traction, Alaska, Northwest Alaska, Anthropological Papers, pp, dog, absence, communication, Sheppard, archaeological record, Alaska Anthropological Association, Smithsonian Institution, Smithsonian Institution Press, W. Sturtevant, implications, Fifth Thule Expedition, Alaska Anthropological Association Monograph Series, Handbook of North American Indians, caribou hunting, University of Alaska Press, American Museum of Natural History, mode of transportation, prehistoric societies, Alaska Journal of Anthropology, David Damas, University of Alaska, societies, Prehistory, sleds, economic cost, Kobuk River, sled, American Antiquity, Traditional Eskimo, D. C. Anderson, Cambridge University Press, Western Arctic, Canadian Museum of Civilization, Marco Polo, Point Hope, aggregation, archaeological terms, travel distances, hunting, search range, dog sleds, exploitation, food supplies, Northwest Alaskan Eskimo Societies, William L. Sheppard Sheppard Research, Eskimo culture, Cultural ecology, Rainey, Eskimo cultures, Eskimo, cultural systems, traction system, subsistence strategies, modes of transportation, settlement strategies, Southwestern Alaska, Mathiassen 1928:79ff
ARTICLES THE SIGNIFICANCE OF DOG TRACTION FOR THE ANALYSIS OF PREHISTORIC ARCTIC SOCIETIES William L.
Sheppard Sheppard Research, 980 SW Hillcroft Ave., Portland, OR 97225. Sheppard @ hevanet.com.
Abstract: Dog traction was a central element of Eskimo cultures from Greenland to Southwestern Alaska, yet has received relatively little attention from northern archaeologists. This Mode of Transport
ation likely emerged in the last 1500 years and was a necessary element in the mobile subsistence strategies and social network
s of historic Eskimo societies. An understanding of the effects of adopting dog traction is necessary for analyzing prehistoric societies that preceded and came after its development. The increased mobility conferred by the use of dogs likely had effects on functional and stylistic variability in archaeological assemblages, the costs of resources and means of procurement, settlement strategies, and other aspects of Eskimo culture. Keywords: Mobility strategies, Cultural ecology, Prehistoric change
INTRODUCTION The use of dogs as draft animals to haul sleds, or dog traction, is one of the traits universally identified with Eskimo culture. Despite the centrality of this mode of transportation in the societies and economies of Eskimos from Greenland to southwest Alaska, surprisingly little attention has been given to its inception and the roles it played in those systems. This is particularly interesting given that it is widely held that dog traction is a relatively recent development in Eskimo prehistory (Anderson 1988; Bandi 1969; Dumond 1977; Giddings 1952, 1964; Hall 1978; Jenness 1940; Larsen and Rainey 1948; Rainey 1941; VanStone 1955). Almost all researchers who have been concerned with prehistoric Eskimo or Eskimo-like cultures have considered dog traction to a limited extent, and some have briefly considered some of its potential implications (Anderson 1988; Larsen and Rainey 1948; Sheppard 1986). But despite the obvious importance of this technological adaptation, to date only Hall (1978) has made a concerted attempt to focus on the many changes that it may have initiated in prehistoric Eskimo culture and ecology. One of the principal objectives of modern anthropologists is to understand how cultures function as systems. This requires one to examine how the social and economic components of the cultures connect and interact, and to do so we must understand the mechanisms that allow the flow of information and resources. For human systems we are talking about communication and transportation and, for most of human history
mechanisms have been equivalent--the actual movement of people has controlled the flow of information and resources across the landscape, by foot or by other means of travel. In the prehistoric Arctic there have been two nonpedestrian modes of transportation: water travel and dog traction. The former must be of equal antiquity with the settlement of most of the North America
n Arctic, which would have been logistically and economically unfeasible without it. Because of that, waterborne travel and subsistence activities can be treated as more of a constant in analyzing cultural systems during the last 5-6000 years of North American
Arctic prehistory. The presence of dog traction, on the other hand, cannot be similarly dealt with because it originated at some intermediate point of time and likely involved significant systemic changes. Therefore, we must consider what the implications are for its adoption, especially for the time before dogs were used-- when we really do not have an adequate ethnographic model on which to base analyses. In this paper, I will look at the social and economic implications of dog traction and explore as well the potential ramifications of those changes for archaeological interpretation. In part, one significant goal is to stimulate more interest in this question, particularly since, as will be seen, my analysis raises as many questions as it answers. It is not my intent to "dot every i and cross every t" regarding the issue, but to start addressing the significant
70 Alaska Journal of Anthropology Volume 2, Numbers 1 - 2
conceptual issues involved. It is also not my intent to discuss all aspects of the domestication, use, and consumption of dogs by Eskimo peoples, most of which have no significant bearing on my analyses. This is not a paper about dogs and dog traction per se, but about the systemic implications of the latter. In the subsequent discussion, I will address the following issues. First, I will examine what are reliable archaeological indicators of dog traction and when this mode of transportation is likely to have arisen. Next, I will look at how dog traction functioned in real societies, using Northwest Alaska as a model. Given that ethnographic framework, I will look at the possible structure of predog traction societies by removing the capabilities provided by dog transport. With that in mind, I will consider how the pre-dog traction system might be materially reflected in the archaeological record
