« Previous Region | Index | Next Region »

Deschutes Region

Table of Contents:

Deschutes Housing:
Northern Paiute

Ochoco National Forest
Laughlin Hills
Newberry Crater
Ferry Canyon
Paquet Gulch 35WS125
Heath Cliffs
Pithouse I and II
Dudley Site 35CR958

Lewis and Clark mention a number of villages on their travels along this stretch of the river. After leaving The Dalles, they noted a small Chiluckittequaw village of eight houses on the right bank. On Oct 29, 1805 they visited a village of seven houses on the right bank that may be The Friendly Village Site tested by Cole (1974:7). They then went to a village of eleven houses, probably at the mouth of the Klikitat River. Lower Memaloose Island was noted to contain thirteen burial vaults. Two more houses were passed on the right bank at about Major Creek and two houses at about Mosier, that may be Smackshops Village. At the mouth of the Hood River were four houses, the first they had seen on the left bank. Fourteen houses were noted on the right in the vicinity of Canoe Creek. On their return trip they noted twenty houses scattered over several miles at this location. Four houses were noted in the vicinity of Mitchell Point. They camped at a three-house village on Oct 29, 1806. They then passed a village of eight houses which had been moved to the other side of the river on their return trip. On April 13 the village consisted of eleven houses, indicating the rapid movement of structures from place to place and across the Columbia River. On Oct 30, 1805, Clark followed an old Indian path to an abandoned village containing very large houses constructed differently than any he had seen before. The next day he noted an old village of eight houses and a second very old village with the remains of only one house still visible and nearby eight burial vaults (Cole 1974:7-9).

[back to top]


"Since 1805, at least three types of housing have been described for the Dalles-Celilo Falls area. The descriptions of the mat, plank or earth lodges indicate a considerable variation in construction. On their maps, Lewis and Clark indicate that the lodges were of mat construction at the various village and camping areas along the river as far west as the Short narrows (Ten Mile Rapids). In the large village at the head of the Long narrows (Five Mile Rapids), they met with people of a new nation, the Ech-e-lute (Skillute, Wishram, Upper Chinook) and saw the first wood houses since they had left Illinois. Clark described the construction of the houses as follows (spelling as in original):

those houses are about the Same Shape Size and form 20 feet Square, (wide and 30 feet long) with one Dore raised 18 inches above the ground, they (the doors-Ed.) are 29-1/2 inches high & 14 wide, forming in a half circle above, those houses were Sunk into the earth six feet, the roofs of them was Supported by (a ridge pole resting on) three Strong pieces of Split timber on which, (thro' one of which the dore was cut) that and the walls, the top of which was just above ground Supported a certain number of Spars which are Covered with the bark of the white Ceader, or Arber Vitae; and the whole attached and Secured by the fibers of the cedar. the eaves at or near the earth, the gable ends and Side Walls are Secured with split boards which is Seported on iner Side with Strong pieces of timber under the eves &c. to keep those pieces errect & the earth from without pressing in the boards, Suported by Strong posts at the corners to which those poles were attached to give aditional Strength, small openings were left in the roof above the ground, for the purpose, as I conjectured, of deschargeing their arrows at a besieging enimey; Light is admited thro' an opening at top which also Serves for the Smoke to pass through, one half of those houses is apropriated for the storeing away Dried & pounded fish which is the principal food, the other part next the dore is the part occupied by the nativs who have beds raised on either side, with a fire place in the center of this Space each house appeared to be occupied by about three families; that part which is apropriated for fish was crouded with that article, and a fiew baskets of burries (Thwaites 1905, III: 154-155)" (Wilke, Wilde, Dalan, James, Weaver & Harvey 1983: 92-93).

Salmon drying at The Dalles
Wilke & Others 1983:86

Clark noted that the timber for houses was obtained in the higher elevations and rafted down to the Columbia River (Wilke, Wilde, Dalan, James, Weaver & Harvey 1983: 94).

"The missionary Daniel Lee, who was at The Dalles Methodist Mission form 1838 to 1843, described similar wood housing for the people from The Dalles (Long Narrows) to the Cascades, apparently including the Wasco village of Cathlasco or Wasco on the south side of the river.

