The architecture of Pueblo Bonito : (with 81 plates) |
 | I. |
THE RISE AND FALL OF PUEBLO BONITO |
 | II. |
| III. |
| IV. |
| V. |
| VI. |
| VII. |
| VIII. |
| IX. |
| A. |
| B. |
| C. |
 | The architecture of Pueblo Bonito : |  |
THE RISE AND FALL OF PUEBLO BONITO
Pueblo Bonito began as a wide-spreading crescent of Pueblo II
houses with storerooms at the rear, several subterranean kivas out in
front, and the village trash pile beyond. After that trash had accumulated
to a depth of 8 feet or more, after 5 feet of sand had
settled against the old P. II houses, another people came to join the
original settlers—a Pueblo III people with a more advanced architecture
and a different pottery complex. Together, the houses these
P. III people built and the pottery they made are now widely accepted
as earmarks of a distinct social development, "The Chaco Culture."
One objective of the Pueblo Bonito Expeditions was to ascertain, if
possible, the origin of this development and its termination.
We reached part of our goal but not all. After seven summers at
Pueblo Bonito, 1921-1927, I am of the opinion that the P. II and the
P. III peoples who formerly dwelt there had come independently from
beyond the San Juan River, but I can only guess at their destination
after leaving Chaco Canyon. This recognition of the broad area north
of the San Juan as the place of origin—an area in which I have done
no field work for many years—has been substantiated by the published
observations of Kidder, Morris, Martin, Roberts, Brew, and others,
and I have placed great confidence in those observations in what
follows. I am aware of no later, equally pertinent researches.
The so-called Chaco Culture was just short of full bloom when
it first came to Chaco Canyon. At Pueblo Bonito it introduced veneerand-core
masonry and architectural precision; it also introduced
Corrugated-coil Culinary ware, Straight-line Hachure and, among
others, a hybrid variety of domestic pottery we designated "the ChacoSan
Juan." Seen by various observers and variously described, the
Chaco Culture and influence attributed to it have been reported far
beyond the borders of Chaco Canyon.
Chaco-like pottery and Chaco-like masonry have been cited repeatedly
as proof that Lowry Ruin, northwest of Mesa Verde
National Park and Aztec Ruin to the southeast, among others, represent
colonies from Chaco Canyon. Mancos Black-on-white and
McElmo Black-on-white were dominant pottery types at Lowry
although Mesa Verde Black-on-white appeared conspicuously in the
upper fill of several rooms (Martin, 1936, p. 94, 205). All Lowry
masonry is of veneer-and-core, or double-coursed, composition but
Martin separated it into 3 classes: Chaco-like, non-Chaco, and intermediate.
Tabular sandstone identifies the first; squarish blocks of
uniform coursing, the third.
Aztec Ruin, with its predetermined ground plan, its large highceilinged
rooms, and kivas sunk within the house mass is even
closer to Chaco architecture than Lowry. Like the latter, however,
Aztec masonry falls short of the Chaco ideal because building stone
comparable to that of Chaco Canyon is not found in the vicinity.
Aztec sandstones, and those generally throughout the northern
country, are tough and cross-bedded; they lack the natural cleavage of
Chaco sandstones.
Eventually both Lowry and Aztec were abandoned by their builders
and thereafter were appropriated, partially repaired, and reoccupied
by Mesa Verde peoples. This was also true of Solomon's Ruin, on the
south bank of the San Juan River, and of various lesser structures to
the northward. So-called Chaco peoples settled this northern country
first and those with a Mesa Verde-like culture moved in later.
Where remains of the two occur at the same site "the Mesa Verde is
always on top" (Morris, 1939, p. 204)—not Classic Mesa Verde
necessarily but its forerunner.
The range of Chaco-like influences and the occurrence of Chacolike
pottery and Mesa Verde-like pottery in the same river valleys
and even in the same ruins were puzzles Earl Morris took for
his own. No one did more than he to define and interpret the factors
in these puzzles. He recognized the West Pueblo at Aztec Ruin as a
Chaco-type building but regarded the pottery of its builders as no
more than "Chaco-esque." He recognized the contemporaneity of
"true Chaco," "Chaco-like," and "non-Chaco" earthenware throughout
La Plata Valley (Morris, 1939, p. 205). Chaco-like pottery and
masonry were especially noticeable at his Sites 36, 37, 39, and 41 and
each of these had been reoccupied by peoples with a Mesa Verde-like
culture.
On his plan of Aztec Ruin, Morris (1924) includes the local Great
Kiva and 28 lesser kivas. Of these latter, 12 are represented with
shallow basal recesses, 8 to 10 pilasters, and subfloor ventilators. Two
(C, N) are shown with the deep south banquette that identifies them
with the period of Mesa Verde, or a proto-Mesa Verde, occupancy. I
use "proto-Mesa Verde" as a synonym because, unlike others, I see
only one Mesa Verde-like culture for the Mesa Verde country—a
culture born in small-house settlements westward toward the Rio
Colorado or beyond, a culture that developed vigorously in McElmo
Canyon and its numerous tributaries along the Colorado-Utah border
communities as Spruce-tree House and Cliff Palace, with their distinctive
pottery and numerous ceremonial chambers or kivas.