. Finally, I will ruminate briefly on how Northwest Alaskan prehistory may reflect the adoption of dog traction and make some general comments about its consideration in relation to broader issues of northern prehistory. ORIGINS The issue of when dog traction first emerged is a thorny one since there are no good indicators of its actual inception. The first use of dogs to pull loads across snow doubtless involved little more than a length of raw hide and whatever else was at hand. This could be as insubstantial as a frozen hide toboggan such as used by the Iglulik Eskimos (Mathiassen 1928:79ff). Thus, early dog traction was probably largely indiscernible if not invisible in the artifactual archaeological record. A variety of biological indicators might be used to signal the expanded use of dogs in a broad sense. These could include skeletal deformations, increased frequency of bones overall, evidence of selective breeding, and other factors. Some researchers have considered spinal deformation of dog bones found on Banks Island dating over 2000 years ago as relating to dog traction (Morrison, personal communication), however, a case can also be made for such damage to have been produced by use as a pack animal (Arnold, personal communication). In contrast to the above, dog traction in its fully developed form has several reliable indicators that may be preserved in the archaeological record. These include whips and whip handles, harness parts, trace buckles, and swivels (Giddings 1952; Hall 1978). The simple presence of sled parts is not adequate evidence since these were clearly used prior to the use of dogs, although more than one researcher has proposed the so-called "built-
up" sled as a possible indicator, an idea which I will return to later. So when do we have evidence for the emergence of developed dog traction? Historical descriptions of dog traction in northEast Asia
date to the late 13th century (~700 B.P.) when Marco Polo recorded the following: In order to travel over the frozen surface of the ground, they construct a sort of vehicle, not unlike that made use of by the natives of the steep and almost inaccessible mountains in the vicinity of our own country, and which is termed a tragula [emphasis in original] or sledge. It is without wheels, is flat at the bottom, but rises with a semicircular curve in front, by which construction it is fitted for running easily on the ice. For drawing these small carriages they keep in readiness certain animals resembling dogs, and which may be called such, although they approach the size of asses. They are very strong and inured to the draught. Six of them, in couples, are harnessed to each carriage, which contains only the driver who manages the dogs, and one merchant, with his package of goods (Polo 1958:325-326). Other sources, originally uncovered by Birket-Smith (1929:169), suggest other contemporaneous use of dog traction in Asia. Most specific is the account of Ch'ang Te who observed the Kirghiz use of dogs as draft animals around A.D. 1259 (Bretschneider 1910 [I]:129). Field and Prostov (1940-1941: 388-406) noted the presence of dog harness parts in a site near the mouth of the Polui River that dates between the first and sixth century A.D. The Asian sources imply at a minimum the contemporaneity of dog traction between Asia and North America, if it did not actually originate on the west side of Bering Strait. Although several authors (Ackerman 1984:110; Ford 1959:156; Rainey 1941:547) have been willing to concede at least the potential for dog traction as early as Birnirk times, others (Giddings 1952:62-63; 1960:124; Hall 1978:212-216; Hickey 1979:427) have only conceded its existence in the last three to five hundred years. On the other hand, in Canada there is long-standing evidence of dog traction accompanying the spread of Thule culture (by whatever means) dating to at least 1000 years ago on Banks Island in the Amundsen Gulf region (Arnold, personal communication) and on Ellesmere Island no later than 900 years ago (McCullough 1989). Thus, unless we believe dog traction emerged in full form immediately after
The Significance of Dog Traction for the Analysis of Prehistoric Arctic Societies 71
the expansion to the east, it must have been adopted in Alaska before 1000 B.P.
DOG TRACTION IN HISTORIC ESKIMO SOCIETIES
Returning to the issue of built-up sleds, I believe, as
The following discussion is based mainly on the vast
suggested by VanStone (1955:115) and others, that the body of data compiled by Burch for Northwest Alaska
change in sled construction is indicative of dog traction. (especially 1998b, but also 1972, 1975, 1976, 1980, 1981,
It provides an intermediate point between no evidence of 1984a, 1984b, 1998a), on my own field research in the
dog traction and the presence of all definitive elements. region (Sheppard 1983, 1986, 1988; tapes and Field notes
1979, 1980, 1982, 1984, 1985 and 1987), and from Uni-
As its name implies, the built-up sled featured a raised versity of Alaska/Bureau of Indian Affairs ANCSA
cargo area, lifted above the runners by curved wood or 14(h)(1) research in Northwest Alaska between 1975 and
antler arched supports which increased clearance by a 1992 (BIA ANCSA Oral History
Collection, Bering Straits
factor of two or more. Built-up sleds featured much nar- tapes 1975-1992, NANA tapes 1987). To avoid cumber-
rower sled shoes and were altogether lighter and more some over-citation, the reader may assume the sources
flexible than their counterparts which had a series of cross of information are citations above unless otherwise noted.
slats connected directly to wider, heavier runners.
