Return to Menu
A History of the Native People of Canada
Volume II (1,000 B.C. to A.D. 500)

Middle Palaeo-Eskimo Culture (Précis, Chapter 30)

Unlike the case with many of the cultures considered here, there are a number of readable and informative accounts of Palaeo-Eskimo culture (e.g. Maxwell 1984; McGhee 1996; Schledermann 1990). There is also a recent synthesis of Palaeo-Eskimo archaeology in Greenland that is directly pertinent to Canada (Grønnow and Pind 1996). For a personal account, that provides insights into the problems faced by Arctic archaeologists as well as information on Palaeo-Eskimos, see Schledermann (1996). A useful summary of Middle Palaeo-Eskimo culture (Dorset complex) can be found in Maxwell (1985: Chapters 6 and 7). The word 'complex' is used here to designate regional or temporal subunits of Palaeo-Eskimo culture and is regarded as a temporary archaeological classification device that is certain to be subjected to future modification. Terms like Pre-Dorset, Dorset, Lagoon, Independence II, Saqqaq, and Groswater fall into this category. Ongoing archaeological research is revealing that the regional diversity apparent in Middle and Late Palaeo-Eskimo culture was already well in process during Early Palaeo-Eskimo time in Period III (e.g. Sutherland 1996). The most widely known Middle Palaeo-Eskimo complex, Dorset, was first identified from National Museum of Canada specimens (Jenness 1925). It was recognized early that all of the aforementioned complexes were related to the Denbigh Flint complex of Alaska. The Arctic Small Tool tradition is the umbrella cultural classification that encompasses all of the regional and temporal groupings. While a close relationship between Pre-Dorset and Dorset was noted by a number of archaeologists (Collins 1956; Harp 1958; Taylor 1959) it remained for Taylor (1968) to demonstrate the nature of the relationship.

Middle Palaeo-Eskimo Implements - Drawings: David Laverie
Middle Palaeo-Eskimo Implements

To the left are two toggling harpoons dating to 900 B.C. and A.D. 500, respectively. To their right are a ground chert burin and a tip fluted triangular point. To the right is a soapstone lamp although Period IV lamps are usually rectanguloid in form.

(Drawings by Mr. David W. Laverie.)

It is necessary to comment on the use of the word 'Eskimo' in Palaeo-Eskimo. I agree with the observation "...that the Dorset population spoke some old variant of the Eskimo language" (Taylor 1968a: 9), a view supported by the oral traditions of the Inuit, more commonly referred to as Eskimos, which state that the Tunit (Dorset) spoke the same language as themselves (Rasmussen 1931: 113-114). Others are not as sanguine regarding the foregoing language association (e.g. McGhee 1996: 41).

During most of Period IV (1,000 B.C. to A.D. 500) the Eastern Arctic was involved in a cooling trend that began around 2,000 B.C. The greatest impact a cooling climate would have had on people was how it influenced sea ice conditions and the pockets of open water called polynyas where concentrations of sea mammals would have been available.

Middle Palaeo-Eskimo culture is used as a rubric for all of the related regional complexes of the Arctic Small Tool tradition dated between 1,000 B.C. and A.D. 500. This involves an occupation that extended from the Mackenzie River Delta in the west to Greenland in the east and north to northern Ellesmere Island and south to the Island of Newfoundland. Such an occupational distribution encompasses an east-west span of 3,700 km and a north-south one approaching 4,000 km (Taylor 1968a: 6). Given the variability in physiography, animal resources, and sea ice conditions across this enormous expanse of land and sea considerable cultural variability is to be expected, but always within a broadly shared cultural pattern. In order to identify regional cultural variability within Middle Palaeo-Eskimo culture the local complex involved will be specified. The term 'subculture' for such complexes was used earlier, but is discontinued here as an unnecessary source of potential confusion. A much expanded data base is going to be required to assess the interrelationships of these complexes to one another and undoubtedly there will be significant classification changes in the future. For example, each complex was undoubtedly composed of a number of independent communities that would have had their own name for themselves and a clear sense of self-identity. Such classificatory fluidity cannot and should not be avoided in a data-driven discipline like archaeology. As a discipline whose classifications are based upon cultural processes, rather than the laws of genetics or physics, archaeology is forced into the situation of having to establish the classifications necessary to control the mass of data but with the realization that these data are always incomplete and must be adjusted with the advent of new information. An increasing interest in Arctic archaeological classification problems should resolve some of the current classificatory difficulties, such as the same site being assigned to different complexes by different researchers.