So-called Mesa Verde-type kivas may differ from place to place
but they retain as fairly constant fixtures an encircling bench about
3 feet high, six masonry pilasters rising 2 to 3 feet higher, a deep
above-bench recess or "banquette" at the south with a floor-level
ventilator underneath and, between fireplace and north bench, a
cylindrical hole in the floor, the sipapu, symbolic passageway from
the underworld (Kidder, 1924, p. 60).
Chaco Canyon kivas, on the other hand, have low, log-enclosed
supports for their cribbed ceilings rather than 3-foot-high masonry
pilasters. They have a shallow in-bench recess at the south, a subfloor
ventilator connecting with an external shaft, and a sunken
"vault" of unknown function west of the fireplace. They lack the
deep, above-bench south banquette of Mesa Verde kivas and they lack
the sipapu. In all the Society's Chaco Canyon investigations no
kiva-floor hole was found that could positively be identified as a
sipapu except, possibly, that in Kiva Q. But, as we shall see, both
Mesa Verde-type and Chaco-type kivas occur at Pueblo Bonito.
Throughout the Mesa Verde country and southward many archeological
sites, early and late, display a mixture of elements considered
distinctive either of the Mesa Verde culture or the Chaco. Morris's
Site 39, at the junction of La Plata River and Barker Arroyo, includes
a number of buildings evidencing occupancy from BM. III to late
P. III, or Mesa Verde, times (Morris, 1939, pp. 50-55). Building I,
a late structure, is noticeably Chaco-like in its planned arrangement;
refuse piled on the north side contained many sherds that "in quality
of paste, surface treatment, and ornamentation, might have come
from the dump of Pueblo Bonito." But, the smaller of its two intramural
kivas, No. 6, had eight masonry pilasters 3 feet high, a subfloor
ventilator, a shallow basal recess at the south, and a 39-inch-deep
banquette above—a combination of Chaco-like and Mesa Verde-like
fixtures.
Beneath Building I were remains of a Pueblo II cobblestone
structure that included four small kivas of which Morris cleared two.
One of these was typical of the Mesa Verde but its companion, Number
1, had a 9-inch-wide bench without pilasters, a south recess 6
inches deep, a sipapu, and an under-floor ventilating system—the only
instance, if I read correctly, in which a subfloor ventilator is reported
in a Pueblo II kiva. Here, then, as in the overlying Pueblo III building,
and Chaco-type kivas appear together in the same chamber.
Presumably all circular kivas, irrespective of period and locality,
evolved from BM. III—P. I pit-houses wherein family living quarters
were combined with an area set apart for rituals. Brew (1946)
describes 14 such combinations at Site 13 on Alkali Ridge, southeastern
Utah. No two are exactly alike. With one exception (N)
all had four roof-supporting posts, a restricted area at the south,
and a low passageway to an antechamber. Some had a sipapu or
possible sipapu; 3 (B, E, M) were lined at floor edge with 2-inchdiameter
posts, upright or leaning inward; one (H) had a threequarter
bench and upright posts at the rear of it.
Small posts slanting roofward from the rim of a pit-house,
or from its "bench" when the pit was deeper, are characteristic
of the Pueblo I period. They have been noted, north and south,
wherever pit-dwellings are known and will be cited again in our
description of P. III kivas at Pueblo Bonito. By their own charred
timbers, a majority of dated pit-houses apparently were constructed
in the 8th century. As previously noted, at least two of them occur in
Chaco Canyon.
One-story surface structures walled by posts packed between
with mud—a specialty of Pueblo I architects—often accompanied
BM. III and P. I pit-dwellings. Roberts (1930) describes three different
kinds of post-and-mud structures in the Piedra district, east of
the Mesa Verde, each kind grouped crescentically about the north and
west edges of the pit that supplied mud for house walls. Pueblo I
post-stone-and-mud construction persisted on Alkali Ridge even
after local pottery had developed into types generally recognized as
Pueblo II (Brew, 1946, p. 222). Rocks crowded into the mud between
posts led to coursed stonework and single-coursed masonry has
long been regarded, sometimes incorrectly, as a badge of Pueblo II
civilization.
Lancaster and Pinkley (1954) describe a remarkable sequence
of three superposed P. II kivas at Site 16, Mesa Verde National Park,
each provided with sipapu and lateral ventilator. The first was a
simple 4-post jacal structure while the second was walled with "singlecoursed"
masonry and the third, with "double-coursed" stonework
including blocks dressed by pecking. Alone among the three, this
uppermost Site 16 kiva possessed masonry pilasters (eight in number,
2 feet high) and a deep south recess—features that thenceforth
were to distinguish Mesa Verde kivas.