The role of dog traction in historic Eskimo societies
The built-up sled featured several advantages in before the late 1800s was quite different from how it
moving about snow-covered terrain. One improvement was used during the gold rush and later periods, when it
would be in traversing snow and snow-covered ice, where filled much the same niche as the modern snowmachine.
a built-up sled would be less likely to accumulate snow Rather than being a mode of personal transport, it can
between the runners. It would also be less likely to hang more correctly be conceived of as an improvement in
up on rough terrain, especially the roots and snags one freight hauling associated with pedestrian movements. In
would find in interior areas. The shorter, narrower sled general, it allowed people to move more pounds of food
would also be better fitted for close, windy
trails that would be found in forested or
brushy conditions. For Northwest Alaska,
Burch (1998a:196) notes that the built-up
sled or uniapiaq (Figure 1) was preferred
in interior districts while the long, low
qamun (Figure 2) was preferred in coastal areas. In general, it is hard to visualize Figure 1 - Uniapiaq or built-up sled (from Murdoch 1892: 354 [Figure 356]).
much travel through soft conditions with-
out the assistance of dogs. If we accept built-up sleds as and possessions at a faster pace. Eskimo families through-
evidence (or at least a necessary precursor) for the be- out the Arctic typically had small dog teams and often
ginnings of developed dog traction, then it began as early augmented canine power with that of women and chil-
as 1500 B.P. during Birnirk times. In any event, it is very dren. Estimates of the number of dogs per household are
quite variable, but I believe about three per family is rea-
sonable. People seldom rode on the sleds. The limitation
in the number of dogs was reflective of the economic
cost of the teams themselves, which required sustenance.
Figure 2 - Qamun or low sled (from Murdoch 1892: 355 [Figure 357]).
In Siberia, the size of a family's team was directly correlated to its economic well being (Schnirelman 1994:183185) and it is reasonable to suppose this relationship ex-
hard to make a case for its adoption any later than 1200 isted elsewhere. When times were good the number of
B.P., assuming the eastward Thule expansion with dog dogs expanded and when times were bad the numbers
traction post-dates somewhat the actual beginnings of that decreased, by neglect and human consumption
transportation. Interestingly, as far as I can tell, built-up individual families might have few dogs, they did aug-
sleds which show up in Alaskan Birnirk assemblages ment their teams with those of other families for specific
never made an appearance in the eastern Arctic, even in trips or activities.
historic times. Thus, there is a situation where people, if
they actually migrated to the eastern Arctic as late as 1000
It is important to re-emphasize that the husbandry
B.P., left behind or discarded a significant component of and maintenance of dog teams was not simply a labor
dog traction technology.
72 Alaska Journal of Anthropology Volume 2, Numbers 1 - 2
saving development, it did have a significant economic cost. The actual investment in labor and resources could be very difficult to estimate. Historically, the pattern was generally not to feed the dogs in the royal manner in which they were kept in modern times. They were often left to fend for themselves (particularly in the summer) or ate items, bones, etc., that were generally not fit for human consumption. However, working dogs could not function on a maintenance diet and even their consumption of bones took away from potential starvation foods of their human keepers. Before the introduction of the snowmachine, a significant proportion of the subsistence effort was devoted simply to acquiring dog food, and after the adoption of mechanized transport some major species actually dropped out of the subsistence regime in certain locales. In Norton Bay, for example, almost all seal hunting ceased except for bearded seals (Erignathus barbatus); much of this decrease was attributed directly to the adoption of snowmachines and decreased need for dog food (Sheppard, tapes and field notes, 1979). Burch has shown that most of the 30 percent decline in daily subsistence consumption between 1966 and 1984 could be accounted for by the subtraction of working dogs from the community (1985:110-112). Clearly, modern people faced much greater dietary demands from large teams used for trapping and other activities, but still one must factor in a significant level of subsistence effort for pre-contact societies. As will be seen below, dog traction facilitated a more diverse diet, but it also required it. Dogs were important for both summer and winter travel. During the summer, families harnessed the animals to umiaqs and used them to help move the heavily laden craft upstream. They were also used as pack animals. The discussion below will, however, concentrate on winter use of dogs with sleds--partly for the sake of simplicity, and because of the relative invisibility of summer use in the archaeological record. Winter use of dog teams can be roughly classified into three main areas: household movements, retrieval, and visiting. Families within most Northwest Alaskan societies had two types of household movements, small-scale point to point camp shifts in mid- to late-winter and longer-distance, major household movements from winter areas to break-up-time subsistence sites. The former consisted of the movement of nuclear or small extended families from one resource area to another and did not necessitate moving all possessions that had been hauled to winter sites. In contrast, the latter involved the movement of many people and all their goods, including kayaks and umiaqs, over many miles.
