In the absence of contemporary field studies, some evidence for domestic architecture can come from the careful analysis of contemporary images. Even images produced for commercial, artistic, or other non-scholarly motives can be utilized.
Unfortunately, the early images of tipis often show only a few lodges,
or a single view of a village. Perhaps the earliest ethnographically
accurate European representation of tipis are on the maps of the New
Mexican cartographer Bernardo Miera y Pacheco between 1760 and 1778, on
which he uses tipis as icons for the 'roving tribes'. 
.
The earliest Anglo-American depiction of tipis is apparently Edwin James
1819 portrayal of three Plains Apache lodges.
There are numerous tipis and villages in George Catlin's paintings
and cartoons, and in his near contemporary Carl Bodmer.
Soon after the invention of photography, intrepid western artists were making views of the local Indians. But photographic views of pre-reservation Plains Indian villages are not only rare, due in part to the inherent danger of the operation. But as with the paint media, they often show no more than a single tipi or single views of villages, and thus preclude the detailed examination of contemporary architectural features.
In the winter of 1872-1873, William S. Soule made a series of twelve photographs of a Comanche village which shows a number of lodges, with associated features, and which allow a two-dimensional examination of the spatial arrangement and domestic architecture of a Comanche village in the late pre- reservation period (Kavanagh 1991).
Although Soule, and/or his brother, John S., who marketed the images, deliberately attempted to mislead his viewers by misattributing the views, they are clearly of a single village, probably that of the Comanche chief Horseback, located on Medicine Creek, north of Fort Sill, in the shadow of Mt. Scott. And although the order in which they were made is unclear, and it is possible that they were made over a period of several days, the photographs can be arranged in a series based upon elements which appear in the background of one and in the foreground of the next.
Figure 1
A group of white men and Indians on the banks of a creek with a number of tipis clustered along the banks, with more tipis on the right.
Figure 2
The tipis from the right hand side of fig. 1 are here shown close-up. Tipi 2 is a skin lodge with a strip of cloth or canvas wrapped around the lower portion. To the right of center, more tipis go off into the distance.
Figure 3
Taken from up the hill from the position of fig. 2, this
vantage gives a wider and deeper view of the village.
Fig 3a
An even wider view is Fig 3a (Courtesy, Fort Sill Museum).
Figure 4
Four Indian men are seated on a log in front of Tipi 2. It is smokey with age and the cloth strip is probably intended to patch its lower edge. In the right background is a storage rack carefully wrapped and tied.
Figure 5
Some of the white men and Indians from fig. 1 are lounging around Tipi 1, with Tipi 2 in the background. The Indian fourth from the right is possibly the chief, Horseback; one of the other Indians is his son, later known as Johnny Horseback. The white man on the far left, holding the Remington 'rolling block' rifle, is Horace P. Jones, post interpreter at Fort Sill.
Figure 6
Three Indian men are lounging at the rear of Tipi 1, identifiable by the peculiar curved seam above their heads. The distinctive storage rack beside Tipi 3 is clearly evident in the background, covered with a buffalo robe.
Figure 7
Horace P. Jones poses in front of Tipi 3, wearing a buffalo robe, possibly the robe which covered the rack in fig. 6. On the left is a small unnumbered tipi and an unidentified hump-backed structure which was erected in the interval between figs. 6 and 7. On the right, behind the rack, is Tipi 4 and a tripod with what appears to be a covered shield hanging on it. A tall vertical pole leans against the tripod. Beyond them is a small conical structure, possibly a temporary lodge. In the right background are Tipis 5 and 6.
Figure 8
Interpreter Jones, still wearing the buffalo robe, sits with a family in front of Tipi 5. On the left is Tipi 4, and the shield tripod with the vertical member seen in fig. 7. The conical structure is in the right behind Tipi 6. Next to it is what appears to be another covered shield hanging from a single pole. More tipis appear in the right distance. The white patch at the right has been identified by modern Comanches as an ash pile.
Figure 9
Mount Scott fills the horizon behind Tipi 7, made from two separate covers around a single framework whose poles extend only a little way above the cover. The view of Mt. Scott clearly identifies this as being in the valley of Medicine Creek some 10 miles north of Fort Sill. Although Tipi 7 is not identifiable in any previous image, Tipis 4, 5, and 6 are visible, as is the shield tripod, single shield pole and conical structure.
On the right is Tipi 8 with one of its smoke-flap poles leaning against the front of the tipi.
Figure 10
A dozen lodges are shown from the rear. Tipi 8, in the left foreground, is identifiable from the distinctively shaped tips of its door poles, which also identify this tipi as having a four-pole base. Tipi 7 is identifiable by the looseness of its two covers and its short poles.
Figure 11
A child peers out at the photographer from Tipi 9. A flimsy
storage rack is in the right foreground. Except for the
unmistakable bulk of Mount Scott in the distance, the relation of
this image to the rest of the series is unclear. None of its
cultural features can be matched with the rest of the images.
However, the different perspective on Mount Scott and the trees
along Medicine Creek suggest that this image was made somewhat to
the right (north) of fig. 9.