In addition to complications arising from the relationships between regional variants of Middle Palaeo-Eskimo culture there is evidence that a number of different migrations took place in various places at different times. The antecedents of the Middle Palaeo-Eskimo Groswater complex of northern Québec, Labrador, and Newfoundland, for example, were derived from cultural developments in Ellesmere Island and adjacent Greenland. Colonization of the Hudson Bay region, on the other hand, originated with populations further to the west in the Foxe Basin.

Middle Palaeo-Eskimo culture descended from the Early Palaeo-Eskimo culture of Period III (4,000 to 1,000 B.C.). Controversy surrounding the transition of the Pre-Dorset complex into the Dorset complex continues into Period IV. For example, there is still argument concerning exactly when and how the Pre-Dorset complex became the Dorset complex. Here the period of the Pre-Dorset transition into Dorset is included in Middle Palaeo-Eskimo and, thus, begins at 1,000 B.C. rather than 500 years later as many would advocate for the beginning of Dorset (McGhee 1996). Differentiation of the two complexes has been mainly based upon the presence or absence of certain technological elements, with a clear emphasis being placed upon the differences rather than the similarities. A number of assumptions of a questionable nature have been made regarding the relationship between the two complexes, such as the suggestion that Early Palaeo-Eskimos had the bow and arrow but that Middle Palaeo-Eskimos abandoned the weapon system. Another assumption is that all Palaeo-Eskimos lacked the technological capabilities to capture large sea mammals from boats. This appears to have been mainly based on the absence of later Inuit specialized whaling gear that survives archaeologically. Large whale remains are occasionally recovered from Middle Palaeo-Eskimo sites but are assumed to have been the result of people scavenging the bodies of stranded whales. The accuracy of this assumption has been questioned by the evidence of large whale hunting at the late Period III Saqqaq complex site of Queqertasussuk in western Greenland (Grønnow 1994: Table 1). It is necessary to accept the fact that many elements of ancient technologies are archaeologically invisible. There is no doubt that a reliance on the rich ethnographic documentation of Inuit technology and lifeways has influenced interpretations of Palaeo-Eskimo technology, especially when it deviates from the Inuit norm. Such a one-to-one assumption contains many pit-falls.

A major difference between the Early Palaeo-Eskimo of Period III and the Middle Palaeo-Eskimo of Period IV was the abandonment by the latter of formerly occupied territories in the interiors of northern Alaska and the barrengrounds of Keewatin and Mackenzie districts in the Northwest Territories (McGhee 1987). This shift in settlement strategy away from the interior suggests a reduced focus on caribou and a concentration on sea mammal hunting. Indeed, the transition into the Dorset complex has been regarded as a direct product of an adaptation to the advantages of greater sea ice coverage resulting from climate change.

A central tenant of Eastern Arctic archaeology during Period IV and later is that a central core area, represented by Hudson Strait, the Foxe Basin, and the Baffin Island region, was continuously occupied and functioned as a launching pad for population migrations to more remote regions as well as a place to retreat to during unfavourable situations. While the concept still has some utility, it appears to be an over-simplification of events. Other core areas, such as the High Arctic and northern Greenland, Labrador, and the Central Arctic, existed for considerable periods of time and interacted with one another as well as the Core area. Interaction, however, was variable in intensity. For example, it now appears that the earliest Middle Palaeo-Eskimos of Labrador and Newfoundland had more in common with people in the High Arctic, adjacent Greenland, and most of northern coastal Québec than they did with people in the Core area proper.

The Arctic Small Tool tradition (Irving 1957), of which Middle Palaeo-Eskimo culture represents a segment of the eastern branch of the tradition, was well-named. Stone tools are typically minuscule in size and maintain an exceptionally conservative adherence to culturally approved stylistic models. Tools were also fashioned with meticulous care. A wide range of burin varieties are represented, most with ground working edges unlike the burin blow sharpened implements of Period III. Burins appear to have served a wide range of cutting, gouging, and etching functions and, as they were always fashioned from very hard stone like chert and nephrite, were ideally suited for fashioning implements and ornaments out of ivory, antler, and bone. Microblades may or may not be present or abundant on Middle Palaeo-Eskimo sites. On sites with wood preservation, microblades have been found hafted in wooden handles. Their main function would have been as knives for cutting up hides for clothing and for slicing meat. Points are generally assumed to have tipped toggling harpoon heads rather than arrow shafts but this assumption is questionable. Chipped side blades were fitted into slots along the sides of bone and antler lances, harpoons and, on rare occasions, arrows. Ground slate points and knives become common and are frequently side-notched like many of the chipped stone points. There is a wide range of scraper, knife, and chipped adze forms, the latter generally ground only at the bit. Important bone and ivory implements are needles with gouged-out rather than drilled eyes and a range of toggling harpoons, the latter being particularly useful as horizon markers. Some of the ivory and antler harpoons were tipped with chipped or ground stone points while others were self-armed in the sense that their piercing tips were simply ground to a sharpened point. Thanks to permafrost conditions, wood preservation occurs at some sites where items like handles for stone burins and microblades and boat and sled parts have been recovered. Art objects in ivory, bone, antler, wood, and soapstone are present in increasing numbers but do not achieve the degree of florescence apparent during Period V. The first detailed quantitative and qualitative descriptions of Early and Middle Palaeo-Eskimo tool kits were done by Taylor (1968) in his effort to demonstrate technological continuity between the Pre-Dorset and Dorset complexes.