Retrieval took two forms as well. Long distance, seasonal retrieval involved the recovery of cached food acquired earlier in the year. The best example of this is the retrieval of caribou meat which was stored near kill sites while items of higher priority, mainly hides and sinew, were hauled back to settlements on foot. Short-distance retrieval involved the fetching back, usually by women, of harvests by pedestrian hunters. Visiting varied in scale depending on the abundance of harvests, ranging from the aggregation of members of a single society, to larger inter-societal gatherings at messenger feasts, and occasionally even larger trade fairs involving several societies. Visiting, in effect, amounted to household movements, but typically of intermediate to long distances without the need to haul all of a household's possessions. But sleds could be used to haul food, needed possessions, trade goods, and occasionally people. To give some quantitative perception of the mobility provided by dog traction, the following table (using data from Burch 1998b) presents some estimates of distances between winter and spring settlements (the second column) and maximum distances for intra-societal aggregation (the third column). Looking just at the ability of these societies to aggregate, several have maximal distances of over 30 miles, two or more days by dog team and perhaps double that without the teams. Figure 3 shows the distributions of the societies named below.
The Significance of Dog Traction for the Analysis of Prehistoric Arctic Societies 73
Figure 3 - Northwest Alaskan Eskimo Societies mentioned in text.
MODELING THE ABSENCE OF DOG TRACTION If one took away dog traction from an early historic Eskimo society, what would be the result? It should be recognized that there was nothing that was done with dog sleds that could not have been done, at least theoretically, by humans pulling sleds. However, the use of dogs decreased the effort involved in moving people and goods and increased range and speed. In looking at the probable effects of the absence of dog traction it is most reasonable to propose general trends. Some of the logical implications of a more burdened, pedestrian society would be as follows: (1) smaller overall yearly-ranges; (2) increased coastal orientation away from inland resources; (3) a reduction in the amount of tools and supplies carried away from coastal areas; (4) longer
stays at more widely separated locales; and (5) decreased winter visiting activities, especially between members of different groups. In terms of more specific impacts, one of the greatest would be on mid-winter caribou hunting carried out from scattered small settlements. By all accounts such an enterprise was problematic at best with dogs and would be very difficult without that increased mobility. Dog traction allowed a greater search range in looking for prey as well as quicker and more efficient retrieval of the cached proceeds of hunts. In the absence of dog traction, caribou exploitation would track much more closely to up-cycles when peaks in the prey population would mean greater likelihood of pedestrian hunters encountering the animals at widely distributed locations. In down cycles, exploitation might be limited to late summer hunts, which were indispensable in providing good hides,
74 Alaska Journal of Anthropology Volume 2, Numbers 1 - 2
sinew, and other products but less important for the food supplies. The absence of dogs would require pedestrian packing of the summer harvest in a less efficient manner in terms of total time, number of round trips, and potential loss of cached items. A variety of secondary impacts can be predicted; among these would be a less buffered economy. As Halstead and O'Shea (1989:123) have noted, there are a variety of ways in which hunter-gatherers can buffer themselves against economic uncertainty, including: (1) a more heterogeneous resource base; (2) exploitation of a broader hunting range; and (3) increased exchange. The absence of dog traction limits all of the above because fewer resources could be effectively exploited and people would have less ability to increase their overall hunting range. The absence of improved transportation would also limit the ability of families and larger groups to disperse in hard times. It would also be more difficult to simply pack up and leave an entire territory, which was the last resort for historic Eskimos. The impact of smaller ranges would be in smaller group sizes. This in turn would likely result in a greater percentage of exogamy. Although historic groups had a decided preference for societal endogamy, in practice a great deal of exogamy occurred. For smaller groups, out-marriage would be an absolute necessity and thus the level of relatedness between adjacent populations would be high. The combination of small group size and increased kin connections suggests that violence at both local and regional levels would be much more rare. This commonsense statement is supported by observations of modern tribal warfare. Amongst the Yanomamo, stable alliances between villages, hence less violence between them, were predicated on the exchange of women (Chagnon 1992:160-162). Elsewhere, Pospisil noted for the Kapauku of Papua New Guinea an inverse relationship between relatedness and the level of violent conflict. Additionally, there are obviously basic logistical problems for small groups to organize the type of large-scale, longdistance conflict that characterized the immediate precontact and early historic period. The decrease in group size, general mobility, and reduced ability for broader inter-societal interaction would mean that information flow would be slower and more localized. This might lead to more pronounced local differences in dialect, art, clothing and other cultural aspects that are not so closely tied to local economy and material problems, and thus have a greater ability to drift. Tech-