On the basis of our explorations at Pueblo Bonito the two bark
dates, A.D. 1074, cited by Lancaster and Pinkley (ibid., p. 78)
seem to me a bit late for a Pueblo II building although quite in keeping
with "double coursed" masonry. Both single- and double-coursed
stonework appear at Pueblo Bonito, but the latter was not introduced
until after 8 feet or more of household rubbish had accumulated in
front of the original settlement. That original settlement was a
crescent-shaped Pueblo II village of single-coursed masonry; 12 of
the 13 datable timbers we recovered from its ruins were felled between
A.D. 828 and 935. In contrast, beams, ceiling poles, and lintels
salvaged from later portions of the same pueblo, those with doublecoursed
masonry, bear tree-ring dates from A.D. 1011 to 1126.
These latest rooms, despite their superior construction, were
first to be abandoned as I read the record, and they were stripped
of their furnishings in the process. The Old Bonitians, on the other
hand, remained in residence and amassed the cultural treasures
and the foodstuffs cataloged by the Hyde Expeditions. If eight Old
Bonitian rooms eventually were converted into burial vaults for a
hundred dead that could not be interred in the accustomed place the
fact merely evidences continuing occupancy under adverse conditions
(Judd, 1954, pp. 325-341). We found no adult burial in Late
Bonitian dwellings.
Pueblo II masonry at its very best is found in Old Bonito—its
equal has not been reported elsewhere—wall-wide slabs of sandstone
spalled around the edges and bedded one upon another in a surplus
of mud mortar (pl. 10, 1). Interior walls may include upright
slabs at the base or posts with mud and rocks between in the old
P. I. tradition; exteriors may slant inward after the manner of pithouse
walls and may be studdied with stone chips, presumably to
lessen erosion by wind and water.
In his tabulation of architectural features observed on Alkali
Ridge, Brew (1946, pp. 204-205) notes that 8 of the 14 BM. III—P. I
pit-dwellings at Site 13 were equipped with 4 roof-supporting posts
while five had six posts each and one (N) had none. Nine Pueblo II
kivas on Alkali Ridge, like nearby pit-houses, had been dug down into
native earth and plastered. Each of the nine was provided with a
lateral, above-floor ventilator; the sipapu was present in five and
absent in four; the deep south banquette appeared in two only, those
at Site 11 and in Unit 2, Site 13.
Five of Brew's nine P. II kivas had six masonry pilasters each, two
had eight, one had four, and the ninth retained the four free-standing
plausible thought that stones piled behind a weakened roof support
led to masonry pilasters and that these latter eventually supplanted
posts altogether. Kivas at Brew's Sites 3 and 9, one with four masonry
pilasters and the other with four roof-supporting posts, were both
bowl-shaped and seem likely forerunners of the bowl-shaped P. II
kivas at Old Bonito.
From his investigations at Whitewater, Roberts (1939) points to
two bench-wide adobe ridges in Structure 12 as possible antecedents
of the low Chaco-type pilaster and to the subfloor ventilator in
Kiva B, Unit 2—the only one observed at Whitewater—as one of the
earliest of its kind. Both Kiva B and Structure 12 are described
as Developmental Pueblo, or P. II; both were provided with the symbolic
sipapu. But the subfloor ventilator was installed when Kiva B
was reconditioned and a new floor laid. The possibility remains, therefore,
although Roberts does not hazard the guess, that this renovation
of a P. II kiva occurred coincident with construction of two nearby
P. III house units and a Great Kiva, each of which possesses undeniable
Chaco affinities and may be more or less contemporaneous with
the Village of the Great Kivas on the Zuñi Reservation, which is
definitely P. III in time and possibly even post-Bonito (Roberts,
1932, p. 169).
The subfloor ventilator in the P. II kiva beneath Morris's Building
I, Site 39, appears to be fully developed (Morris, 1939, p. 53).
Since there is no hint here of reconstruction, this second example may
be older than that in Roberts's rebuilt Kiva B. In either case the two
apparently identify the subfloor ventilator—one of the most pronounced
differences between Mesa Verde and Chaco kivas—as a P. II
innovation. I know of none earlier. Two earthenware vessels crushed
upon the floor of his buried kiva are also identified by Morris as
Pueblo II.
There is still disagreement as to the actuality of P. II pottery
just as there is dispute as to what constitutes Mancos Black-on-white
and McElmo Black-on-white. Martin (1936, pp. 80-94) first described
Mancos Black-on-white from Lowry Ruin where it was
present "from earliest times, rising and then gradually decreasing in
percentage" and, as it decreased, McElmo Black-on-white, "formerly
known as proto-Mesa Verde," rose in favor (ibid., p. 113). In part
because these two varieties were associated at Lowry, Martin classified
the ruin as "late P. II—early P. III" in time and culture but with
strong Chaco affiliations. Indeed, his Mancos Black-on-white resembled
a sherd in hand was one or the other (ibid., p. 112). The nine known
tree-ring dates from Lowry, A.D. 987-1086 (Smiley, 1951, p. 23), lie
within the Pueblo Bonito bracket.
