By: William Fitzhugh. Originally Published in ASC Newsletter No. 22, pg 29-33.
Harp seals have been intertwined with human history ever since people began living along the Arctic and Subarctic shores of the North Atlantic. Harps were quite likely a resource for Upper Paleolithic cultures of Europe and for hundreds of years and more recently have been a mainstay for Saami, Finns, and Russians living around the White Sea. In the Northwest Atlantic, harps have been important for Maritime Archaic Indian cultures between 8000-4000 years ago from Maine to northern Labrador and have sustained Paleoeskimo, Inuit and Innu peoples who occupied the regions in the Canadian Eastern Arctic and Greenland south to Newfoundland and the northern Gulf of St. Lawrence. The latter regions have been investigated by the Smithsonian for more than thirty years. The presence or absence of Harp seals may have been a major factor, along with climate change, for cultural migrations and boundary changes between these culturally-distinct populations.
Field Program 2014
Testing this hypothesis became the focus of a sub-project of the ASC’s “Arctic Crashes” project in 2014-2015 as part of the author’s on-going research into Eskimo and Inuit culture development in Labrador, Newfoundland, and the Quebec Lower North shore. Arctic Crashes is exploring the causes and effects of fluctuations in northern animal populations and its impact on human societies. The recent discovery of Inuit winter occupations on the LNS west of Blanc Sablon has provided a new data-set with which to test the climate/pack-ice/southern Eskimo migration model in which three culturally and chronologically distinct Eskimo/Inuit groups occupied—and then abandoned—the northern Gulf of St. Lawrence and Island of Newfoundland: Groswater Paleoeskimo 2500-2200 BP; Newfoundland Dorset 1800-1400 BP; and Labrador Inuit AD 1500-1750.
Hart Chalet site under excavation by Alaina Harmon and Mariel Kennedy in August 2014. View north through entry passage into house interior.
With support from the Smithsonian’s Grand Challenge Program, we conducted field surveys and excavations in July and August 2014, from Hamilton Inlet (Labrador) to Brador and St. Paul Bay on Quebec’s LNS. Fieldwork was facilitated by the ASC’s research vessel Pitsiulak was staffed by a field team including Alaina Harmon and Notre Dame student Marielle Kennedy. Ted Timreck and Sandra Kingsbury produced video documentation for the ASC and NMNH’s Q?rius Education Center. Our activities concentrated on excavations at the Hart Chalet Inuit winter village site near Brador (Quebec) where we spent ten days conducting excavations in Houses 1 and 2 and recovered a large sample of bone and shell midden material dated to ca. 1700. This sample is now being analyzed by Claire St. Germaine of University of Montreal. After species identifications have been made we will be submitting samples for isotopic analysis to determine water temperature and other characters suitable for environmental reconstruction. (See here for related Crashes studies of the paleo-marine environment conducted by Walter Adey and colleagues.)
Thirty years ago when we identified major north-south movements in Labrador’s Eskimo-Indian boundaries, correlations between these changes and climate cycles identified in the pollen records and Greenland ice cores suggested climate as the primary causal factor. The correlation was particularly strong with the distribution of Eskimo groups, who were heavily dependent on sea ice and its associated fauna. The mechanism suggested was shifts in the duration and southward extent of seasonal pack ice. Cooler weather brought more pack ice south and produced longer winters in coastal regions. Eskimo resources that came with the pack ice were ring, harp, bearded, and bladdernose seals, and walrus and bowhead whales. We also knew that the historic period Labrador Inuit had expanded their whale-hunting culture into areas of central and southern Labrador formerly occupied by the Innu. But earlier Paleoeskimo groups like the Dorset and Pre-Dorset were walrus and seal hunters, not whalers. Dorset Paleoeskimos expanded far south of the Thule/Labrador Inuit boundary, occupying the entire Island of Newfoundland and nearby northeastern shore of the Gulf of St. Lawrence. What was it about the pack ice that enabled this Dorset expansion about 2000 years ago, as well as an earlier Groswater expansion, also including all of Newfoundland around 600 B.C.? Walrus remains are not common in Groswater and Dorset sites in Newfoundland, but harp seals are present in great numbers. The more that we researched the question of early Eskimo expansions and retreats from their maximal southern limits, the more it became evident that the answer must be found in changes in the distribution of pack ice and harp seals.