nological innovations developed in one area would also move more slowly across long distances. It is simplistic to contrast only the total absence of fully developed dog traction with its most sophisticated late prehistoric form. It is very arguable, however, that technological changes following the development of builtup sleds are minor, thus most of the significant structural changes resulting from fully developed dog traction are the result of the new sled form. On the other hand, the use of heavier, lower coastal sleds pulled by dogs could have enhanced coastal adaptations and settlement with little definitive appearance archaeologically. One could visualize the changes as "Dog Traction Lite," with changes restricted mainly to the coastal zone. ARCHAEOLOGICAL MANIFESTATIONS How would the preceding affect what one might find in the archaeological context? From a settlement standpoint there would be a greater concentration of winter sites in coastal locations. In the absence of dog traction, it would be harder for families to move people and supplies overland. Fewer sites would occur away from the coast and those that were found in interior locations would be more restricted to ones that were more readily accessible by waterborne transport. Fewer interior sites would reflect winter occupations because use of the sites, even if accessed initially by water, would require pedestrian movement back to the coast carrying all goods, supplies, boats, and people moved there earlier in the year. Bascially, the absence of dog traction would encourage coastal settlement and discourage settlement away from the coast, but it would not absolutely require or eliminate either option. Coastal and interior sites would likely differ significantly in assemblage size and diversity. It stands to reason that if one must bear the full burden of one's tool kit and use it for a shorter period of time, fewer, more multifunctional tools would be preferable over many tools with very specific uses. One would also be encouraged to manufacture tools on site using locally available materials. Both of those factors would encourage the use of chipped stone over ground slate in interior contexts. The latter would be more time consuming to make and would be less widely distributed. Ground stone tools also tend to have been manufactured for single or limited functions. Increased local stylistic variability should be reflected in the archaeological context. Such variability would be most apparent in more permanent coastal settlements where there would be a more elaborate and diverse tool kit.
The Significance of Dog Traction for the Analysis of Prehistoric Arctic Societies 75
DOG TRACTION AND NORTHWEST ALASKAN PREHISTORY I believe there is a strong case to be made that many of the major prehistoric changes in the last 1500 years are the consequence of the adoption of dog traction. In particular, I would contend that reduction in stylistic variability in late prehistoric assemblages owes much to the leveling influence of significantly increased information flow. Prior to about 1000 B.P., Northwest Alaska can be characterized as having regional styles of art and tool forms, including Okvik, Old Bering Sea, Ipiutak, and Birnirk. This stylistic variation was conventionally interpreted in terms of cultural sequences, but there is now good reason to believe all of those archaeological classifications are essentially contemporaneous (Gerlach and Mason 1992). Regional stylistic variation would be encouraged by the absence of dog traction, which would limit long-distance interaction to summer months when people could travel by boats.
The ability to exploit caribou more efficiently in the winter is likely, as Hall (1978) suggested, the factor that promoted the gradual expansion of Eskimos up the Kobuk and Noatak drainages. Winter caribou hunting provides an explanatory basis for settlement shifts observed by Harritt (1994) at Kuzitrin Lake and Hall and Gerlach (1988) at Tukuto Lake. In both cases, earlier occupations of these inland locales was evidently the result of short-term summer use, whereas late prehistoric occupations were characterized by winter use long-term enough to warrant the construction of more substantial and more enduring semisubterranean structures. One of the most significant general issues facing archaeologists in northern Alaska and indeed throughout the Arctic is the adoption of whaling. Previously, it has been suggested (for example McCartney 1984) that surpluses generated by whaling would have allowed the maintenance of dog teams. I believe it is worthwhile to turn this question around and consider the transportation and
Figure 4 - Maximum aggregation distances for Point Hope whale hunters. 76 Alaska Journal of Anthropology Volume 2, Numbers 1 - 2
logistical issues related to whaling. At best, even with modern harpoon technology, indigenous whaling can be characterized as highly stochastic. The chances of an individual strike yielding a capture are low, as are the chances of an individual crew making a kill. Thus, the success and effectiveness of whale hunting, particularly with prehistoric technologies, would be contingent on the degree to which multiple crews could be mustered. For example, consider the hypothetical aggregation of six crews. Each crew of eight fit adults would likely represent the manpower contribution of several households composed of four to six people. If households supplied on average two adult hunters then the total manpower of 48 whaling crew members for six boats might involve the total aggregation of at least 24 families, or 96 to 144 people. Even in the best cases, this concentration of people would require the coalescence of families dispersed over literally thousands of square miles. For example, Figure 4 shows the maximum travel distances for Point Hope society members to aggregate for whaling at Tikigaq (Point Hope). While such aggregation is technically feasible by foot travelers, it would have been greatly enhanced by dog traction and may have significantly increased the ability of people to be at the right place at the right time. Naturally, there would be feedback in terms of acquisition of surplus and increased team size; however, I believe there is a strong case for dog traction to be considered integral to the adoption of whaling, not simply a consequence. This is even more clearly the case if one adopts such complex formulations as Sheehan's (1995) model of the interrelatedness of whaling and interior subsistence. The implications of dog traction for mobility, interaction, group size and other factors also cannot be neglected in analyzing changes in prehistoric social relations
hips. In a recent paper, Mason (1998) has used the concept of "polities" in analyzing Birnirk, Old Bering Sea, and Ipiutak site distributions and assemblages. Mason's polities are equated with the territorially defined "societies" as developed by Burch (1980, 1984, 1998a, 1998b). While it is laudable to treat the regional prehistory in cultural rather than archaeological terms
, it is evident from the preceding discussion that such polities in the historical sense were largely predicated on the integrative ability of dog traction. One wonders whether anything remotely resembling the historical model could have been present without it. I am in greater accord with Burch (1998b:316-317) that a regional system of societies and boundaries likely may have existed as early as Punuk times, and this is more consistent, for example, with the best concrete evidence of warfare, that is armor, which does not occur until that time.