Presence of Mancos and McElmo pottery in association at northern
ruins has long puzzled archeologists of the Pueblo area. Both are
primarily sherd-tempered, but the one is ornamented with mineral
paint and the other with organic. Mineral paint persisted from P. I
through the Chaco-like phase of P. III, but meanwhile the use of
organic paint increased progressively and became dominant by the
end of the period, when Classic Mesa Verde was in its prime
(Shepard, 1939, p. 254).
From Mesa Verde National Park Deric O'Bryan (1950) contributes
to the definitions of both Mancos Black-on-white and McElmo
Black-on-white. He identifies the first with small, one-kiva house
units dated approximately A.D. 900-1050; the McElmo Phase, about
1050-1150, is identified with larger masonry settlements whose kivas
in addition to the six pilaster-sipapu-fireplace-deflector and above-floor
ventilator combination of Mancos Phase kivas, have the deep south
banquette as an established feature. Here, then, in O'Bryan's
post-1050 McElmo Phase is the fully developed Mesa Verde kiva of
Kidder's definition, the one with the deep south banquette. O'Bryan
found no pure McElmo site but noted that McElmo Black-on-white
pottery sometimes occurs on Mancos ruins and even on those of
post-McElmo times.
At their Site 16, also on Mesa Verde, Lancaster and Pinkley
(1954, p. 70) noted that "90 percent of the pottery . . . is assignable
to the P. II period, or the Mancos Mesa phase." Reed (1958)
recognized both Mancos and McElmo among Chaco-like sherds at 4
late P. III, or Mesa Verde, sites he excavated in Mancos Canyon but
regarded the McElmo as merely an improved Mancos. "Generally,"
he wrote (ibid., p. 83) "Mancos Black-on-white has been called
`Chaco' pottery or thought of as closely related to Chaco pottery" and,
again, as though clarifying Morris, "The so-called non-Chaco pottery
of the Chaco period on the La Plata is clearly Mancos Black-on-white
decorated with solid elements, lines and dots, and parallel stripes; the
so-called Chaco-like is hachure-style Mancos" (ibid., p. 97).
When Brew (1946, p. 285) found Mancos Black-on-white and
McElmo Black-on-white intermixed in household waste on Alkali
Ridge he listed the former as P. II and the latter as P. III but added
"the Mancos . . . had begun to show Mesa Verde features. The
McElmo." Under the circumstances Brew took a second look at the
McElmo and decided on the spot "to call it early Mesa Verde."
Early Mesa Verde, or McElmo, Black-on-white is a conspicuous
variety at Chaco Canyon ruins, large and small. It was abundant in
late deposits at Pueblo Bonito and upcanyon; it was preponderant at
Pueblo del Arroyo (Judd, 1959, p. 175; Vivian, 1959, p. 26). Its
off-color white slip, its rounded or flattened and tick-marked rim, its
black organic paint, and its near-Mesa Verde designs separate it from
other local types. It was a late arrival at Pueblo Bonito since, in 12
feet of West Court rubbish, Roberts and Amsden recovered no sherd
of it below the upper 4 feet. Hence McElmo Black-on-white serves
as an index to the comparative age of household sweepings wherever
found in the valley.
At Łeyit Kin and Bc 50-51, small-house sites opposite Pueblo
Bonito, the presence of Mancos Black-on-white and McElmo Blackon-white
proved puzzling to Brand, Kluckhohn, Dutton, and their
colleagues from the University of New Mexico because, as I read
their evidence (Brand, et al., 1937; Dutton, 1938; Kluckhohn and
Reiter, 1939), all were too intent upon a greater antiquity. Casual
sherd samples I collected in 1920 at half a dozen small sites on the
south side of the canyon between The Gap and Wirito's Rincon
(U.S.N.M. Nos. 315841-867), and perhaps including Bc 50-51 and
Łeyit Kin, contained such a large proportion with Mesa Verde-like
designs I classified them at the time as P. III and thus contemporaneous
more or less with the major Chaco ruins (Judd, 1921,
p. 102).
Based on this 1920 judgment, our Pueblo Bonito stratigraphy,
and excavation data since published, Łeyit Kin and Bc 50-51 appear
to me no more than P. III offshoots from Pueblo Bonito or Chettro
Kettle. Use of cottonwood and pinyon vigas (Kluckhohn and Reiter,
1939, p. 33) was a P. II trait at Old Bonito; "keyhole" kivas with
high masonry pilasters and above-floor ventilators echo the Mesa
Verde country. Only one kiva, No. 4 at Bc 51, had a subfloor
ventilating system; all pottery types reported, irrespective of name,
are varieties represented in the 12-foot-deep rubbish in the West
Court at Pueblo Bonito. The preponderance of McElmo Black-onwhite
at Bc 50-51 together with rude masonry when tabular sandstone
was readily accessible, adult burials in rooms, and use of potsherds
as wall chinking combine to suggest a late P. III, Mesa Verde-like
occupancy. Seven timbers from Łeyit Kin were all felled in A.D.
