University of Texas at Austin wordmarkCollege of Liberal Arts wordmark
Texas Beyond History
TBH Home
Plains Villagers Main

Reconstructing Hank's House

drawing of cutaway view of Hank's house
Artist's depiction of cut-away view of Hank's house as it might have looked 700 years ago. Drawing by Ryan Wilkens.
photo of center roof support post
This wall post (just to the left of the striped photo stick) burned into the ground, but the charring did not continue all the way. The pointed soil discoloration below the charred section shows the shape and extent of the original sharpened post. Photo by Doug Boyd

Click images to enlarge

photo of stunted juniper
Stunted juniper growing near top of caprock overlooking Canadian Breaks near Hank's house. Juniper was the favored tree for house posts, probably because it was one of the few types of wood available and it lasts longer before rotting than cottonwood and other available trees in the area. Photo by Doug Boyd.

Once the preserved half of Hank's house was excavated, we began trying to reconstruct how the whole house might have looked. Although, of course, we will never know for certain what the missing half of the house looked like, many prehistoric Plains Village houses have been excavated in the Texas Panhandle and the Southern Plains. Almost all the houses that look similar to Hank's house are very symmetrical. Thus, we assume that the north half was a mirror image of the south half. Granted this assumption and the preserved facts, the full house was rectangular with its long axis oriented east-west and its entrance facing east. The dimensions (length x width) and interior floor space of the house were 6 meters (19 feet, 8 inches) east-west by 5.6 meters (18 feet, 4 inches) and had a total area of 33.6 square meters (362 square feet).

Accurate measurements of the depths of all postholes (from the house floor level) were made, and accurate measurements of the maximum post diameters were made for all of the charred posts. When no charred wood was found, it was still possible to estimate the approximate width or diameter of the post based on a comparison with the charred posts on the west wall. Because the postholes along the south and east walls were approximately the same diameters and depths as those on the west wall, it is fair to assume that the posts in these holes were in the same size range too. When the sizes and depths of all of the posts and postholes are compared, it reveals some interesting things about how Hank's house was built. (See Post Dimension Table).

It is safe to assume that the posts that were structurally most important were larger and set deeper in the ground because they had to hold up the most weight. As might be expected, then, the two central posts that held up the main central frame in the center of the house were much larger (diameter = 15 centimeters or 6 inches) than all of the other posts and were set deeply in the ground (average depth = 57 centimeters or 22.5 inches). In contrast, the posts along the walls were smaller (average diameter = 8.75 centimeters or 3.6 inches), more variable in size, and were all set less that 43 centimeters (17 inches) into the ground.

Careful excavation of the charred posts and postholes along the west wall, along with detailed observations, led to another conclusion about house construction. All six of the posts that were set into postholes below the floor level were leaning inward toward the center of the house. Measurements of the angles of the charred posts and postholes revealed that all of the posts were leaning slightly eastward ranging from 11.5 to 17.5 degrees off of vertical. When all of the angles were averaged, it shows that the west wall of the house leaned inward about 14 degrees.

photo of posthole
Notice the slight incline of the outline of the posthole of this wall post. The wall posts were purposefully set angled out to help support the roof. Photo by Doug Boyd.
drawing of cross-section of Hank's house
Artist's reconstruction of cross-section through Hank's house depicted with a relatively low roof. Graphic by Sandy Hannum, courtesy Prewitt and Associates, Inc.

Paleobotanist (and archeologist) Phil Dering identified the following plants among the charred archeological remains representing architectural elements:

Charred Material

Architectural Element

Plant Identification

Large posts (n=2)

central roof supports

Juniper (Juniperus sp.)

Small posts (n=9)

wall posts

Juniper (Juniperus sp.)

Wood on house floor

roof components

Juniper (Juniperus sp.)—90 %
Elm (Ulmus sp.)
Cottonwood or Willow (Populus sp.)

Grass on house floor

roof material

Yellow Indian Grass (Sorghastrum nutans)

cover of 1975 report on the Black Dog Village site
The cover of a 1975 report by John Keller on the Black Dog Village site. Based on the excavations of a large (26-x-18 feet) rectangular dwelling that was excavated at this Hutchinson County site (41HC30), an unidentified artist drew a reconstruction of the “big house” that may date to about A.D. 1300. It shows walls consisting of vertical rock slabs and mud (or adobe), with a high roof covered in thatch (grass). Evidence of burned grass and brush was found in this house. Like Hank’s site, this site was also named for a canine, and the unnamed “black dog” is shown in the picture.

Based on all of the architectural details described above, two different reconstruction scenarios may be formulated for Hank's house. Both are the same except for the type of roof. Hank's house may have had a grass-covered roof, in which case it must have been quite tall so that its roof was steep enough for water to run off and not collect in the grass layer. The alternative is that the roof was lower and had a layer of earth covering over the layer of grass. Such houses are called earth lodges and were a common form of house among the Plains Indian peoples in the central and northern Plains.

Several lines of evidence suggest that Hank's house was an earth lodge rather than a grass house. The most important evidence is burned debris inside the house. Modern fire investigators look at the evidence in burned structures to determine when and where the fire started. Remembering that only the southern half of Hank's house survived, the southwest corner where the evidence of intensive burning was found suggests that the fire began in the back of the house or near the center, possibly the roof. The blaze probably caught quickly, but the house fire was abruptly snuffed out when the roof collapsed. If the roof had been made only of branches and straws—all good fuel—it would have continued to burn until the entire roof and the wooden portions of the walls were completely engulfed in flames. Conversely, if the roof had a substantial layer of earth on top of it when it collapsed, this would probably have covered over the burning debris when the roof collapsed and stopped the oxygen flow to inside the pit. This earth lodge scenario would explain why so many of the posts along the south and east wall were not burned.

A similar phenomenon has been suggested for burned pithouses in the southwestern United States. In his 1986 book, Prehistory of the American Southwest, famed archeologist Emil Haury observed that of the 14 "semi-subterranean" houses at an Hohokam site (Roosevelt 9:6) in Arizona:

draing of Antelope Creek house
Antelope Creek houses, as envisioned by Floyd Studer in this undated pencil drawing, closely resembled Pueblo architecture. While the lower wall and exterior details were based on archeological evidence, high adobe and rock walls, the flat roof, and the roof-top entrance were entirely conjectural. Studer went to considerable lengths to reinforce his notion of the "Pueblo Culture of the Texas Panhandle." Courtesy of the Panhandle-Plains Historical Museum in Canyon. (Color added.)

Click images to enlarge


all the lodges except one had been destroyed by fire, … . The burning of an earth lodge, while destructive in itself, does much to preserve the details of construction. The collapse of the roof smothers the fire and preserves the incompletely burned timbers and brush in the form of charcoal. Once in this state, the evidence lasts indefinitely, as long as it is underground.

drawing of Hank's house
Artist's depiction of Hank's house as it might have looked 700 years ago. Drawing by Ryan Wilkens.

Follow Hank's House link to Building It