One of the subtler considerations about the effects of dog traction in changing prehistoric societies relates to the recognition that overtly similar archaeological reflections of prehistoric ecologies [faunal remains, technology and the like] can be the result of dramatically different systems. This recognition, or lack thereof, can involve striking differences in interpretation. For example, Giddings and Anderson (1985:318-322) characterize the levels of Choris, Birnirk, and Western Thule caribou exploitation at Cape Krusenstern as essentially the same. Despite the apparent similarities in utilizing this terrestrial resource, the comparable faunal assemblages could have been produced in very different ways. For example, the relatively high level of caribou exploitation by Choris people was likely sustained by local availability of the animals during periods of higher resource levels. Birnirk and Thule occupants of the area could have sustained similar harvest levels either through local hunting or in inland hunts. Another problem to be considered is how almost exactly the same activity could reflect a system of very different costs and benefits. Harritt (1994:372), for example, notes very similar levels of harvesting of ringed seal (Phoca hispida) in Ipiutak and Early Western Thule occupations on Cape Espenberg. The historic pattern for exploiting that resource from Cape Espenberg was to haul tents and other gear far out onto the ice where entire families camped for sustained periods (Burch 1998b:301). Such a pattern was predicated on the use of dog traction for moving camps, hauling harvested seals, and enlarging the breadth of the hunting area. In the absence of dog traction all of those activities would consume more time and energy. Thus, while the archaeological reflection of the activities between the two periods is comparable, one might propose different underlying structural differences in the exploitation of that resource, with and without dog traction, such as size and organization of the hunting group
s or perhaps a more protracted period of seal hunting at the expense of other seasonal activities. In the examples above, the lack of consideration of differences related to dog traction produces a situation where similar archaeological remains can be interpreted as reflecting essentially the same cultural ecology. Looking outside Northwest Alaska, a study by Helmer (1992) illustrates an inverse phenomena, where the same lack of consideration exaggerates differences between two prehistoric traditions. Looking at Devon Island settlement strategies, Helmer concluded that Early Paleo-Eskimo inhabitants did not locate their settlements optimally with respect to resource distributions, while Thule people did
The Significance of Dog Traction for the Analysis of Prehistoric Arctic Societies 77
do so. In his analysis, optimal locations were those that minimized the distances to the overall range of resources. But clearly the very definition of optimality would vary for people with markedly different mobility strategies. It makes sense for highly mobile Thule foragers to situate themselves to make most efficient use of several resources from a more fixed residential base. Whereas fixed residential bases make less sense for earlier pedestrian hunters with doubled or tripled travel times and significantly less ability to either haul the proceeds from a hunt long distances or to cache and retrieve them at a later date. The utility of Helmer's analysis would be considerably improved by looking at optimality in terms of both residential versus logistic mobility strategies, in the sense discussed by Binford (1980). It is my contention that many of the other apparent differences between Dorset and Thule are exaggerated by the effects of changes in transportation and derivative effects on settlement patterns, assemblage diversity, and other areas. But that is another paper.
tory. More serious consideration of this issue has, I believe, the potential to overturn some of the main pedestals of northern prehistory. One must underscore the essential need to consider all analyses of prehistoric Arctic cultural systems in light of how dog traction, or its absence, fundamentally effects the functioning of those systems and how they might be reflected in the archaeological record. ACKNOWLEDGMENTS My thanks go to Ken Pratt for repeated substantial help and encouragement, particularly during recent very difficult times. I am also grateful to Pete Bowers for directing me to the Marco Polo reference. Robert Drozda has demonstrated to me over and over why one does not want to be a musher or put a guest cabin in the middle of a dog lot.