1039 (Dutton, 1938, p. 23). That an underlying pit-house was encountered
typical P. I shelter lies at the base of a near-by slope and P. I or
older peoples were long resident in Chaco Canyon (Judd, 1924b;
Roberts, 1929; Bryan, 1954).
The association of P. II and P. III cultural traits, including
kiva fixtures, is apparent at other sites, large and small, throughout
the Chaco area. Still others, as Kidder (1924, p. 57) anticipated,
may represent an earlier or a later horizon. There are the ruins
Amsden examined south of Pueblo Bonito, and there is the one
Roberts partially excavated in 1926 about 10 miles to the east (Judd,
1927a, p. 166). This latter contained so many adult burials and so
many pieces of late Mesa Verde pottery (U.S.N.M. Nos. 334123-154)
is was dubbed at the time "the Mesa Verde house."
Under special permits from the Department of the Interior, the
Pueblo Bonito Expedition in 1925 extended its inquiries beyond Chaco
Canyon. Monroe Amsden that year examined 16 small-house ruins in
Kinbiniyol Valley, south of the Chaco (U.S.N.M. Nos. 329803-845)
and the following summer, 1926, Frank H. H. Roberts, Jr., cut two
stratigraphic sections, one 10 feet deep and the other 12, through
village waste at Pueblo Alto, on the cliff north of Pueblo Bonito,
and three similar tests, varying in depth from 3 feet 3 inches to 8 feet
8 inches, at Peñasco Blanco (Judd, 1927a, p. 168).
Amsden's data remain unpublished, but together his 16 small
ruins impress me as being hastily built, briefly occupied refuges
of post-P. II family groups, harried and on the run. Masonry,
for example, is primarily of wall-wide sandstone blocks, relatively
thick, not face-dressed but amply chinked, and with upright slabs
at the base. Ruin 13, a compact unit in one corner of a slab-enclosed
court, includes a kiva without pilasters but with a shallow basal
recess at the south and a deeper banquette above, a square ventilator
opening 2 inches above floor, a wattled deflector banked with adobe,
a masonry-lined fireplace, and a probable sipapu. Ruin 13 potsherds
(U.S.N.M. No. 329823) include Straight-line Hachure but those
with solid lines, stepped triangles, ticked lines, and checkerboard
figures are more numerous.
The stratigraphic data collected by Roberts at Peñasco Blanco
and Pueblo Alto were, with my permission, included in his 1927
doctoral dissertation at Harvard and have since been cited repeatedly
by other investigators. For our present review of Chaco Canyon
history, however, it is important to note that the Pueblo Alto sherds
(U.S.N.M. Nos. 334161-162) evidence construction and abandonment
while Pueblo Bonito was in its prime since Old Bonitian types and
missing. At Peñasco Blanco, on the other hand, Roberts's stratigraphy
reveals Transitional and Degenerate-Transitional from bottom to top,
Chaco-San Juan and other late varieties in the upper strata only
(U.S.N.M. Nos. 334166-168). This sherd record, coupled with a
section of first-type, or Old Bonitian, masonry visible on the northwest
side (the only ruin other than Pueblo Bonito where such
masonry is known to occur in Chaco Canyon) suggested to Roberts
that Peñasco Blanco was founded about the same time as Pueblo
Bonito but may have been abandoned a bit earlier.
Potsherds we gathered in 1920 and 1925 from the surface at
Chettro Kettle, Hungo Pavie, Una Vida, Weje-gi, and Pueblo Pintado
likewise include relatively few Transitional and associated Old
Bonitian fragments in proportion to later varieties such as the three
hachured types, our organic-paint Chaco-San Juan or McElmo
Black-on-white, and Corrugated-coil Culinary. This preponderance
of late over early varieties of pottery suggests to me an outward movement
from Pueblo Bonito—an outward movement which other students
will dispute and which, admittedly, is not fully substantiated by
our too-short list of constructional dates.
