Buried City

In this section, we take a look at the characteristic Buried City architecture, particularly the large stone-based houses which have attracted attention for at least a century. Buried City architecture differs significantly from that known at other Plains Villager sites in the Texas Panhandle and adjacent areas. Until the Courson family sponsored a new series of excavations in 1985, archeologists had assumed that the Buried City houses were much like those of the better-studied Antelope Creek focus along the Canadian River. Our work showed that this assumption was only partially true. There are many houses and many different kinds of houses at the Buried City locality. They range from small scooped-out pithouses 2 to 3 meters (6.5-10 feet) in diameter, to the recently discovered deep, circular pithouses, to surface houses with post walls and poorly prepared floors, to large, carefully planned, stone-based houses that are the hallmark of the Buried City complex. And then there is the misnamed "temple."

Construction and Layout of a Typical Buried City House

Based on the four large houses excavated in the 1980s, the typical Buried City house had a square to rectangular outline and measured about 7-9 meters (about 23-30 feet) on each side. The houses appear to have been constructed by the jacal technique, that is the walls were formed of closely spaced upright pickets, woven together with thin sticks/vines and then plastered over. But the walls are unusual in that they were double-pen walls with two parallel rows of pickets, which were anchored by caliche boulders and slabs, earth-filled, and plastered with mud. The roofs were most likely capped with a thick layer of grass thatch. The result was a sturdy, well-insulated house that provided protection from winter cold and summer heat (and possibly from raiders as well).

House construction began with the excavation of a contiguous wall trench about 2-feet thick that outlined a rectangular area about 8 meters (26 feet) on a side. Small-diameter posts (pickets) were then set against both sides of the wall trench so as to form parallel rows about 2 feet apart. Next, the trench was backfilled to the level of the original soil and then capped with small to large caliche boulders and slabs of caliche placed between the rows of pickets. The boulders formed the lower core of the wall and acted to stabilize (anchor) the walls and keep the two sets of pickets from collapsing toward each other. Next the pickets were interwoven with flexible sticks (probably of plum, native grapevine, and willow) that ran along walls, essentially forming a pair of wickerwork fences about 20 inches (50 centimeters) apart. When these wall forms were finished, the prehistoric builders began excavations inside the house.

The bench areas along the north and west sides of the house were dug down a scant 10 centimeters (4 inches) below the original ground surface along the walls, and sloping deeper toward the center, eventually reaching a depth of about 18 to 23 centimeters (7-9.5 inches). Between the benches, a deeper central channel was dug as much as 50 centimeters (almost 20 inches) below the original ground surface. The inner edges of the benches were vertical and 20 to 28 centimeters (8.3-11 inches) high. The edges of the north and south benches were carefully plastered with caliche mud to form smoothly rounded edges, but lacked the raised lips that occur along the edges of the benches in many Antelope Creek Phase houses.

If the earth dug from the interior of the house had been used to fill the double-pen picket walls, as seems likely, the amount of fill would have been enough to create an earth-filled wall about 1.5 meters (4.9 feet) above the original ground surface. Inside the house four large posts about one foot in diameter and possibly six to eight feet tall were set into the channel against the benches to provide roof support. The roof support posts were probably topped by horizontal braces. Roof-support stringers would have run from the braces to the tops of the walls. The result would have been a square, flat-topped roof in the center of the house surrounded by a sloping roof.

Little real evidence survived of what the roof may have been made of, but we have hints of two possibilities. The first is that the roof was a grass-thatch roof with white caliche plaster around a smoke-hole in the center, directly above the central fireplace. Chunks of burned plaster were found on the floor of several houses. Alternatively, the roof may have been covered by a layer of earth, possibly cut sod blocks. Only minimal evidence of burned earth was observed in the house ruins, leading us to suspect the roofs were made of grass.

Life in the Kit Courson House

The Kit Courson house is the most thoroughly investigated Buried City house. A detailed model of the house was built by Danny Witt based on archeological documentation and comparative data from historic earth lodges in the Northern Plains. Although the Buried City houses may have had grass-thatch roofs instead of being earth-covered and differed from Plains earth lodges in other important details, the basic layout and roof-framing may have been similar. Putting it all together—documented details and supposition—we offer a sketch of life in the Kit Courson house (drawing on details from other Buried City houses as well).