CONCLUSIONS Given the critical importance of the dog traction issue, several related questions need to be addressed. First, we need the ability to establish the use of dog traction in the absence of definitive technological indicators. Until that problem is solved, the date for the actual inception of this technology is educated guesswork. A first step might be an examination of skeletal deformation in modern working sled dogs and comparison to prehistoric remains. Second, differences in the technological development
of dog traction in the eastern Arctic versus the west need to be addressed. The development proceeded differently in the two areas; the east was more advanced in some respects with the early adoption of swivels and buckles but very conservative in the late use of pegged versus lashed runners. Third, if indeed I am correct in my assertion about the use of dog sleds before 1000 B.P., we need to focus on factors involved in the significant time lag before the Eskimo expansion into interior areas such as the Kobuk and Noatak rivers. Finally, there remains another nagging question. If dog traction was so advantageous to Eskimo societies that they were willing to incur the cost of adopting this technology, why was it not adopted by neighboring subarctic Athapaskan and Algonkian peoples as well? I have long believed that dog traction is the "polar bear in the living room" for Arctic archaeology. Its largely ignored presence has significant implications for virtually every previous discussion and analysis of Eskimo prehis-
78 Alaska Journal of Anthropology Volume 2, Numbers 1 - 2
REFERENCES Ackerman, R.E. 1984 Prehistory of the Asian Eskimo Zone. In Handbook of North American Indians
, Volume 5, Arctic, edited by David Damas (pp. 106-119). W. Sturtevant, General Editor. Smithsonian Institution
Press, Washington, D. C. Anderson, D.D. 1988 Onion Portage: the Archaeology of a Stratified Site from the Kobuk River, Northwest Alaska. Anthropological Papers of the University of Alaska 22(1-2). Bandi, H.J. 1969 Eskimo Prehistory. University of Alaska Press, College. Binford, L.R. 1980 Willow Smoke and Dog's Tails: Hunter-Gatherer Settlement Systems and archaeological site
Formation. American Antiquity 45(1): 4-20. Birket-Smith, K. 1929 The Caribou Eskimos: Material and Social Life and Their Cultural Position. Report of the Fifth Thule Expedition, 1921-24, Volume 5 (1-2). Gyldendal, Copenhagen. Bretschneider, E. 1910 Mediжval researches from eastern Asiatic sources: Fragments towards the knowledge of the geography and history of central and western Asia from the 13th to the 17th century. London, Paul, Trench, Trьbner & Co., Ltd. Burch, E.S., Jr. 1972 The Caribou/wild Reindeer as a Human Resource
. American Antiquity 37(3):339-368. 1975 Inter-Regional Transportation in Northwest Alaska. Anthropological Papers of the University of Alaska 17(2):1-11. 1976 Overland Travel Routes in Northwest Alaska. Anthropological Papers of the University of Alaska 18(1):1-10. 1980 Traditional Eskimo Societies in Northwest Alaska. In Alaska Native Culture and History, edited by W. Workman and Y. Kotani (pp. 253-304). Senri Ethnological Studies 4, Osaka. 1981 The Traditional Eskimo Hunters of Point Hope, Alaska: 1800-1875. North Slope Borough, Barrow. 1984 Kotzebue Sound Eskimo. In Handbook of North American Indians, Volume 5, Arctic, edited by David Damas (pp. 303-319). W. Sturtevant, General Editor. Smithsonian Institution, Washington, D. C. 1985 Subsistence Production in Kivalina, Alaska: A Twenty Year Perspective. Technical Paper No. 128. Division of Subsistence, Alaska Department of Fish and Game, Juneau. 1998a The Cultural and Natural Heritage of Northwest Alaska, Volume VII, International Affairs. Unpublished manuscript in possession of the author. 1998b The Iсupiat Eskimo Nations of Northwest Alaska. University of Alaska Press, Fairbanks. The Significance of Dog Traction for the Analysis of Prehistoric Arctic Societies 79
Chagnon, N.A. 1992 Yanomamo (4th Edition
). Harcout, Brace, Jovanovich, Orlando. Dumond, D.E. 1977 The Eskimos and Aleuts. Thames and Hudson, London. Field, H. and E. Prostov 1942 Results of Scientific Investigations in Siberia, 1940-1941. American Anthropologist 44(3):388-406. Ford, J.A. 1959 Eskimo Prehistory in the Vicinity of Point Barrow, Alaska. Anthropological Papers of the American Museum of Natural History (47)1. Gerlach, S.C. and O.K. Mason 1992 Calibrated Radiocarbon Dates and Cultural Interaction in the Western Arctic. Arctic Anthropology 29(1):54-81. Giddings, J.L. 1952 The Arctic Woodland Culture of the Kobuk River. University Museum Monographs. University of Pennsylvania
, Philadelphia. 1960 The Archaeology of Bering Strait. Current Anthropology 1(2)121-138. 1964 The Archaeology of Cape Denbigh. Brown University
Press, Providence. Giddings, J.L. and D.D. Anderson 1986 Beach Ridge Archeology of Cape Krusenstern: Eskimo and Pre-Eskimo Settlements Around Kotzebue Sound, Alaska, National Park Service
Publications in Archeology, 20. U.S. Department of the Interior, Washington, D.C.