The number and range of tree-ring dates from major Chaco Canyon
ruins, originally reported for the Pueblo Bonito Expeditions by
Douglass (1929, 1935), have recently been reviewed by Smiley
(1951) and Bannister (1959) and are listed herewith, from Peñasco
Blanco eastward to Pueblo Pintado, the out-of-canyon sites being
grouped at the end:
| Ruin | Number | Range of cutting dates |
| Peñasco Blanco | (18) | 898-1087 |
| Ruin 9 (Casa Chiquita) | (1) | -1060 |
| Ruin 8 (Kin Kletso) | (17) | 1059-1178 |
| Pueblo del Arroyo | (35) | 1025-1117 |
| Pueblo Bonito | (71) | 828-1126 |
| Chettro Kettle | (351) | 911-1116 |
| Hungo Pavie | (14) | 942-1077 |
| Una Vida | (8) | 847-1048 |
| Weje-gi | (1) | -1027 |
| Pueblo Pintado | (2) | -1060 |
| Pueblo Alto (N. cliff) | (0) | |
| Sin-kle-sin (S. cliff) | (2) | -1111 |
| Kinklizhin (Black House) | (1) | -1084 |
| Kinbiniyol | (9) | 941-1124 |
| Kinya-a (Pueblo Viejo) | (6) | 1097-1106 |
Tree-ring dates merely suggest the period during which construction
may have been under way. They cannot be taken at face value,
especially when few in number. Timbers were used and reused, as
Douglass (1935, 1939) observed after examining the material collected
by the National Geographic Society Beam Expeditions of 1923,
1928, and 1929. Chaco Canyon forests indubitably were reduced,
possibly destroyed, by builders of the major pueblos and while these
latter were under construction it is reasonable to believe that, suitable
trees being fewer and farther afield, easily accessible timbers were
being salvaged from one abandoned village and carried to the next,
as happened when Awatobi beams were transported to Hano and
Walpi.
Decimation of the Chaco forests would have hastened formation
of a contemporary arroyo, and this in turn would have brought about
reduction in bordering farmlands. In arroyo formation, as Bryan
(1954, p. 12) pointed out, erosion progresses headward or upstream
and because there is progressive reduction in the number of rooms
and in the quantity of visible rubbish especially at Hungo Pavie, Una
Vida, Weje-gi, and Pueblo Pintado, it is my theory these east-lying
ruins, reflect an up-canyon shift of a reduced population. As their
fields failed, the village dwellers moved. And they moved just far
enough, a mile or two at a time, to keep beyond the annually advancing
arroyo. Food has always been a strong incentive to migration!
With fewer data available, Kidder (1924, p. 55) doubted that
more than four or five major Chaco pueblos were inhabited simultaneously
or that the population of the valley ever exceeded 6,000. On
the basis of our later observations, I would reduce those estimates
by half, to two great houses or three at most. I find no reason to
believe the Old Bonitians were involved in this theoretical upcanyon
population shift. They stayed behind, at least for a time, and
stubbornly tilled their ancestral acreage, however curtailed. Malnutrition
is evidenced in Old Bonitian skeletal remains recovered by the
National Geographic Society (Judd, 1954). It was the Late Bonitians
who moved and rebuilt and moved again.
Stonework associated with datable timbers is a further index
to the age of a Chaco ruin. Florence Hawley (1938, p. 250) saw 10
distinct variations in Chaco Canyon masonry. At Pueblo Bonito I
recognized four principal varieties: the oldest, P. II or Old Bonitian;
the other three, Late Bonitian. Twelve tree-ring dates recovered from
the older part of town extend from A.D. 828 to 935; 44 dates from
Late Bonitian houses range from 1011 to 1126.
To judge solely from this Pueblo Bonito sequence, upcanyon
masonry is all late. If my presumed eastward movement were spurred
in any degree by enemy peoples, Weje-gi displays the only evidence—
a row of cliffside portholes. Pueblo Pintado, easternmost of the major
Chaco ruins and a prominent landmark from every direction, stands
astride the Continental Divide.
Once the Divide had been attained which path was taken by bearers
of the Chaco Culture? I do not know. Dispersal was by clan or
family groups, the Late Bonitians first and the Old Bonitians sometime
after. Nowhere is there evidence of mass migration. The two
peoples did not necessarily follow the same trail but both left Chaco
Canyon. Small pure Chaco sites are reported along the Continental
Divide, southward from Pueblo Pintado and West of Mount Taylor.
Reed (1950, p. 92) postulates a population shift eastward to the
upper Rio Grande but elsewhere (1955, p. 179) recognizes among
potsherds collected in the Zuñi country "true Chaco Black-on-white of
the twelfth century."
A twelfth century migration southward from the Chaco country
seems entirely reasonable. The latest known growth-ring from
Pueblo Bonito is A.D. 1126; only one later Chaco date has been
reported, A.D. 1178 from Ruin 8 (Kin Kletso) a half-mile west of
Pueblo Bonito (Bannister, 1960, p. 20). All Late Bonitian rooms we
explored had been vacated and stripped of their furnishings while
the Old Bonitians continued in residence, storing their autumn
harvests and, contrary to their cultural heritage, burying at least
some of their dead in unused groundfloor rooms.
The small clustered rooms of Hopi towns have always seemed
to me a reflection of those in Old Bonito just as the large, highceilinged
rooms of Acoma and Zuñi have seemed to echo those of Late
Bonito. This is only a personal impression, to be sure, but Chaco
Canyon influences are stronger in south central New Mexico than in
any other area personally known to me. And there remains the
intriguing fact that the Zuñi, culturally Puebloan, are an isolated
linguistic group.