Harp Seal Biology and Ecology
Harp seals are the most abundant marine mammal in the northwestern Atlantic—some 6-9 million animals. Their biology, ecology, and migratory behavior (Sargeant 1991) have been investigated in detail due to the species’ economic importance to traditional and commercial hunters from Greenland to Newfoundland, and because of the controversy over the commercial hunt of its new-born ‘whitecoats’ around Newfoundland and in the Gulf of St. Lawrence. Harp seals migrate annually from Baffin Bay and Davis Strait in large companies of 20 to 100 or more individuals. The migration strikes the northern Labrador coast in late October or November and proceeds south in waves, with animals hugging the shore and entering the exact same bays and island passages year-after-year just as ice begins to form. Labrador Thule and 16-18th C. Labrador Inuit sites contain large numbers of harp seal bones. During the 19-20th C. thousands of harps were caught annually by Inuit and Europeans with rifles and nets along the Labrador coast and the Quebec Lower North Shore. A Newfoundland hunt (both traditional and commercial) for adult harps and white-coats has been conducted off-shore on the floating pack-ice by ship-borne hunters since the mid-19th century.
The main mass of the harp migration takes several weeks to pass any given location. Reaching southern Labrador, part of the herd remains on the newly-formed pack ice east of southern Labrador and northern Newfoundland in a region called “The Front.” The other segment passes with the drifting ice through the Strait of Belle Isle into the Gulf of St. Lawrence. Part of this group hugs the Quebec coast west to Natashquan and Mingan where they remain a few weeks feeding before turning south to their birthing area on the ice floes north of the Magdalen Islands. The rest of the Gulf herd passes south along the west coast of Newfoundland before re-grouping north of the Magdalens. They remain here and in other areas of stable ice throughout the winter. In February and March, the females give birth on the ice to pups known as white-coats. The mothers tend and feed their pups for several weeks as they cannot feed themselves or even dive because the thick furry white coats that keep them from freezing on the open ice are too buoyant. When their blubber has thickened and the white-coats have been replaced by shorter hair, they begin to swim and feed on their own.
In April, the adult harps gather again, this time to bask in the sun and to moult, and when the pack ice melts in April and May, they head north in small companies. Adults leave first, then the young, following a hydrographic feature known as the “Eskimo Channel” that parallels the west coast of Newfoundland. It is this northward migration that was the primary target of Port au Choix Dorset hunters, as their route passes close to shore at Pointe Riche. After leaving Newfoundland the migration is generally too far off-shore in the outer pack off Labrador to be accessible to shore-based hunters and reaches the summering grounds around Greenland and Baffin Bay in June and July.
Like the arrival and departure of geese and of salmon, the harp migration was a relatively dependable phenomenon during the historical period. Catch statistics varied considerably though, as a result of variable hunting access due to storms, dangerous ice, or inaccessible locations far from shore. Throughout the historical era the harp catch was a crucial early winter and spring resource to the Inuit, European settlers, and some Indian groups throughout Labrador, along the Quebec LNS, and northern and western Newfoundland. When unavailable due to population decline, abandonment of the Gulf, or inaccessibility, the loss of harp seals caused hardship for European settlers, and for traditional cultures, it could spell disaster.
Whale bone and caribou antler implements from Hart Chalet Inuit winter village site (House 1).