Hall, E.S. 1978 Technological Change in Northern Alaska. In Archaeological Essays in Honor of Irving B. Rouse, edited by R.C. Dunnell and E.S. Hall, Jr. (pp. 209-229). Mouton, The Hague. Hall, E.S. and S.C. Gerlach 1988 The Later Prehistory of Northern Alaska: The View from Tukuto Lake. In The Late Prehistoric Development of Alaska's Native People, edited by R.D. Shaw, R.K. Harritt, and D.E. Dumond (pp. 107136). Aurora: Alaska Anthropological Association Monograph Series, No. 4. Alaska Anthropological Association, Anchorage. Halstead, P. and J. O'Shea 1989 Introduction: Cultural Responses to Risk and Uncertainty. In Bad Year Economics: Cultural Responses to Risk and Uncertainty, edited by P. Halstead and J. O'Shea (pp. 1-7). Cambridge University Press, Cambridge. Harritt, R.K. 1994 Eskimo Prehistory on the Seward Peninsula, Alaska. Resources Report NPS/ARORCR/CRR-93/21. National Park Service, Alaska Region, Anchorage. Helmer, J.W. 1992 Prehistoric Site Location Strategies in the North Devon Lowlands, High Arctic Canada. Journal of Field Archaeology 19: 291-313. 80 Alaska Journal of Anthropology Volume 2, Numbers 1 - 2
Hickey, C.G. 1979 The Historic Beringian Trade Network: Its History and Origins. In Thule Eskimo Culture: An Anthropological Retrospective, edited by Allen P. McCartney, pp. 411-434. Mercury Series Paper No. 88. Canadian Museum of Civilization, Archaeological Survey of Canada, Hull, Ontario. Jenness, D. 1940 Prehistoric Culture Waves from Asia to America. Smithsonian Institution Annual Report for 1940, pp. 383-396. Smithsonian Institution Press, Washington, D.C. Mason, O.K. 1998 The Contest between Ipiutak, Old Bering Sea, and Birnirk Polities and the Origin of Whaling during the First Millennium A.D. along Bering Strait. Journal of Anthropological Archaeology 17: 240-325. Mathiassen, T. 1928 material culture
of the Iglulik Eskimos. Report of the Fifth Thule Expedition, 1921-24, Volume 6 (1). Gyldendal, Copenhagen. McCartney, A.P. 1984 History of Native Whaling in the Arctic and Subarctic. In Arctic Whaling: Proceedings of the International Symposium on Arctic Whaling, February 1983, pp. 79-112. Arctic Centre, University of Groningen
, Groningen. McCullough, K. 1989 The Ruin Islanders. Mercury Series Paper No. 141. Canadian Museum of Civilization, Archaeological Survey of Canada, Hull, Ontario. Murdoch, John 1892 Ethnological Results of the Point Barrow Expedition. In 9th Annual Report of the Bureau of American Ethnology for the Years 1887-1888 (pp. 19-441). U.S. Government Printing Office
, Washington, D.C. Polo, M. 1958 The Travels of Marco Polo. First Published in 1299. Orion Press, New York. Pospisil, L. 1993 "I Am Very Sorry I Cannot Kill You Anymore": War and Peace among the Kapauku. In Studying War: Anthropological Perspectives, edited by S.P. Reyna and R.E. Downs, pp. 113-126. Gordon and Breach, Langhorne, PA. Rainey, F.G. 1941 Eskimo Prehistory: the Okvik Site on the Punuk Islands. Anthropological Papers of the American Museum of Natural History (37), Part 4. Schnirelman, V.A. 1994 Cherchez Le Chien: Perspectives on the Economy of the Traditional Fishing-oriented People of Kamchatka. In Key Issues in Hunter-Gatherer Research, edited by E.S. Burch, Jr. and Linda J. Ellanna, pp. 169-188. Berg, Oxford. Sheehan, G.W. 1995 Whaling, Surplus, Trade, and War and the Integration of Prehistoric Northern and Northwest Alaskan Economies, A. D. 1200-1826. In Hunting the Largest Animals: Native Whaling in the Western Arctic and Subarctic, edited by A.P. McCartney, pp.185-206. Canadian Circumpolar Institute, University of Alberta, Edmonton. The Significance of Dog Traction for the Analysis of Prehistoric Arctic Societies 81
Sheppard, W.L. 1983 Continuity and Change in Norton Sound: historic sites
and Their Contexts. Occasional Paper
No. 37. Anthropology and Historic Preservation, Cooperative Park Studies Unit, University of Alaska, Fairbanks. 1986 Variability in Historic Norton Bay Subsistence and Settlement. Ph. D. Dissertation, Department of Anthropology, Northwestern University, Evanston. 1988 The Later Prehistory of Norton Sound as Viewed from Norton Bay. In Late Prehistoric Development of Alaska's Native People, edited by R.D. Shaw, R.K. Harritt, and D.E. Dumond (pp. 137-147). Aurora: Alaska Anthropological Association Monograph Series, No. 4. Alaska Anthropological Association, Anchorage. VanStone, J.W. 1955 Archaeological Excavations at Kotzebue, Alaska. Anthropological Papers of the University of Alaska 3(2):75-155. 82 Alaska Journal of Anthropology Volume 2, Numbers 1 - 2