The paired kivas Hodge (1923) excavated back of Hawikuh are
pre-Zuñi and follow the Chaco tradition with their subfloor ventilators
and sunken vaults west of the fireplace. Ruins underlying Ketchipauan,
one of the Seven Cities of Cibola, are of excellent masonry
and the equal of that at the paired kivas near Hawikuh.
In a letter of August 18, 1921, addressed to the first symposium
held at the National Geographic Society's Pueblo Bonito camp, N. C.
and orientation reminded him of ruins in Chaco Canyon
although he saw no comparable affiliations in local pottery.
At Kiatuthlanna (1931), the Village of the Great Kivas (1932),
and at Whitewater (1939, 1940), Roberts laid bare a cultural sequence
extending from BM. III and P. I pit-houses to the terraced buildings
of Pueblo III. Echoes from the Chaco were everywhere present and
so were traits from the Mesa Verde country. With the sipapu an
almost constant feature and base slabs overlain by coursed masonry
occurring more frequently than is customary for their kind, the
Whitewater pit-houses may represent a local architectural advance
over those elsewhere or they may reflect usurpation of existing dwellings
by later immigrants. Pit-house Number 2, for example, provided
a tree-ring date of A.D. 814 while a room in Unit 3, a P. II building,
gave one 200 years later, A.D. 1014, obviously late for a P. II structure.
Summarizing his researches in this area, Roberts wrote (1939,
p. 263) "the ruins . . . represent a peripheral lag in the Chaco
pattern and, despite many recent expressions of opinion to the contrary,
the flow of influence was from the Chaco . . . and not the
reverse."
The Village of the Great Kivas on the Zuñi Reservation with its late
Chaco masonry and pottery (Roberts, 1932); the two pre-Zuñi kivas
near Hawikuh (Hodge, 1923); the typically Chaco black-on-white
pottery from small-house sites south of Fort Defiance, Ariz. (Kidder,
1924, p. 56); the two earth-walled kivas at Site LA 2505 about 20
miles north of Gallup, N. Mex., one (B) with a deep south banquette,
sipapu, above-floor ventilator, and tree-ring dates of A.D. 1020 and
1047 (Smiley, 1951, p. 26), and the other (A) with a lateral ventilator
sealed and replaced by one of the subfloor variety (Bullard and
Cassidy, 1956); the late Chaco masonry, Chaco-like pottery, and a
Great Kiva in Manuelito Wash, south of Gallup (Reed, 1944, p. 167;
Judd, 1954, p. 34), and Seltzer's comparative data (1944, p. 17) on
Old Zuñi and Pueblo Bonito skulls, all offer strong support for the
theory of a southward trek from Chaco Canyon in the twelfth
century or thereabout.
Gladwin (1945), Martin (1936), and O'Bryan (1950) are among
those who see the Chaco Culture rising from BM. III and P. I pithouses
of the Little Colorado-Puerco drainage and spreading thence
northward through increasingly larger settlements to its demise in
Chaco Canyon. Roberts and the present writer see distribution in the
to put a finger on the place of origin.
At Pueblo Bonito we have the distinct P. II culture of Old Bonito
and the better known P. III culture of the Late Bonitians. Both were
born somewhere among the sage-covered mesas and valleys of southeastern
Utah and southwestern Colorado. There, among those valleys
and mesas, the whole panorama of Pueblo architectural development
lies exposed to view—a development that extends from single, earthwalled
pit-houses to the wide-curving post-and-mud surface communities
of P. I, to the wall-width masonry and crescentic grouping of
P. II dwellings and storerooms and, finally, to the many-roomed,
multiple-storied towns of Pueblo III. Somewhere in that far-reaching
scene, and most likely where Pueblo II flourished, eventually
will be found the cross-road from which the so-called Mesa Verde
and the Chaco peoples took their separate ways.
The masonry-lined subfloor vault of Chaco kivas, although of
unknown purpose, seems so unusual a feature its origin and development
should be traceable. But nowhere among published descriptions
do I find anything even remotely comparable except the oval depressions,
filled and floored over, reported by Roberts (1939, p. 106) in
P. I Structure 12 at Whitewater, Arizona, and by O'Bryan (1950,
p. 34) in a P. II kiva at Site 102, Mesa Verde National Park, Colo.,
150 miles distant.
