The most common form of permanent housing in the prehistoric period for arctic regions was the semi-subterranean winter house. First built in the American arctic about 800 BC, by the Norton or Dorset Paleo-Eskimo groups, semi-subterranean houses were essentially dugouts, houses excavated partially or completely below the ground surface to take advantage of geothermal protections during the harshest of climates.
While there are several versions of this form of house over time in the American arctic regions, and in fact there are several related forms in other polar regions (Gressbakken Houses in Scandinavia) and even in the great plains of North American and Asia (arguably earth lodges and pit houses), semi-subterranean houses reached their highest pinnacle in the Arctic. The homes were heavily insulated to ward off the bitter cold, and constructed to maintain both privacy and social contact for large groups of people despite that harsh climate.
Semi-subterranean houses were built of a combination of cut sod, stone, and whalebone, insulated with sea mammal or reindeer skins and animal fats and covered with a bank of snow. Their interiors possessed cold-traps and sometimes dual seasonal entrance tunnels, rear sleeping platforms, kitchen areas (either spatially discrete or integrated into the main living area) and various storage areas (shelves, boxes) for stowing food, tools and other household goods. They were large enough to include members of extended families and their sled dogs, and they were connected to their relatives and the rest of the community via passageways and tunnels.
The real genius of semi-subterranean homes, however, resided in their layouts. At Cape Espenberg, Alaska, a survey of beach ridge communities (Darwent and colleagues) identified a total of 117 Thule-Inupiat houses, occupied between 1300 and 1700 AD. They found the most common house layout was a linear house with one oval room, which was accessed by a long tunnel and between 1-2 side spurs used as kitchens or food-processing areas.
Layouts for Community Contact
A substantial minority, however, were multiple large-roomed houses, or single houses built side-by-side in groups of four or more. Interestingly, the house clusters, with multiple rooms and long entrance tunnels are all more common attributes at the early end of occupation at Cape Espenberg. That has been attributed by Darwent et al. to a shift from a dependence on whaling to localized resources, and the transition to a sharp downturn in climate called the Little Ice Age (AD 1550-1850).
But the most extreme cases of below-ground communal connections in the Arctic was during the 18th and 19th century, during the Bow and Arrow Wars in Alaska.
The Bow and Arrow Wars
The Bow and Arrow wars were a long-lasting conflict between different tribes including the Alaskan Yup'ik villagers. The conflict could be compared to the 100 Years War in Europe: Caroline Funk says it imperiled lives and made legends of great men and women, with a range of conflicts from deadly to merely threatening. Yup'ik historians do not know when this conflict started: it may have begun with the Thule migration of 1,000 years ago and it may have been instigated in the 1700s by competition for long distance trading opportunities with the Russians. Most likely it began at some point in between. The Bow and Arrow Wars ended at or just prior to the arrival of Russians traders and explorers in Alaska in the 1840s.
Based on oral histories, subterranean structures took on a new importance during the wars: not only did people need to conduct family and communal life inside because of weather demands, but to protect themselves from attack. According to Frink (2006), historic period semi-subterranean tunnels connected the members of the village in an underground system. The tunnels - some as long as 27 meters - were formed by horizontal logs of planks shored up by short vertical retainer logs. Roofs were constructed of short split logs and sod blocks covered the structure. The tunnel system included dwelling entrances and exits, escape routes and tunnels that linked village structures.
Coltrain JB. 2009. Sealing, whaling Journal of Archaeological Science 36(3):764-775. doi: 10.1016/j.jas.2008.10.022and caribou revisited: additional insights from the skeletal isotope chemistry of eastern Arctic foragers.
Darwent J, Mason O, Hoffecker J, and Darwent C. 2013. 1,000 Years of House Change at Cape Espenberg, Alaska: A Case Study in Horizontal Stratigraphy. American Antiquity 78(3):433-455. 10.7183/0002-73184.108.40.2063
Dawson PC. 2001. Interpreting Variability in Thule Inuit Architecture: A Case Study from the Canadian High Arctic. American Antiquity 66(3):453-470.
Frink L. 2006. Social Identity and the Yup'ik Eskimo Village Tunnel System in Precolonial and Colonial Western Coastal Alaska. Archeological Papers of the American Anthropological Association 16(1):109-125. doi: 10.1525/ap3a.2006.16.1.109
Funk CL. 2010. The Bow and Arrow War days on the Yukon-Kuskokwim . Ethnohistory 57(4):523-569. doi: 10.1215/00141801-2010-036delta of Alaska
Harritt RK. 2010. Variations of Late Prehistoric Houses in Coastal Northwest Alaska: A View from Wales. Arctic Anthropology 47(1):57-70.
Harritt RK. 2013. Toward an archaeology of late prehistoric Eskimo bands in coastal northwest Alaska. Journal of Anthropological Archaeology 32(4):659-674. doi: 10.1016/j.jaa.2013.04.001
Nelson EW. 1900. The Eskimo about Bering Strait. Washington DC: Government Printing Office. Free download