They are covered with bark, and the walls are made of boards split out of the cedar. The boards are set on end, and support the plates, rafters, and ribs. The floor is about three feet below the ground, and the fire-place in the middle, sometimes a little lower, and the smoke escapes through an opening at the top. Several families often unite together and occupy one house. Twenty or thirty are sometimes seen living together thus (Lee and Frost 1968: 180-181)" (Wilke, Wilde, Dalan, James, Weaver & Harvey 1983: 93-94).

"Most information that is available on housing and habitation patterns in The Dalles-Celilo Falls area is based on the writings of Curtis and Spier and Sapir regarding the Wishram. Spier and Sapir write that two forms of housing were used by the Wishram, a 'semi-subterranean earth lodge' used mainly in the winter, and a mat lodge" (Wilke, Wilde, Dalan, James, Weaver & Harvey 1983: 94).

"According to Spier and Sapir, the earth lodge was described as 'hemispherical superstructure built over a circular pit.' The size of the lodge depended on the number of people living in the house, which varies from one to six families. The diameter of the pit was described as 16 or more feet with a depth of from 4 to 6 or more feet. The framework '... was of poles, covered with tule mats, grass, and dirt, or with cedar bark.' The framework was not described in detail, but Spier and Sapir assumed that it resembled that of houses of neighboring peoples with two or more central posts supporting ridge poles that held poles that extended from the edges of the pit. Access to and from the lodge was by a hole in the roof with a ladder to the floor of the lodge. The floor and the sides of the lodge were lined with mats and low bed platforms were placed against the walls. The space under the bed platforms was used for dried food storage (Spier and Sapir 1930: 202)" (Wilke, Wilde, Dalan, James, Weaver & Harvey 1983: 95).

[back to top]

Northern Paiute:

"In the Great Basin, the typical winter house was a dome-shaped structure. The frame was formed by bending willow poles and lashing them together. It was then covered with bark, pine needles, earth, and/or rushes. In the summer a simple 'lean-to' structure was often used" (Hartmann & Honey 1977:5-6).

[back to top]


"The Tenino consisted of four separate groups. Each group was an independent unit and had its own distinct territory. However, movement between groups was unrestricted. Each group had two village sites. One was along the Columbia River which was occupied during the fishing season. Dwellings at this site were rectangular above ground structures, constructed of poles and mats. Half of the house was used as a living area while the other half was used to dry salmon. The winter villages were several miles away from the Columbia River and occupied from November to March. Here each family owned two structures. The most important was the oval semi-subterranean earth lodge which provided shelter and was the sleeping quarters. The other structure was the rectangular pole and mat lodge which had been used during the summer. This structure was now used for cooking and other daytime activities" (Hartmann and Honey 1977:7-8).


[back to top]

Ochoco NF Timber Sale, Crook County

In a timber sale in Crook county on the Ochoco National Forest, Hartmann and Honey (1977) recorded two sites each with two shallow depressions, one site with a single shallow depression, and one site with "many" shallow depressions that may have been housepits. No measurements were taken on any of the depressions.

[back to top]

Laughlin Hills Village, Wasco County

Follansbee (1979) recorded a large village site on a very minor stream of a large creek flowing out of the Laughlin Hills in Wasco county. Twenty-eight housepits were described, two separated from the main cluster. The depressions averaged twenty-one feet in diameter and the two smallest about eight feet. All of the houses were at an elevation of about 2350 feet. Jan Peterson, then archaeologist with BPA, felt the location of the village may have been defensive and that it may have been placed at this location after the horse was introduced. The location of the site did not correspond to any winter village pattern associated with the Tenino, Mollala or Northern Paiute. The site is located in an oak grove and is close to a major root gathering area. Twenty-six of the houses were grouped inside an area that was roughly circular in form, the house scattered fairly evenly apart. The site was determined eligible on Feb 21, 1980.

[back to top]

BPA Project 35WS124

During a BPA project with OSU in 1980, site 35WS124 was recorded. Between 7 to 10 depressions were observed with scattered lithic debris. No measurements were taken or put onto the site form (Snyder, Robbins, Honey & Draper 1980).