Regional complexes of Middle Palaeo-Eskimo culture have been identified mainly on the basis of their tool kits. This not only applies to Middle Palaeo-Eskimo culture but also to the preceding Early Palaeo-Eskimo culture. Thus, during Period III (4,000 to 1,000 B.C.) "In the North American Arctic, the early ASTt is represented by the Proto-Denbigh and the Denbigh Flint complex in Alaska, the Independence I complex found primarily in the Canadian High Arctic and northern Greenland, the Saqqaq complex, found primarily in southwest and southeast Greenland and the Pre-Dorset and Dorset complexes generally found in the Canadian central and eastern Arctic regions, including Labrador and Newfoundland" (Schledermann 1990: 20). Regional differentiation appears to increase during Period IV with such complexes as the Lagoon complex in the Central Arctic and the transitional Pre-Dorset, Independence II, late Saqqaq, Dorset I, and Groswater complexes to the east. As with Period III, however, the similarities between these regional complexes exceed their differences.

There was a significant subsistence shift from Period III to Period IV with Middle Palaeo-Eskimos of the latter period abandoning former interior hunting territories and focusing on sea ice hunting of sea mammals (Maxwell 1980: 170). Terrestrial animals were still important but more as a supplementary resource than a primary one. Colder climate is speculated to have reduced the availability of caribou. Also, marine mammals tend to be a more dependable food source although their availability is affected by sea ice conditions. As Middle Palaeo-Eskimo culture occupied over 1,500,000 square kilometres of quite variable landscape, extending from the Mackenzie River Delta across Arctic Canada into Greenland and south along the Labrador coast onto the Island of Newfoundland, the animal resources and the methods necessary to capture them would also have been variable. Despite regional differences in resource availability and seasonal emphases on specific prey animals, like caribou, seal appear to have been the mainstay of sustenance throughout the entire region.

Concomitant with a concentration on sea mammal hunting was a natural settlement pattern shift onto the coast and away from earlier interior settlements. This is one of the outstanding characteristics of the Period IV occupation of the Arctic. Many sites are situated within easy distance of the sea-ice interface with its sea mammal concentrations. A major problem facing a fuller understanding of settlement patterns is the likelihood of winter settlement on the sea ice in igloos. Such winter settlements would be ideally suited for purposes of either flow edge or seal breathing hole hunting. The possibility of winter sea ice settlement was raised for Early Palaeo-Eskimo culture in such places as the Central Arctic where winter occupation sites on land appear to be missing (Wright 1995: 435). Of course, all evidence of winter sea ice occupation would disappear each year along with the ice. There are also problems with both acquiring faunal samples and drawing inferences on sustenance from them. Large animal bones, for example, were used as fuel and, given the climate and availability of permafrost or snow banks in the summer, foods could be readily cached for winter consumption. Thus, a concentration of aquatic bird remains on a site need not indicate a summer occupation when the birds would have been available but, rather, could represent other seasons of the year when cached bird carcasses were being consumed. In many instances site locations are likely a more reliable indicator of season of occupation than recovered animal remains.

Middle Palaeo-Eskimo dwellings range from tent rings and semi-subterranean rectanguloid structures to dwellings with axially aligned hearth-work areas and adjacent sleeping quarters to either side; the latter being most characteristic of the High Arctic, northern Greenland, and Labrador-Newfoundland.