At some as yet undetermined point in Pueblo prehistory, clan
ritualists replaced the four traditional roof-supporting posts with
masonry columns and later replaced these with low-log-enclosed
pilasters. At some still unidentified stopping place they introduced
a new, subfloor type of ventilator, eliminating the deflector. The
one Morris (1939, p. 53) describes in a P. II kiva under Building I,
Site 39, is the earliest of which I am aware but a companion kiva,
Number 2, was typically Mesa Verde with its 6 masonry pilasters, a
deep south banquette, and a lateral ventilating system. I know of
no horizontal, log-enclosed pilaster earlier than those in the bowlshaped
P. II kivas at Pueblo Bonito.
Lowry Ruin, the West Pueblo at Aztec, Solomon's Ruin near
Bloomfield, N. Mex., and the short-lived structure on Chimney
Rock Mesa near Pagosa Springs, Colo. (Jeancon, 1922; Roberts,
1922), are among those repeatedly described as probable colonies from
Chaco Canyon. Available tree-ring dates as listed by Smiley (1951)
lend credence to this theory of colonization since a majority—49 from
Aztec, A.D. 1110-1125; 9 from Lowry, 987-1086; 3 from Solomon's
Bonito. But the subject is not so easily dismissed.
The East Kiva on Chimney Rock, which seems so isolated and
alone, is undeniably Chaco-like in its lack of a sipapu, its low logenclosed
pilasters, its west-side vault, and a subfloor ventilating system
that was rebuilt to the original plan when the floor was raised.
These Chaco resemblances invite further exploration, but at the time
of his initial visit Roberts (1922, p. 12) recorded his then opinion
that Chimney Rock pottery looked older than that of Chaco Canyon.
If he is correct in this early impression then the associated Chaco-like
masonry must be older and Chimney Rock stands not as a colony from
Chaco Canyon but as a possible way station on a path southward.
"The Chaco-like remains north of the San Juan, both architectural
and ceramic," wrote Earl Morris (1939, p. 204), "are so widespread
and so numerous that I consider it untenable to view them wholly as
an extension of or a backwash from, the Chaco Canyon center. . . .
The most Chaco-like of the vessels from the north country, which
seem so significant when viewed singly or selectively grouped, become
far less so when viewed as the minor component that they are of the
totality of wares among which they occur."
Kidder obviously had the same intangible evidence in mind when
he observed that Chaco-like vessels from Montezuma Valley and
McElmo Canyon, target of commercial and amateur collectors for
half a century, "are seldom of the most pronounced Chaco types; they
give one the impression of being either the product of a peripheral
development affected by Chaco influence, or of an earlier and less
specialized stage of the Chaco culture" (Kidder, 1924, p. 56). They
may, he added, "indicate a northwestern spread or a northwestern
origin" of that culture. Anna Shepard may have seen a like probability
when she hinted (1939, p. 285) a common source in early P. III
times or previously for the mineral-paint Chaco and the carbon-paint
Mesa Verde wares.
Morris doubtless would have regarded Chimney Rock pottery as
"more Chacoesque than Chaco." To him the many small ruins
throughout the San Juan and Animas valleys in which Chaco-like
pottery predominates might be earlier, contemporary with, or even
later than Chaco-like Aztec pueblo (Morris, 1928, p. 418). He does
not so imply but some one of these small ruins may have spawned the
unique ideas in architecture and in pottery ornamentation that brought
about the Pueblo III conquest of Chaco Canyon.
In pursuing our investigations for the National Geographic Society
Chaco Culture apart, but we did discover a great deal previously
unknown about Pueblo Bonito. We learned that it is the architectural
product of two unrelated peoples; that the first of these had been
in residence long enough for 5 or 6 feet of blown sand to pile up
against their homes before the second group arrived and built upon
that sand. Source of the domestic water supply at Pueblo Bonito
remains a mystery, but we turn with increasing favor toward the
Navaho tradition that water could be had with shallow digging almost
anywhere in the valley before erosion of the 1850 arroyo. The great
natural cistern on the north cliff overlooking Pueblo Bonito was of
limited capacity, even if it existed in A.D. 1000.
We learned that the forests which furnished roofing timbers for
Pueblo Bonito flourished when rainfall was more abundant than it is
today; that slow-flowing floodwaters following summer rains had
spread widely across the valley floor annually depositing enough black
alkali to lessen the productivity of village fields before a 12th-century
arroyo lowered the water table beyond reach of surface vegetation.
We learned that each of three Late Bonitian additions to the
original settlement had forced the abandonment and destruction of
dwellings previously built; that plans for a fourth and more extensive
addition were left incomplete and a substitute adopted. We learned
that this substitute was itself abandoned when the Late Bonitians
migrated, leaving their Pueblo II co-residents behind in sole possession
of the compound pueblo. That these original settlers were
last to depart is clear from the foodstuffs, the household utensils,
and the ceremonial paraphernalia they left in their brush-roofed
houses.
Reduction in arable lands, a consequence of reduced rainfall or
erosion, seems a most likely cause for desertion of Chaco Canyon by
the Bonitians. The Great Drought of 1276-1299 occurred 100 years
too late to have been influential, but that of 1090-1101, perhaps an
incentive for Bryan's buried arroyo, could have spurred the outgoing.
In the chapters which follow I shall seek to present Pueblo Bonito
as we now know it, from the original P. II settlement to the last of
the three additions planned and executed by the Late Bonitians.
| The architecture of Pueblo Bonito : | |