[back to top]

Newberry Crater House

The 1997 Web site for OSMA included research at Newberry Crater. They found a residential camp with a "domestic structure, containing a central fire hearth, was rimmed by the charred remains of five stout Lodgepole pine support posts. The area immediately around the hearth (i.e., the area within the structure) was relatively more rock-free than surrounding areas. Rock had been brought to the habitation area, possibly to help anchor the edges of habitation structures or for other purposes. The house floor did not appear to have been excavated below the contemporary ground surface, The interior space is estimated to have been about four meters from side to side, and up to five meters long.

"Three of the support posts from the house perimeter were radiocarbon dated, producing a weighted mean radiocarbon date of 8555±60 BP and a dendrocalibrated calendar age of 9490 years ago. Two dates from the hearth were older than the post dates by several hundred years (8880±110 and 9060±80), but the hearth contained relatively large chunks of charcoal, and this discrepancy is best explained by the use of older wood for hearth fuel than was used for the structural posts. The date derived from the posts is considered a better indicator of the time of occupation".

"The central hearth had been excavated about 30 cm below the contemporary surface, and at its widest point measured nearly a meter in diameter" (OSMA WEB Site prepared by Mark Tveskov and Tom Connolly 1997).

[back to top]

Juniper Control Project 35CR527

In a juniper control project, Marci Enneberg and Suzanne Crowley Thomas (1988) recorded site 35CR527, a "juniper structure". This consisted of medium sized juniper limbs abutted against a living tree branch. Hides or branches were then used to cover this temporary shelter. The lower sides of the branches, limbs and the living tree show signs of charring from a fire within the temporary structure. They note structures of this type were noted in 1876 by a land surveyor, Alonzo Gesner during a survey in the Bear Creek area (Enneberg & Crowley Thomas 1988: site form).

[back to top]

Upper Al Timber Sale 35CR590

Site 35CR590 was added in the Upper Al timber sale by David Stepp. The site had five depressions: 4.7 meters by 15 cm deep; 4.1 meters by 20 cm deep; 4.5 meters by 20 cm deep; 4.6 meters by 35 cm deep; and 2.5 meters by 35 cm deep. They were either house or roasting ovens according to the recorder (Stepp 1989).

[back to top]

Ferry Canyon 35WS261

Thomas (1993b) recorded site 35WS261 during the Ferry Canyon projects. The 11 depressions ranged from 5 to 8 meters across in three rows (top to bottom: 5-4-2) on three terraces overlooking a creek between two smaller drainages. Associated with the depressions were hopper mortars, metates, manos and a Late Archaic point.

[back to top]

Paquet Gulch Bridge 35WS125)

The Paquet Gulch Bridge site (35WS125) was mitigated in 1991. One of 75-100 depressions was tested, and found to be a housepit dating 2290±100 BP. "The house pit depression is oval, measuring about nine meters long (from berm top to berm top), six meters wide, and about 15 cm deep. The berm stands about 70 cm above the down sloping northeastern terrain, but is almost level with the surrounding terrain to the south and west" (Jenkins & Connolly 1994: 35).

The house had a central hearth produced a date of 1390±90 BP (Beta-45973). They felt it was clear the house was reoccupied many times over a span between 2300 and 1300 years ago. Housepit 2 was 9 by 6 meters and 60-70 cm deep, with eight pit features. "Some of the pit features in the HP2 floor were small (roughly 50 to 60 cm in diameter) and may have been post holes for the superstructure. If so, there may have been four central posts (the typical Tenino style) with cross members spanning the space between them" (Jenkins & Connolly 1994:38).

[back to top]

Heath Cliffs

OSMA excavated the Heath Cliffs site on the Warm Springs Reservation and reported it as a University of Oregon Anthropology Paper (# 53) in 1996. Block 4 excavations in the Northern Locus exposed a shallow circular housepit (Feature 7) with a fire hearth or storage pit on the house floor. A bulk carbon sample gave a 3430±130 date. The depressions was roughly five meters in diameter and 35-40 cm deep. The house was defined by the absence of cobbles and piles of cobbles in the berm around the house rim. The house was assigned to Component II, dating between 3700 - 2500 BP. A complex of living floors, hearths and storage pits from the 5500-2250 period of occupation was interpreted as multiple superimposed house floors, but no defined features could be isolated (Jenkins & Connolly 1996).

