The evidence furnished by the kivas is significant in other ways, for in connection with the above theory the claim has also been advanced that the cliff villages were occupied for only brief periods at best; that they do not, as assumed by me, represent a phase—so much as an incident—in the development of a people. Aside from the linguistic, sociologic, and other evidence I have to offer later on that of not only these kivas, but also of certain other features of the ruins themselves, is decidedly indicative of both long and continuous occupancy; and an examination of this evidence helps to an understanding of the culture growth of the early cliff dwellers as being not that of Pueblos at first, but that of Pueblo ancestry, Pueblos developing.
Occurring in the midst of the greater groups of northern cliff dwellings, no less than somewhat more scatteringly and widely distributed to at least as far south as the middle of Arizona, are remains of cave dwellings of an older type. They are usually lower down in the cliffs, although they once occurred also in the larger and more accessible of the caverns now occupied by later cliff-house remains, underneath or amid which remains they may still in places be traced. These rude and very ancient cave dwellings mark the beginnings of the cliff occupancy. In all essentials they correspond to the modern cave dwellings of the Sierra Madre in Sonora, Mexico, so admirably described by my friend, Dr. Carl Lumholtz, as built and still lived in by the Tarahumári and Tepehuani Indians, who survive either in the state of these first cliff dwellers of the north, or, as is more probable, have naturally and independently resorted to a similar mode of life through stress of similar circumstances.
Like the Tarahumári, these ancient people of the north at first resorted to the caves during only portions of the year—during the inclement season after each harvest, as well as in times of great danger. At other times, and during the hunting, planting, and seed-gathering seasons particularly, they dwelt, as do the Tarahumári, in rancherias, the distinctive remains of which lie scattered near and far on the plateaus and plains or in the wide valleys. But the caves were their central abodes, and the rancherias, frequently shifted, were simply outlying stations such as are the farming hamlets of the modern pueblos.
The earliest of these dwellings in the caves were at first simple huts disposed separately along the rear walls of these recesses in the cliffs. They usually had foundation walls, approximately circular in plan, of dry-laid stones, upon which rested upper converging courses of cross-laid logs and sticks, hexagonal and pen-like covers surmounted, as were the rancherias of the open plains, by more or less high-pitched roofs of thatch—here in the shelters added rather for protection from cold than from storms of rain and snow.
But in course of time, as the people dwelling, when needful, in these secure retreats increased in numbers, and available caves became filled, the huts, especially in the more suitable shelters, were crowded together in each, until no longer built separately, but in irregularly continuous rows or groups at the rear, each divided from others by simple, generally straight, partitions, as are the dwelling divisions of the Tarahumári today. But unlike the latter, these hut-like rooms of the northern cave-dwellers were still rounded outwardly, that is, each hut (where not contiguous to or set in the midst of others, as was the case with those along the front), retained its circular form. The partitions and foundation walls were still built low, and still surmounted by converging cross-laid upper courses of logs or saplings and roofs of thatch. As with the Tarahumári, so with these earliest cliff dwellers of the north; their granaries were far more perfectly constructed than their own abiding places. To adequately protect their store or provision from seed-devouring animals, no less than from the elements, it became necessary to place it in dry crannies or pockets of the cliffs near at hand, preferably in recesses as far back in their caves as possible, and also to seal it up in these natural receptacles. At first (as may be seen in connection with the caves of Las Tusas, Arizona, containing some of the oldest and rudest separate hut remains I have yet examined) the mouths of these receptacles were walled up with dry-laid stones, carefully chinked, and plastered inside with mud, precisely as were the granary pockets of the Havasupai Indians seen by me in 1881. Later, while still the houses continued to be mere low-walled and partitioned sheds or huts of dry masonry, these granaries came to be quite well constructed, of mud-laid walls, and were enlarged, as stores increased with increase of settlement and tillage, until they had to be built outward from the niches like good sized, slightly tapering bins, protruding somewhat from the cave walls, and finally forming, as do the granaries of the Tarahumári today, miniature prototypes of the perfected single cliff house of a far later day.
In times of great danger small children were not infrequently bestowed for safe-keeping in the larger of these little granary rooms in the deepest recesses at the rear of the earliest cave villages, as the finding of their remains without burial token in such situation has attested; and thus the folk tales which modern Pueblos tell of children left in the granary rooms and surviving the destruction or flight of their elders by subsisting on the scant store remaining therein (later to emerge—so the stories run—as great warrior-magicians and deliver their captive elders), are not wholly without foundation in the actual past of their ancestry. It was thus that these first cliff dwellers learned to build walls of stone with mud mortar, and thus, as their numbers increased (through immunity from destruction which, ever better, these cliff holds afforded), the women, who from the beginning had built and owned the granaries, learned also to build contiguously to them, in the depths of the caverns, other granary-like cells somewhat larger, not as places of abode, at first, but as retreats for themselves and their children.
It is not needful to trace farther the development of the cliff village proper into a home for the women and children, which first led to the tucking of storerooms far back in the midst of the houses; nor is it necessary to seek outside of such simple beginnings the causes which first led to the construction of the kivas, always by the men for themselves, and nearly always out in front of the house cells, which led to the retention for ages of the circular form in these kivas and to the survival in them for a long time (as chambers of council and mystery, where the souls of the ancients of men communed in these houses of old with the souls of their children's grandchildren) of the cross-laid upper courses of logs and even the roofings of thatch. Indeed, it is only in some way like this, as survival through slow evolution of archaic structures for worship, that the persistence of all these strange features—the retention for use of the men, the position in front of the houses, the converging hexagonal log wall caps, the unplastered roofing of thatch—until long after the building of houses for everyday use by the women, with walls continuous from floor to ceiling, with flat and mud-plastered roofs and smooth finishing inside and out, manifest themselves.
Of equal significance with this persistency of survival in the kiva, as to both structural type and function, of the earliest cave-dwelling hut-rooms through successively higher stages in the development of cliff architecture, is the trace of its growth ever outward; for in nearly or quite all of the largest cliff ruins, while as a rule the kivas occur, as stated, along the fronts of the houses—that is, farthest out toward the mouths of the caverns—some are found quite far back in the midst of the houses. But in every instance of this kind which I have examined these kivas farthest back within the cell cluster proper are not only the oldest, but in other ways plainly mark the line of the original boundary or frontage of the entire village. And in some of the largest of these ruins this frontage line has thus been extended; that is, the houses have grown outward around and past the kivas first built in front of them, and then, to accommodate increased assemblies, successively built in front of them and in greater numbers, not once or twice, but in some cases as many as three, four, and in one instance five times.
All this makes it plain, I think, that the cave and cliff dweller mode of life was a phase, not an incident merely, in the development of a people, and that this same people in general occupied these same caves continuously or successively for generations—how long it is needless here to ask, but long enough to work up adaptively, and hence by very slow degrees, each one of the little natural hints they received from the circumstances and necessities of their situation in the caves and cliffs into structural and other contrivances, so ingenious and suitable and so far-fetched, apparently, so long used, too, as to give rise to permanent usages, customs, and sociologic institutions, that it has been well-nigh impossible to trace them to such original simple beginnings as have been pointed out in the case of a few of them.
The art remains of both the earliest cave dwellers and of the cliff dwellers exhibit a like continuity of adaptive development; for even where uses of implements, etc., changed with changing conditions, they still show survivals of their original, diverse uses, thus revealing the antecedent condition to which they were adapted.