The villagers would have entered the house from the east via an extended entryway (slightly sloping down into the house), possibly a crawl-way that was lined with posts and covered like the house roof. The narrow entranceway minimized the amount of snow and cold air that entered the house. Near the entrance, but outside the house was a smaller, perhaps semi-attached room on the southeast corner that had apparently been added after the main house was built. This added room may have served special purposes such as that of a menstrual room or perhaps a mother-in-law's room. The houses of several Plains Indian groups are known to have had such features.

Between the entryway and the attached room was a food processing and work area where in good weather people would have performed routine tasks like food preparation, tool maintenance and repair, and other daily functions. (In this area, archeologists found many deer-antler tool handles and flint-knapping tools, a deer-jaw sickle for cutting grass, grinding stones, and large fragments of pottery.) Weather permitting, most daily activities and work probably took place outside the house. When the north wind was blowing, outside work likely was done on the leeward (protected) side of the structure or in sheltered nooks like that created by the room addition and the entryway wall.

Entering the house, you would have stepped over a small raised sill at the bottom of the entranceway. When inside the house, proper, you would have been standing in the central channel, the lowest floor level. The sill you stepped over may have served two purposes; it probably keep water from seeping into the room from the entranceway and it would have been a stop for an entryway flap door, probably made of bison hide. Such a flap door would have been needed in the winter to trap cold air and keep it from entering the room. Interestingly, just inside several houses in the southeast corner of central channel, we have found several smooth rounded stones, sometimes fragments of grinding stones. Jack Hughes thought these served as weights for the entryway flap door.

In the center of the central channel (which is also the center of the house) was a fireplace or hearth confined within a raised ring of plaster. Smoke from fires would have exited through a hole in the roof that was probably lined with plaster to guard against an accidental fire (a grass-thatched roof would have been highly flammable). Not surprisingly, numerous charcoal flecks are found in the center of the houses surrounding the fireplace, often embedded into the plaster. Other cooking-related detritus found in this area includes many small (½-inch diameter or so) fragments of pottery and small scraps of bone or mussel shell. In the center of the channel, south of the firepit in one house, we found a mid-portion of a buffalo rib lying against the bench margin. The rib was worn in such a way as to suggest it may have been used as a pot stirrer in food preparation.

Surrounding the central channel area are wide, low, elevated platforms or benches occupying the north and south thirds of the room and a narrower rear bench occupying the western one-third. At the intersection of all these thirds are the four massive posts, buried as much as 60 to 75 cm below the lowest surface of the living floor.

In the front section of the house, the south bench and adjacent channel area was apparently an interior work area. At several other Buried City houses, we have found flint-knapping debris, broken grinding stones, and heavy compaction and wear on the floor. Most of the debris is in the central channel, although evidence of flint knapping occurs on the bench itself. In the Kit Courson house we found the sharp end of a flint knife embedded in the floor—it looked like someone had plunged a knife into the floor and then snapped it off, leaving the broken end just below the plaster surface of the floor in this area. In the central channel adjacent was a large quantity of knapping debris from the same kind and color of chert.

The front part of the north bench was consistently cleaner than the south bench but also had some minor knapping and bone debris (small fragments ½ to 2 inches long, possibly from meal leftovers or bone tool manufacture). On the surface of the front-north bench areas we found fragments of charred food items, in one instance a small pit, possibly a storage pit, and occasional bits of pottery and tips of bone awls.

In contrast to the front bench sections, the central bench sections on either side of the hearth near the center of the house are normally clean and artifact-free. In the houses where plaster is preserved, it is heavily worn in the central bench sections, suggesting much traffic. This is probably where people sat while cooking and eating and through which they passed when heading to the rear bench corners.

The rear benches, sometimes had greasy or burned areas about 50 centimeters (20 inches) in diameter. This could be where small warming fires (or perhaps hot stones) were placed or perhaps the rotted remnants of a buffalo robe pile? This pattern was very obvious at the Courson B House. In general, the rear north and south benches (corner sections) of Buried City houses have never produced any artifacts on the plastered floor, and the plaster is normally well maintained with little or no wear. These corner sections are separated from the central rear bench by a shallow U-shaped channel that runs along the north and south margins of the rear bench and slopes from near the floor surface at the west end down to the level of the central channel abutting the main support post at that point. At the Courson B house we found unusually complete deer metapodial (ankle bone) awls, or weaving tools, in the channel that separates the rear bench from the north and south benches. It is not known whether these channels served a mundane function, such as ventilation channels leading to small holes in the outside wall (possible indication of such was found in one house.) Alternatively, they could be dividers that set aside the central rear bench area as sacred space, where ritual paraphernalia was kept. The pattern designating the central west/rear section of houses as sacred space (often used for altars) was widespread among Plains Village groups.

