on the excavation of two cairns on San Juan Island (Caldwell 1955). This report discussed the location of these cairns as on peninsulas of land overlooking saltwater bays, but provides little other detail.
In 1961, amateur archaeologist A.C. Milliken excavated a very rectangular "stone mound" or cairn at Yale Creek Bridge, which contained no grave goods or skeletal remains (Milliken 1961). Little description of this feature was reported.
Recent History of Mound and Cairn Excavation: Description and Dating
Since the 1960s, there have been a number of excavations done by professional archaeologists which have uncovered burial mounds and cairns. None of these investigations have made the discussion of mounds or cairns the focus of their study. Below I provide a description of the features found at each site, and discuss the dates that the burials are associated with. This information is also summarized in Table B.
DcRu 52, Victoria
Skeletal material from this burial was directly radiocarbon dated to 640 + 440 BP. The large standard error on this date (taken to 2 sigmas would put it into the future) put the security of this date in serious doubt.
The cairn itself was large and visible from the surface, composed of many small cobbles and very large boulders (Keddie 1984:8)
False Narrows Site
"A single radiocarbon date of A.D. 240 + 90 [1710 + 90 BP] is associated with False Narrows II, the late Marpole transitional component" (Burley 1988:63). The burial found in this component, FN-49-1967, was excavated into the lower reaches of this component, and should date between 1890-1500 BP, however the exact age remains unclear. Burial FN-41-1967 was found in the False Narrows I deposits which "on the basis of diagnostic artifact types...is interpreted to fall in a mid-Marpole time range between circa 100 B.C. and A.D. 100. [2150-1850 BP]" (Burley 1988:36).
Burial FN-41-1967 was covered two large sandstone slabs, while burial FN-49-1967 consists of "several rocks" which covered the bodies (Burley 1988:55-56).
Montague Harbour Site
Burial MH-6-1964 was located in the Montague Harbour I component, which has the dates of 940 B.C. + 140 [2890 + 140 BP] and 1210 B.C. + 130 [3160 + 130 BP] (Mitchell 1971:221) and should date between 3420 - 2600 BP. Mitchell notes that Burials MH-4-1964 and MH-7-1964 likely belongs to the MH III component as they were "shallow, dug into the surface of the MH II midden, and covered by a cairn" (Mitchell 1971:179). Component MH III has the dates of A.D. 1160 + 130 [790 + 130] and A.D. 1220 + 130 [730 + 130] associated with it, providing 1050-470 BP as a range of dates for these burials. MH-9-1964 clearly belongs to the Montague Harbour II component which was not dated. None of these dates are directly associated with the burials.
Two large rocks are associated with MH-4-1964, a distinct cairn of large beach cobbles and waterworn slabs of sandstone covered MH-7-1964, 1 slab of sandstone was placed over the head of MH-9-1964, and 2 boulders surrounded by several cobbles covered the body of MH-6-1964.
Hill Site
No radiocarbon dates were obtained for the deposits at the Hill site, but "on the basis of burial patterns and selected artifact class" Hall and Haggarty estimate a range of dates from "400 B.C. - A.D. 300" [2350-1650 BP] (Hall & Haggarty 1981:64). Burial HL-2-1973, a large cairn burial visible from the surface, comes from between 8 and 21 meters away from the other burials which were taken from excavations below the surface, and thus may not be of the same expected age.
Burials 1, 4, 10 and 16 are all associated with numerous rocks and boulders, and were excavated from midden deposits. Burials number 3 and 7 are noted as "probable" cairn burials, due to their association with large rocks, but are not "certain" because of back-hoe disturbance. Hall and Haggarty mention burials 6, 11, and 18 may be cairn burials because they are located on bedrock, however, no evidence of boulders was found so I have not included these graves in my analysis.
Tsawwassen Site
Several radiocarbon dates were made directly from skeletal material: TW-D26-1989 dated to 1150 + 60 BP; TW-D16-1989 was dated 3800 + 60 BP; TW-D40-1989 dated to 1160 + 50 BP. The remaining burials are placed in the range of 1150 - 1670 BP range on the basis of other dates of burials in similar stratigraphic units (Arcas 1991:6).
TW-C8-1989 was found "in a shallow, rock-lined pit which was capped by a hearth". (Arcas 1991:159). TW-D15a-1989, TW-D16-1989 and TW-D26-1989 were found rock-lined pits, although the first two were disturbed by back-hoe (Arcas 1991:215). TW-D22-1989 was found "placed in an oval pit and covered by a single layer of boulders and a large rock slab" (Arcas 1991:227). A cairn composed of "large cobbles and small boulders" covered TW-D34b-1989, which was placed in a shallow pit (Arcas 1991:249). TW-D40-1989 was placed "in a wooden box, which was placed in a shallow pit and covered with a large boulder" (Arcas 1991:259). Finally, TW-D52-1989 was damaged by back-hoe, but a rock cairn of "at least 7 boulders and numerous cobbles" were recorded (Arcas 1991:275).
Helen Point Site
Both burials from the Helen Point site are given their dates by their stratigraphic association of the Helen Point Ib component. This component was dated on the basis of comparison with other similar dated assemblages and is assigned to the period of 3300-2800 BP (McMurdo 1974:iv).
Burial HP-2-1968 "appears to have been contained within a pit" with a sandstone slab located to the south of the cranium (McMurdo 1974:128). HP-3-1968 had a "large flat slab of granite" over the head.
Crescent Beach Site
A date range of 3170 - 2500 BP was assigned CB-8-1971 on its stratigraphic association with component II, which had a Locarno Beach aged radiocarbon date of 2980 + 80 BP made on some charcoal.
The single cairn burial from the Crescent Beach site, CB-8-1971, was "marked by the placement of an irregular feature of stones about 15 cm above the bones" (Percy 1975).