It has long been known that the economy of the Phillip’s Garden Dorset site at Port au Choix, one of the largest Paleoeskimo sites in the Eastern Arctic and Subarctic, was based predominantly on harp seals. This dependence, particularly at the key site of Port au Choix, has led to speculation that a change in migration route or a precipitous population crash may have caused the site’s abandonment, and subsequently, in a domino-like effect, the disappearance of Dorset culture throughout the rest of Newfoundland (Bell and Renouf 2008, 2011; Renouf and Bell 2009). In earlier years, the discussion was all about the ice—how close and how thick it was; where was it moving; and how to get to it—because this was where harp seals congregated. Every year conditions varied from region-to-region, but western Newfoundland in early spring was where seals could be expected most dependably (Hodgetts (2003, 2005; Hodgetts et al. 2003), especially at Port au Choix where the cape bordered the Harp seal migration north following the Eskimo Channel (LeBlanc 1996, 2000). For many years local hunters have reported that shifting spring winds and currents in the Gulf ice sometimes caused harp migrations to shift from western Newfoundland across to the Quebec Lower North Shore, taking the animals out of reach of Newfoundland hunters (D. Sargeant pers. comm. 1972). Similarly, LNS hunters frequently speak about winters when harps become unavailable during their early winter migration because of lack of ice or from ice having been blown too far off-shore to reach with small boats (pers. comm. with Harrington Harbor hunters, 2001-10; Murray 2011).
Archaeozoological studies have made cultural and environmental reconstructions more specific. Hodgetts et al. (2003), citing a decreasing percentage of harp seal bones and diversification of diet to include more fish and birds in the later Dorset components at Port au Choix, suggest a broadening of the diet and less dependence on harp seals than in earlier years. Changes like this could be a response to reduced harp seal availability. Citing chronomid midge frequency changes in sediments from nearby Bass Pond, (Rosenberg et al. 2005) suggested that terrestrial warming at Port au Choix peaked at 1100 BP, coincident with the end of the Dorset occupation. Marine pollen transfer function studies off southwestern Newfoundland (Levac 2003) indicated a warming of Gulf waters at this time. Based on these studies, Renouf and Bell (Renouf and Bell 2009; Bell and Renouf 2011:37) speculated that climate warming may have undermined sea ice conditions and destabilized the harp seal population and its migration routes, ending Dorset tenure at Port au Choix, and through cascade effects, severing Dorset contacts with Labrador and bringing an end to Dorset culture throughout Newfoundland.
Seal mandible from Hart site H1 midden which produced shells and large amounts of caribou bones.
Based on observations of the past few years, a variation of this hypothesis may be suggested that more explicitly links advances and retreats of Groswater, Dorset, and southern Inuit occupations south of Cartwright to cycles of harp seal availability. Johnston et al. (2005, 2012) report that, since 1996, the formation of pack ice in the Gulf has declined dramatically, such that in many areas there is no ice at all, and where it is present it is weak and breaks up in storms. This situation has become even more dramatic since 2007 and has been widely reported in the press. If ice thins or disappears before the white-coats have molted, they usually drown. The winters of 2010-2012 in the northern Gulf were so mild that many areas had no ice, and female seals had to give birth in the water or on shore. When this happens pups drown or are abandoned and die on shore or are lost to gulls and other predators. Poor ice conditions are thought to have resulted in a large losses of pups in 1981, and in 1998-2005.
Hart Chalet House 1 west midden, where shells, mammal and fish bones were recovered.
In July 2010, during fieldwork on the Quebec LNS, we found harp pup carcasses on-shore, and local hunters told of “thousands” dying in the vicinity of their villages. Without the winter ice platform, wildlife officials cannot conduct aerial population counts, so the effect of these recent low-ice winters on the population is not easily quantified. Johnston et al. (2005) documented a significant reduction in sea ice cover on the east coast of Canada since 1995. These data show cyclicity in ice presence and absence that seems to be keyed to the North Atlantic Oscillation. A more recent study (Johnston et al. 2012) using satellite photography has shown that “warming in the North Atlantic over the last 32 years has significantly reduced winter sea ice cover in harp seal breeding grounds, resulting in sharply higher death rates among seal pups in recent years.” This study found that seasonal sea ice cover in all four harp seal breeding regions around the North Atlantic has declined by up to 6 percent each decade since 1979, when satellite records of ice conditions began, and that in low ice years virtually all the young of the year die. Whether the current pattern will persist long enough to have a significant impact on harp seal population remains to be seen, because these losses can take a decade to have an effect, after the current cohorts reach sexual maturity. If the ice does not return, the Gulf portion of the herd will decline or disappear, and the remaining animals will have to shift to the Labrador Front or to other locations where pack ice remains. If this happens, it will result in the loss of the most dependable marine mammal resource in the eastern Gulf and the one that has been the sustaining resource for southern Dorset and Inuit population extensions. Its negative impact on Labrador Eskimo populations would diminish northwards, since harps would still be migrating south, though in smaller numbers, to whelp on the Labrador Front. Its importance to Maritime Archaic and later Indian populations is difficult to determine, because their economies were more diversified, judging from their settlement systems and rare instances when faunal remains or organic tools have survived.