As a final note, considering that the exact site of this village is known, in the valley of Medicine Creek in the shadow of Mt. Scott, it might be suggested that archaeological investigations be conducted to examine in more detail the material remains of this village. Unfortunately, the very features which made the valley of Medicine Creek attractive to the Comanche village planners of the 1870s, also made it attractive to the engineers of the 1900s: it has been under the waters of Lake Lawtonka since 1903.
Soule's Medicine Creek village series allows a two - dimensional examination of the spatial arrangement and domestic architecture of a Comanche village in the late pre-reservation period.
Details of Comanche style, four-pole Tipis are clearly shown. Comanche tipis use a base of four poles, two of which frame the door, the other two extending north and south of the framework. Thus a distinguishing feature of a Comanche tipi are the two door poles extending beyond and below the main bundle; these are visible on Tipi 3 in figs. 6 and 7. The rest of the poles are laid in the crotches formed by the base framework. Of the eight tipis which can be closely examined, the arrangement of the poles is evenly divided between left- and right-hand order.
The number of poles varies. Many have a framework of 17 poles, the large Tipi 2 has 21, Tipi 7 has only 11.
There is a clear pattern in the way the hides are sewn together to form the apex of the covers: their long edges form the front overlap, with the tails providing ornaments on each side of the door [fig. 8; drawing]. Beyond these, however, few other seams are visible, and other details of sewing are not available.
Thus it is not possible to state how the smoke flaps, or 'ears' are sewn on the cover. On several tipis, the lack of visible seams makes it appear that the ears were an integral part of the apex hides, rather than sewn onto the main cover.
The ears are fitted with a pocket to receive the end of a pole, allowing adjustments of the draft for variations in the wind. Interestingly, the bottom of the ears are not clearly defined, and on none of the tipis is there an ear rope and pole as is common on modern canvas tipis. Many of the poles are drilled near the tips to allow tying into drag bundles, and several have either ties or streamers through these holes.
None of the covers are painted. But many show evidence of hard use. Almost all are smoke stained, particularly around the ears. Two have cloth strips around the bottoms, i.e. Tipi 2 and one unnumbered in the background of fig. 8; Tipi 7 has two separate covers. One of the ear poles in on Tipi 3 has been carefully spliced.
A number of tipis, particularly Tipi 8 in fig. 10, seem to be trenched to channel rainwater away from the interior. This tipi also clearly shows the use of pegs around the base, rather than stones which might leave the so-called tipi rings.
All the tipis face east or southeast, probably due more to the prevailing winds on the Plains than to an ideological meaning.
There are several clusters of lodges, the apparent circle of six tipis to the left in fig. 1, and the cluster of tipis in fig. 10. These may represent nemenakanes, extended family groups, with each nuclear family in one or more tipis. Other than this, however, there does not appear to be an overall pattern to the spacing of the lodges. In part, this may be due to the topography of the southern end of the Medicine Creek Valley which constrains the camping area: on the east is the height of land from which fig. 3 was taken, extending northeastward to the background of fig. 11. On the west is Medicine Creek itself, visible in fig. 1, and indicated by the lines of trees in the background of figs. 2-10. On the other hand, there are numerous references to Comanche villages following a dispersed pattern, for example, the village visited by the U.S. Dragoons in 1834 in which several hundred tipis were spread out for some twenty miles along Cache and Chandler Creeks, ten miles east of the valley of Medicine Creek.
A number of storage racks, windbreaks, and other structures are visible. The most substantial rack stands beside tipi 3. Here the posts are sunk into the ground; with this effort, the builders were clearly not intending to move for some time. Another substantial rack is in the right background of fig. 4, carefully covered and tied.
The racks in figs. 2 - 3, and 11 are more flimsy, and may have been intended as meat-drying racks rather than as storage racks, although the rack in figs. 2 - 3 does have things on it, including what is apparently a coffee pot. There are several windbreaks or other canvas covered brush structures evident in the background of figs. 2, 7, and 11, but only the windbreak constructed in the interval between figs. 2 and 3 is definitely associated with a fire.
At least two shield or other medicine racks are visible in figs. 7-9, one a tripod, the other a single pole.
On the far right of figure 3 is what appears to be an army 'A' tent.
On the other hand, there are several features which are not present. Comanche informants in the 1930s spoke of large nuclear families having two tipis, one for sleeping, the other, the tekakahne, for eating. With the exception of the two large Tipis 1 and 2 in figs. 2 - 3, there is no apparent evidence for such groupings here.
While there are several features which modern Comanches have interpreted as ash piles, there are no wood piles except for the small pile in figs. 2 and 3.
One significant lack are people. This is accountable by two historical facts. On the one hand, over one hundred of their women and children had been captured the previous September and were held captive at Fort Concho, Texas, some two hundred miles to the south. At the same time, much of the remaining population had gone on a late winter buffalo hunt. This might also explain Soule's ability to move freely about the village.
Thomas W. Kavanagh
Curator of Collections
Mathers Museum
Indiana University