Deep Bay Site
Burial DB-1-1977 was dated by direct association with charcoal which dated to 790 + 80 BP. The burials DB-3, 4, 5, 8, 9-1977 should date between 1080-630 BP on the basis of dates above and below their associated stratum. Burial DB-7-1977 was found in a stratigraphic layer which lay between two strata dating to 1910 + 110 and 2630 + 100 BP, and thus was likely buried between 2830 - 1690 BP.
The burials "were placed in a shallow pit then covered with a cairn", excepting DB-9-1977 which was "placed in a grave excavated into natural stratum P ... [and] surrounded by moderate sized cobbles as if the grave perimeter had been lined with them (Monks 1977:365). Photographs included in the report indicate these cairns generally consist of several large boulders and many cobbles placed directly over the body.
Mission Hill Mounds Site
A carbon date obtained from a sample of charcoal from the upper level of burial mound MHM-III-1959 dates to 1980+/-70 years BP. It is possible that this charcoal which gave this date was not in its original stratigraphic context, as it came from part of the mound fill, which presumably consisted of soil taken from the nearby midden deposits and redeposited as the bulk of the mound.
In a recent letter to The Midden, Capes described these mounds: "Built of pebbles, cobbles and soil, the mounds were characteristically dome-shaped except for one large, oblong-shaped mound that covered a floor completely surrounded by large boulders which extended in a long snake-like appendage." (Capes 1993:1) As these mounds bear few similarities to the ones found in the Lower Fraser River and southeastern Vancouver Island areas, and as no human remains or burial goods were found, it is questionable whether this represents a burial feature.
Georgeson Bay Site
The burial GB-1-1968 is dated by stratigraphic association with component I, which has a corrected radiocarbon date of 820-1100 BC [2770-3050 BP].
The cairn feature "consisted mainly of fire-cracked rocks broken prior to inclusion in this formation" (Haggarty & Sendy 1978:18-19), although "evidence in the form of articulated or even concentrated human skeletal remains in direct association with the cairn was clearly lacking" (Haggarty & Sendy 1978:66). The evidence of human remains that was found consisted of a few adult cranial pieces, bringing into question whether this feature is burial or not.
North Saanich Site
In a trip I took in 1993 to the American Museum of Natural History, I obtained small samples of a few ribs and miscellaneous bones from burials NS-2-1898, NS-4-1898, NS-6-1898, NS-12-1898, and NS-1-1899, all of which came from cairns excavated by Harlan I. Smith at North Saanich. The uncorrected dates for these burials is as follows: 1200 + 170 BP (WSU 4547); 960 + 80 BP (WSU 4546); 1006 + 90 BP (WSU 4545); 1065 + 80 BP (WSU 4544); and 980 + 200 BP (WSU 4543). The corrected dates (assuming a sigma 13 value of -13.75 o/oo - Brian Chisholm 1994, personal communication) are as follows: 1380 + 170 BP (WSU 4547); 1140 + 80 BP (WSU 4546); 1186 + 90 BP (WSU 4545); 1245 + 80 BP (WSU 4544); and 1160 + 200 BP (WSU 4543), respectively. The dates with the large standard deviation were obtained from extended counts of very small amounts of carbon, but all are fairly secure.
98-NS-2 was made of large stones piled 1 meter wide and 1.5 meters long. 98-NS-4 and 98-NS-6 were made of large stones heaped over the body in a pile 1.2 meters wide and 1.5 meters long. The cairn that made 98-NS-12 was 2.7 meters wide and 3.7 meters long, with the sides and ends of the cairn being straight. 99-NS-1 was "poorly made...cyst ill-defined, covered with bowlders weighing about 300 pounds" (Smith & Fowke 1901:65).
Hatzic Site
I recently obtained an accelerator mass spectrometer date of 840 + 60 BP (Beta-78983) on a sample of wood fibres preserved on a copper disk found in a burial mound from the site Hill-Tout excavated in the Hatzic area (HZ-1-1894; see Figure 13). Although the wood fibres are presumed to be cultural, serving as a tie around the copper disk, the relatively recent date (compared to the very similar Mound 1 from Scowlitz) brings the solidity of this date into question.
Lazenby (1898) describes the mound as:
...roughly 24 feet in diameter at the base, tapering evenly to a crown of 12 or 14 feet at a height of seven feet... Around the base were placed large water warn boulders separated from each other by an inch or two only... a second row of boulders similar in all respects to the first circle [was encountered 6 feet in]. Four feet farther in was the chamber roughly square in shape, and, of course, within the chamber, the relics.
The mound described is similar in most respects to Mound 1 at Scowlitz.
Harbour House Site
The burial (HH-1-1994) recovered by Arcas Consulting Archaeologists at the Harbour House Site on Salt Spring Island was dated by the AMS technique on human bone collagen. The date of 770 + 60 BP, which was "calibrated to AD 1153 - 1311" [797-639 BP] is one of the most recent secure dates yet obtained on below ground interments. Stable carbon isotope fractionation analysis recently completed revealed a sigma 13C (o/oo) reading of -13.4 (R. Brolly, personal communication 1995). Given this stable carbon isotope figure, the corrected date for this sample would be 955 + 60 BP. Given that this date was taken on human bone collagen, this date has been accepted with confidence by the investigators, who site no reasons for its age to be re-evaluated.
Arcas Consulting describes the burial as being a clear "circular pit" which was "intrusive into the midden deposits." Over the burial was constructed "a stone cairn composed of angular boulders...which extended to bedrock". No grave inclusions were found in the burial. Two individuals were buried lying flexed on their left sides directly over top of each other. One was an adult male and the other an adult of unidentifiable gender.