Ice cover is the sine qua non for harp seal availability in the Gulf. Warmer temperatures, both of sea water and air, have been steadily reducing the winter and spring build-up and persistence of pack ice in the Labrador Current. Owing to the narrow and shallow Strait of Belle Isle most of this winter ice does not enter the Gulf but rather follows the south-moving Labrador Current along the northeast coast of Newfoundland. For this reason the amount of Gulf pack ice that forms is mostly dependent on local conditions, especially wind and temperature, which can vary depending on whether air masses are Arctic or Atlantic in origin. For the past several years conditions have produced little or no ice, and a strong correlation has been found between Gulf ice and the North Atlantic Oscillation (Johnston et al. 2005, 2012). According to this research we may expect the trend toward low ice years in the Gulf to continue for some time. Since rising temperatures are generally thought not to have reached the peaks known from the Hypsithermal or Medieval Warm levels, the loss of ice in the Gulf in recent years suggests that these waters may have been free of winter ice even in periods of moderate warmth. If so, the Gulf harp herd may be seen as a marginal or episodic population that comes and goes in step with climatic cycles. While the loss of the Gulf harp population may not have serious consequences for Labrador and possibly eastern Newfoundland, which are ‘upstream’ in the harp southern migration, it would cripple intensive adaptations to this resource in the northern and eastern Gulf. As a result, it seems likely that climatic conditions controlling the appearance and disappearance of winter ice in the Gulf have also governed whether cultures with a high degree of dependence on this one marine resource, most particularly Groswater and Dorset Paleoeskimo and Historic 17-19th C. Inuit cultures, could survive here over the long-term. There is therefore a good chance that these climate/ice/seal cycles explain the southern Groswater expansion and at least the disappearance of Newfoundland Dorset. Absence of large, dependable harp populations in the Gulf and around Newfoundland may also offer a possible explanation for the dominance or resurgence of Indian cultures on the Central Labrador coast during warm climatic periods.
New research techniques and more local studies are beginning to allow us to investigate these issues. The development of more paleoenvironmental records from Newfoundland noted above have contributed to understanding human-environmental interactions in the island’s prehistory (Bell and Renouf 2008; Renouf and Bell 2009). New studies from the Gulf that document changes in the annual monthly duration of sea ice cover in the Gulf, on the Labrador coast, and around Newfoundland will provide the key data for substantiating the hypothesis presented here. Recent studies of corraline algae, a slow-growing coral-like species that formed encrustations on underwater rocks, has provided information on marine climate along the Labrador and Newfoundland coasts (Halfar et al. 2014) that begins to corroborate other proxies with data specifically keyed to seasonal sea ice duration and overall reductions in southern extent of pack ice. If physical conditions can be correlated with modern population numbers we may have a solid foundation for understanding southern Eskimo territorial expansions and retreats. Another line of inquiry presently being followed is reconstruction of local paleo-marine temperatures from isotopic studies of harp and other marine mammal bones from dated archaeological deposits. These tests are currently being conducted under the ASC’s ‘Arctic Crashes’ project using fauna from our Labrador and Lower North Shore (Quebec) collections.
For more on our Arctic Crashes project, please visit our website, and check out the other posts on this blog.
Bell, Trevor, and M.A.P. Renouf
2008 The Domino Effect: Culture Change and Environmental Change in Newfoundland, 1500-1100 cal. BP. The Northern Review 28:72-94.