Architectural Variation

So far we've described only a single kind of house at Wolf Creek. These are the large, stone-based houses of about 64 square meters (689 square feet) enclosing a large single room that can be divided into several discrete areas based on apparent structural features (benches, channels, posts, etc.). Such houses are found in isolated locations, well separated from one another, and were probably built adjacent to one or more garden plots. These are the houses that most observers have noted and dug into, mainly because they are large and obvious. They appear to date in the middle to late 14th century A.D. and are limited to a few miles along Wolf Creek, primarily on the south side of the creek.

Other styles of structures have been identified during our studies and by the ongoing work by the University of Oklahoma. Four different house styles have been identified so far: single-post-wall structures, pit-houses or houses in pits, large single-room stone foundation houses, and one extremely large, multi-roomed, stone-based house. Design variation in the houses in a community is accepted as the norm today, yet archeologists seem to assume that, in the past, people made all their buildings in the same way in each area. Much of the variation we see along Wolf Creek may be simply the range of seasonal variation within a single community. Some structures may have been relatively light and porous for use in mild weather while others (the larger stone-based houses) were durable and resistant for use during inclement weather. The use of summer arbors and more substantial houses for winter is well documented by archeology and ethnohistory throughout the Southeastern U.S. It is, however, also possible that some of the variation seen along Wolf Creek may have developed over time; as more houses are excavated and dated, this idea can be evaluated. Regardless of what the source of the architectural variation may be, the differences are quite real.

The Courson A house shows the first variation in architectural style. It is a small house with walls made of a single row of posts, a central firepit, and two central support posts. The floor is irregular and lacks clear-cut evidence of plastering or smoothing. The entryway was probably to the southeast, although there was no obvious entranceway. Work areas associated with the house may have surrounded it, with a prominent work area west of the house, probably even then on or near the bluff overlooking Wolf Creek. This house was probably used in the mid to late 1200s. The finding of small pieces of stick-impressed daub suggests that the house's walls may have been constructed of wattle and daub: vertical posts set in the ground, then woven with thin limbs, vines, and other pliable substances, and then the whole thing plastered with a layer of mud. There were no rocks in the base of the walls.

The Courson B site contained a large typical Buried City house with stone-base walls that were underlain by two pits that had features reminiscent of pithouses. There were no obvious postholes associated with them, but their shallow basin form (about 2.5 meters in diameter) and a hearth in one, suggest they may have been small pithouses. Radiocarbon dates suggest the pithouses were in use during the early 1200s, slightly earlier than the overlying, large, stone-based house at Courson B. Both possible pithouses had been filled with trash and then after the passage of some time, the large stone-based house at Courson B was built over them.

A third house style was also documented at Courson B, a small square structure with no central channel, no formalized entry, and only a single line of postholes associated with a foundation of single rocks. No datable features (firepits, charred posts, etc.) were found in association with this structure, so we only know that it dates after the large stone-based house and pre-dates the terminal event at the site: a mass burial covered by a rock cairn burial.

Eyerly's "Temple"

The most unusual and atypical "house" at the Buried City is the one Eyerly called the "Temple" and Moorehead labeled Gould Ruin. While it doesn't seem to have been a temple in any meaningful sense, Eyerly's Temple is almost three times the size of any other prehistoric structure along Wolf Creek.

The basic dimensions of Eyerly's Temple have been known since the late 19th century: The long-axis of the structure is oriented 30° east of north and extends for about 22 meters (72 feet). It was some 7-8 meters (23 to 26 feet) wide (reports vary and some details were obscured or destroyed by repeated digging). Observations and opinions of the building varied, depending upon the observer and his prejudices. Eyerly suggested the structure was a burial mound. Moorehead defined it as a multiple-room structure with a central burial chamber. Our work in 1988-1990 confirmed neither of these, but produced a third alternative still and left us with far more questions than it did answers.

Our work at Eyerly's Temple was limited largely to exploring the past archeological and less formal digs in the site and only secondarily allowed us an opportunity to investigate the prehistoric site and its contents. By clearing the disturbed matrix from previous excavations, tidying up the results, and doing some limited and partial exploration of the features previous workers didn't identify or explore, we hoped to gain insights into the structure while limiting our own damage. This limited objective was intentional since we will probably never see another Buried City building quite like Eyerly's Temple.