[back to top]

Pithouse I and Pithouse II dating

"Ames (1988) and Chatters (1989) have noted that radiocarbon dates associated with residence in pithouses on the southern Columbia Plateau (middle Columbia and lower Snake River area) appear to fall into three discontinuous clusters. The earliest of these periods, which Chatters (1989) identified as Pithouse I and Pithouse II, date from about 4500-3700 BP and 3300-1900 BP, respectively. The Pithouse I period is largely contemporary with Warm Springs I at the Heath Cliffs site.

Chatters (1989:9) indicates that during Pithouse I the settlement-subsistence system involved three types of sites, including pithouse (residential) sites, resource gathering locations, and short term camps. Pithouses, which were usually shallow compared with housepits of later periods, appear to have been occupied year-round. During Pithouse I, sites with house features outnumber all other types of sites by nearly 2:1. This pattern is interpreted as one of low subannual mobility (residential bases being occupied throughout the year), high supra-annual mobility (residence sites changed every few years), with relatively long periods of non-use between occupation episodes in a single residential (i.e. pithouse) site. Residential sites were situated in ecotone locations from which a diverse array of environments could effectively be exploited. Faunal assemblages indicate a broad, taxonomically rich subsistence base, relying heavily on small bodied prey. Food preservation and storage was used, but was not a primary component of the subsistence strategy. At the short term residence camps, a limited number of resource types may be intensively collected and prepared for transport back to the long term residential site. Ultimately, all available resources within effective travel distance were exhausted or reduced sufficiently to be impractical for exploitation and the residential site (pithouse) was abandoned in favor of a new, unimpacted central location.

During Pithouse II, Chatters reports, year-round residential bases were abandoned in favor of paired residential sites (winter and summer base camps), and a variety of specialized resource extraction camps. This structure implies, he argues, a more logistical subsistence strategy, with greater scheduling of activities, more focused attention to a narrower range of prey, and a well-developed storage technology.

In general, the Heath Cliffs site occupations reflect some broad parallels to the Pithouse I and II sequence outlined by Chatters (1989), but also some important differences. During the Warm Springs I Phase the Heath Cliffs site appears to have served as a stable residential base, from which a variety of resources was gathered. The range of activities at the site narrowed during the Warm Springs II, and the site may have served primarily as a hunting camp. This change parallels the increasingly logistical subsistence strategies for Pithouse I and II. However, there is no evidence to suggest that winter occupations ever took place at the Heath Cliffs site, and it is suspected that the settlement pattern may have more closely approximated the pattern of winter/summer village pairs suggested by chatters for Pithouse II. Further, there is virtually no evidence for a broad spectrum approach to hunting during Warm Springs I. There is, however, evidence for use of a broad range of plant foods. berries, small seeds, and probably roots were systematically harvested, processed, and stored during the Warm Springs I occupations. While it is difficult to gauge the relative importance of food storage from the Heath Cliffs site excavations, it is likely that it was more important here, than the marginal value suggested by Chatters for his study area to the north. Edible plant seeds and fish have also been recovered from large storage pits, radiocarbon dated between 4700 and 4200 BP, in the Fort Rock basin to the south of the present area (Jenkins et al. 1994c). It may be that subsistence patterns in the upper Deschutes River basin were more similar to those of contemporaneous occupations in the northern Great Basin than to those of the mid-Columbia basin during this period" (Jenkins & Connolly 1996: 148-149).

[back to top]

Dudley Site 35CR958 Ochoco NF

Barber and Holtzapple (1998) reported on ASCO testing of some pit houses at 35CR958 (Dudley Site) through the Ochoco National Forest during 1989-1990. The site had over 40 depressions recorded as possible housepits. Testing indicated that depressions 2 and 15 were single housepit features while 1 contained two houses and depression 3 was not a housepit, but a possible hearth in feature 3 gave a Middle Archaic date of 4930±70. House 2 had a bench with an interior floor diameter of 4 meters and a surface diameter of 7.9 meters. Depression 1 measured 10.3 meters in diameter and the test profile indicated an earlier housepit had been modified by a later one, producing the enlarged surface depression. A hearth from house 1A gave a date of 1520±50. The majority of the projectile points were narrow-necked Late Archaic arrow heads.

« Previous Region | Index | Next Region »