2011 By Land and Sea: Landscape and Marine Environment Perspectives on Port au Choix. In The Cultural Landscapes of Port au Choix: Precontact Hunter-Gathers of Northwestern Newfoundland, edited by M.A.P. Renouf, pp. 21-41. Springer.
Halfar, Jochen, Adey, Walter H., Kronz, Andreas, Hetzinger, Steffen, Edinger, Evan, and Fitzhugh, William W.
2014 Arctic sea-ice decline archived by multicentury annual-resolution record from crustose coralline algal proxy. Proceedings of the National Academy of Sciences, 110(49): 19737-19741. doi:10.1073/pnas.1313775110.
Hodgetts, Lisa M.
2005 Using Bone Measurements to Determine the Season of Harp Seal Hunting at the Dorset Palaeoeskimo Site of Phillip’s Garden. Newfoundland and Labrador Studies 20(1):91-106.
2007 The Changing Pre-Dorset Landscape of Southwestern Hudson Bay, Canada. Journal of Field Archaeology 42 (4) 353-367.
Hodgetts, L.M., M.A.P. Renouf, M.S. Murray, D. McCuaig-Balwil, and L. Howse
2003 Changing Subsistence Practices at the Dorset Palaeoeskimo site of Phillip’s Garden, Newfoundland. Arctic Anthropology 40(1):106-120.
Johnston, David W., Ari S. Friedlaender, L.G. Torres, and David M. Lavigne
2005 Variation in Sea Ice Cover on the East Coast of Canada from 1969 to 2002. Climate Variability and Implications for Harp and Hooded Seals. Climate Research 29:209-222.
Johnston David W., Matthew T. Bowers, Ari S. Friedlaender,and David M. Lavigne
2012 The Effects of Climate Change on Harp Seals (Pagophilus groenlandicus). PLoS ONE, 7(1): e29158 DOI: 10.1371/journal.pone.0029158)
1996 A Place with a View: Groswater Subsistence-Settlement Systems in the Gulf of St. Lawrence. MA Thesis, Department of Anthropology, Memorial University of Newfoundland, St. John’s.
2000 Groswater Technological Organization: A Decision-Making Approach. Arctic Anthropology 37(2):23-37.
2003 Palynological Records from Bay of Islands, Newfoundland: Direct Correlation of Holocene Paleoceanographic and Climatic Change. Palynology 27:135-154.
Murray, Maribeth S.
2011 Whitecoats, Beaters, and Turners: Dorset Paleoeskimo Harp Seal Hunting from Phillip’s Garden, Port au Choix. In The Cultural Landscapes of Port au Choix: Precontact Hunter-Gathers of Northwestern Newfoundland, edited by M.A.P. Renouf, 209-226. Springer.
Renouf, M.A.P., and Trevor Bell
2009 Contraction and Expansion in Newfoundland Prehistory, AD 900-1500. In The Northern World AD 900-1400, edited by Herbert Maschner, Owen Mason, and Robert McGhee, pp. 263-278. Salt Lake City: University of Utah Press.
Renouf, M.A.P., Trevor Bell, and M.A. Teal
2000 Making Contact: Recent Indians and Paleoeskimos on the Island of Newfoundland. In Identities and Cultural Contacts in the Arctic, edited by Martin Appelt, Joel Berglund, and H.-C. Gulløv, pp. 106-119. Danish National Museum and Danish Polar Center. Copenhagen.
Renouf, M.A.P., Michael A. Teal, and Trevor Bell
2011 In the Woods: the Cow Head Complex Occupation of the Gould Site, Port au Choix. In The Cultural Landscapes of Port au Choix: Precontact Hunter-Gathers of Northwestern Newfoundland, edited by M.A.P. Renouf, pp. 251-269. Springer.
Rosenberg, S.M., I.R. Walker, and J.B. MacPherson
2005 Environmental Changes at Port Au Choix as Reconstructed by Faunal Midges. Newfoundland and Labrador Studies 20(1):57-73.
Sargeant, David E.
1991 Harp Seals, Man, and Ice. Canada Special Publication of Fisheries and Aquatic Sciences, 114. Canada: Department of Fisheries and Oceans. Ottawa.