In practice, our intended approach was difficult to accomplish because, even when records of past work existed, they didn't seem to accurately reflect the situation on the ground. T. L. Eyerly mentions only that his crew dug into this structure a little. Correspondence from Warren Moorehead to Fred Sterns suggests that Sterns spent some time excavating at the Temple but, without Sterns's notes and photographs, we do not know where he worked, how much he dug, or what he recovered. Franklin mentioned doing "some work" there, but didn't provide any maps or photographs showing exactly what he did.

With Moorehead's work in 1920 we are on somewhat firmer ground, as he published sketches of his work in 1920 and there are some photographs. Moorehead noted that he paid Sam Handley to backfill the excavations at the end of the season. During the 1990 season, we began to encounter bits of barbed wire intertwined with the rocks of the Temple and other puzzling occurances. Fortunately, a long-time local resident informed us that Handley did not backfill Moorehead's trenches, but had instead left them open for years so that the locals could see what they looked like. Handley used strands of barbed wire to keep the rocks from collapsing into the open trenches and, according to the informant, from time-to-time restacked the rocks after they had collapsed.

Finally, there are some errors in the published record. Moorehead notes four openings or doorways in the west wall and five in the east wall and refers the reader to his Figures 32 and 34 for evidence. Close examination of these photographs shows them to be the same wall viewed from slightly different perspectives. Fortunately, we were able to secure a copy of Moorehead's original field notes from the Peabody Museum in Andover, Massachusetts and these plus other sources provided a somewhat better guide his work.

The south wall of the Temple may have been destroyed before Moorehead arrived—he does not show firm rock alignments marking its position. Second, he notes at one point that near the south end "of this [central to the ruin] trench we found a fragmentary bunched burial, down 20 inches—no skull and no large bones." His field notes discuss at length his difficulty in finding qualified labor and eventual decision to use a Fresno scraper and mule team (predecessor to the modern road grader) to excavate the trenches. This might explain the partial bundle burial containing only a upper torso and skull that we found in a similar location. Moorehead's expedient use of mechanical (or mule-powered!) excavation may have limited what he could observe in his excavations.

During the 1988 and 1990 work, we found sufficient features to suggest that Moorehead's (and everyone else's) work did not finally reach the primary occupation floor of Eyerly's Temple except where the walls were trenched through the house floor. We were able to identify postmolds, storage and cache pits, firepits, and some other features from the prehistoric construction as well as numerous features from known and unknown historic excavators.

The walls of the house appear to have been lined with posts set about 25 to 40 cm (10-15 inches) inside the stone facing of the wall. In the geometric center of the structure was a very large firepit almost 75 centimeters (30 inches) in diameter and about the same in depth. Flanking this firepit were six large postmolds that probably represent the primary roof support posts of the structure. In the center of the southern half of the house was a cylindrical firepit of about 30 to 50 centimeters (12-20 inches) in diameter that was bracketed on the east and west by two postmolds 10 to 15 centimeters (4-6 inches) in diameter. Scattered throughout the building were the cylindrical trash or cache pits about a meter in diameter, which seem so characteristic of the Buried City.

Moorehead reported several openings in the east and west walls of the structure, but because of the extreme disturbance of the stone walls since Moorehead's work, we could neither confirm nor deny his observation. What openings we did find seemed to extend irregularly beyond the structure as though they represent "pothole" or informal excavation pits more than architectural elements. Along the east wall are three areas that might represent possible architectural openings. These seem to be aligned with the two confirmed firepits and possibly a third that Moorehead found.

Eyerly's Temple represents the most unusual structure in the Buried City. Consider, however, that it seems to contain 3 fireplaces, 3 sets of center posts, and is roughly 3 times the size of the "typical" stone-based house. Perhaps the Temple was a large tripartite house, holding 2 or 3 families under a single roof with informal or light partition walls dividing the space into three units: north, central, and south. Since the central unit has a substantially larger hearth than the north or south, then perhaps it served as a community room for two related, extended families or lineages occupying the north and south areas. If complete excavation of the structure is ever undertaken, this hypothesis would certainly be worthy of exploration.

Eyerly's Temple, like most Buried City structures, probably had a complicated history of use that we will never be able to fully sort out because of how it has been treated in historic times. The overlapping hearths show there were at least two occupational periods. But Eyerly surely thought it was a burial mound because he found burials in the upper, mounded part as it appeared in 1906. The mounded earth probably represents the decayed earth-filled upper walls and possibly the earth-covered roof of the "big house." Thus, the burials Eyerly dug into were probably interred long after the big house had fallen into ruin, perhaps even centuries later by unrelated Native American peoples.