Middle Palaeo-Eskimo shamanism is undoubtedly reflected in the mainly naturalistic carvings in ivory, antler, bone, soapstone, and wood. A number of these objects may also have functioned as personal amulets. While such art had its beginnings in Period III and became increasingly common in Period IV, it was not until Period V (A.D. 500 to European contact) that they achieved the peak of their development. In addition to animal and human forms other relatively common objects are containers for shaman paraphernalia carved from walrus tusk and miniature toggling harpoons. Some stylistic correspondences in the art forms exist with the Okvik and Ipiutak complexes of the Bering Strait region of Alaska. The petroglyph sites on the Québec side of Hudson Strait cannot be dated but on stylistic grounds should pertain to Period IV. Not only archaeologists but also art historians have been fascinated by Palaeo-Eskimo art (e.g. Taylor and Swinton 1967). Indeed, it has elicited the observation that "The importance of magico-religious beliefs and their association with Dorset technology is apparent in the most striking accomplishment of the Dorset people: the creation of a body of art that is unique, unexpected, and remarkable" (McGhee 1996: 148).

Evidence of Middle Palaeo-Eskimo burials is extremely rare, coming mainly from vandalized crevice burials in the limestone cliffs of Newfoundland. A few more northerly sites have produced occasional human remains. For some unexplained and presumably symbolic reason these remains are often represented by mandibles. Dog remains are equally rare. Possibly dogs were accorded similar treatment as humans; probably exposure or placement on the sea ice. Physical anthropological studies of the limited evidence attribute the individuals to a northern Mongoloid racial stock to which the present Inuit people belong.

It has been popular to consider Middle Palaeo-Eskimos as being isolated in their Eastern Arctic homeland. There is clear evidence, however, of influences coming from related peoples in Alaska. Middle Palaeo-Eskimo depopulation of the Labrador coast around A.D. 200 has been attributed to Indian intrusions to the coast as well as to natural disasters (Tuck 1976: 101). Direct evidence of Indian-Middle Palaeo-Eskimo contact, however, is lacking and indeed, "The overall conclusions regarding Dorset-Indian contacts in Labrador must be that there is very little evidence to suggest that it occurred..." (Fitzhugh 1980: 29).

The highly mobile, small family groups composing Middle Palaeo-Eskimo society were likely organized into local bands which, in turn, would be related by blood to neighbouring bands. Given the necessity for people to find food and that prey animals can be affected by natural and not totally predictable circumstances, considerable mobility can be anticipated including occasional long distance population shifts. Such moves, however, were likely in conjunction with other groups and thus a mutual support system would still be in place. Throughout most of the year it can be anticipated that the nuclear family was the major social unit and that two or three such families either moved together or were in close proximity to one another. At favourable times of the year, such as during the fall char (Salvelinus alpinus) spawning runs in lakes and rivers or near a summer bird rookery, a number of these family groups could come together to reaffirm their social identity, arrange marriages, perform ceremonies, and exchange information and gossip.

To judge from the nature of the Middle Palaeo-Eskimo tool kit and, indeed, that of all Palaeo-Eskimos, it can be inferred that the people placed a high value on tightly constrained behaviour. There was only one way to do something and that was the Palaeo-Eskimo way. In this respect, there appears to have been little tolerance of idiosyncratic behaviour, even in terms of tool production. It has been suggested that the Middle Palaeo-Eskimo obsession with ultra conservative ways of doing things may have been largely responsible for their disappearance as a recognizable cultural entity when the ancestors of the Inuit occupied their territories around the beginning of the 10th century (Nash 1976: 155).

Limitations in the archaeological evidence are much the same as elsewhere in Canada, such as a scarcity of detailed published regional accounts, component mixture due to repeated occupations of the same sites, and so forth. There are, however, two circumstances largely unique to the Arctic; the probability of archaeologically invisible winter settlements in snow houses on the sea ice and, as a degree of compensation for the former, the visibility of archaeological remains from the air. As the Inuit of Period V, with their highly visible large stone, whale bone, and sod winter dwellings, frequently occupied the same sites as Palaeo-Eskimos aerial survey will accidentally lead to the discovery of the earlier occupations. It has been suggested that Arctic archaeology "...shares certain patterns with mythological belief: acceptance of prior authority, intolerance of alternative views, and a search for simple explanations for complex phenomena" (McGhee 1983: 21). There has also been a tendency to assume that the absence of direct archaeological evidence of a technological or cultural element equates with an actual absence. This has been the basis for questioning the existence of elements of technology such as kayaks, umiaks, snow houses, dog traction, and the bow and arrow. In other words, there has been an over reliance on Inuit ethnographic analogy and the tacit assumption that it can be applied to Palaeo-Eskimo technology. As recent ethnographic studies amply demonstrate, there are numerous ingenious ways that people have been able to achieve their ends without leaving any archaeological record.

Volume IVolume II

Back Menu - A History of the Native People of Canada Continue