Puebloans - Aztec Regional System

At a glance

Why is it called "Aztec"?

The park service answers the question in one sentence: "The name is a misnomer." In the nineteenth century, popular and scholarly opinion held that the ancestors of the Aztecs had migrated south into Mexico from the American Southwest. Euro-American explorers therefore gave Southwestern ruins names like "Aztec", "Montezuma" and "Toltec" — names that flattered the ruins and said nothing about the people who built them.

The mechanism is worth getting right, because the lazy version — "the settlers thought the Aztecs built it" — misses the source. The Spaniards did not name it. The Aztecs were 1,500 miles away in the central Mexican highlands, and as Robert and Florence Lister point out in the monument's administrative history, the Spaniards would not have known that the Aztecs had not emerged as a cultural entity until long after the Animas houses had been deserted. The name came from United States settlers and explorers, and behind them from William H. Prescott's History of the Conquest of Mexico, a bestseller of 1843–44 that repeated the old idea of an Aztec origin in a vague "north land". Mountain men in the Animas Valley began calling the wrecked walls "the Aztec houses"; a citizen of Animas City wrote in 1861 of "Stone buildings three stories high are yet standing, of Aztec architecture"; and when settlers arrived in the valley about 1876, frequent repetition fixed the name. The town of Wallace renamed itself Aztec after the ruin, not the other way round.

The first trained eyewitness got it right. John S. Newberry, a geologist with the Corps of Topographical Engineers, visited in 1859 and described "yellow Cretaceous sandstone masonry, with some exterior walls standing 25 feet high and interior walls retaining plaster" — and attributed the ruins correctly to the forebears of the Pueblo Indians, with no Aztec conjecture at all. In 1875 F. M. Endlich of the Hayden Survey proposed a better name, "Acropolis", and a bigger plan: a squared horseshoe opening south, with perhaps 500 rooms. Neither name stuck.

What is the Totah?

"Totah" is a Navajo word, Tóta', meaning "where the waters come together", and it describes the country where the San Juan, Animas and La Plata rivers converge at modern Farmington. It is a landscape term with an unusually good fit: three permanent streams in a region where water is the limiting factor, and the largest concentration of Chacoan great houses outside Chaco Canyon.

It is also a term with a politics. Archaeologists began using the anglicised "Totah" in the 1990s for the archaeological region formed by the three valleys. Many now prefer "Middle San Juan", on the grounds that it is older — Cynthia Irwin-Williams used it during the original Salmon project — and neutral, in a region where Pueblo and Navajo land claims are live. This entry uses both, and means the same thing by them.

The distance to Chaco is usually given as 55 miles or 85 km; site-to-site measurements give about 89 km north of Pueblo Bonito, at a bearing of three hundred and fifty-seven degrees. "About 80–90 km north of Chaco Canyon" is the safe wording.

Who built the great houses?

For eighty years the answer was simple: Chacoans did. Earl Morris, who excavated Aztec West for the American Museum of Natural History from 1916, believed Chacoans built it; Cynthia Irwin-Williams, who dug Salmon Pueblo in 1970–78, called that site a Chacoan colony. That colonial model is still what most popular accounts repeat.

The current answer is more interesting, and it comes from a deliberate test. Between 2005 and 2009 an NSF-funded project run by Archaeology Southwest, Salmon Ruins Museum and Aztec Ruins National Monument asked whether Chacoans migrated north and built these places, or whether local people copied Chacoan architecture. The method is worth understanding, because it explains how archaeologists argue about migration at all. Highly visible things — public architecture, ceremonial objects, pottery designs — are meant to be seen and are easily copied, so their spread proves nothing about who moved. Low-visibility things — how a wall is built, what temper is in the clay, where the timber came from — are not displayed and tend to travel only with people.

Applied at Aztec, the low-visibility evidence gives a mixed and sequenced answer.

Pottery. Lori Stephens Reed examined several hundred whole vessels and thousands of sherds and found three distinct technological signatures, not two: pots made in Chaco Canyon and carried north; pots made by Chacoan potters using local clay but keeping Chacoan sand-and-sherd temper; and pots made by local potters imitating Chacoan style with crushed-rock temper. Most tellingly, the local potters could not accurately reproduce the complex Chacoan designs. Her conclusion is that the great houses were built through "mutual engagement on the part of local and Chacoan artisans and specialists".

Architecture. Gary Brown and Cheryl Paddock found a stick-reinforced adobe wall in the south wing of West Ruin — parallel thin poles set inside adobe and masonry, a technique not documented in Chaco Canyon and close to local vernacular building. Brown reads it as a hybrid, a local Animas tradition into which Chacoan masonry was introduced. That is about as direct as evidence gets that local builders were on the scaffold, not just in the fields.

Tree rings. The sourcing study overturns the park's own signage. Of 319 beams sourced at Aztec, 45 per cent came from the Chuska Mountains more than 100 km south-west — 64 per cent of Aztec West's assemblage — and 45 per cent were local juniper from the San Juan valley, which supplied 80 per cent of Aztec East. Only 5 per cent came from the San Juan Mountains 25 km north, the range the park's web page still credits. The authors rule out the nearby range, and suggest floating logs down the Animas was harder and less predictable than assumed.

Put together, the sequence is: local people built a Chacoan-plan great house in adobe at Aztec North around 1060–1080, before any Chacoans arrived; Salmon Pueblo was founded about 1090 as a genuine Chacoan enclave with perhaps 50–75 Chacoans among 150–200 residents; Aztec West followed from about 1100, built by a larger Chacoan group working with local masons and adobe specialists; and Aztec East continued the programme from about 1120. In total the Aztec great houses held nearly 900 rooms, with more than 600 more in the surrounding small pueblos — a clustering unprecedented outside Chaco Canyon.

Why move north?

Five explanations are live, and they are not mutually exclusive.

Chaco failed. Construction in Chaco Canyon stops by about 1130; the last beam date from a Chacoan great house is 1128; the system's regional role ends by about 1140. A leadership that had run the canyon for a century needed somewhere to go.

The rivers pulled. Salmon and Aztec were built in fertile alluvial valleys beside the largest rivers in the northern Southwest, in country where the water-management techniques developed at Chaco could be applied at a larger scale. Both sites produced large quantities of corn. The park service notes that Chaco has far less water and arable land, and infers that Aztec farmers helped feed the canyon — an inference, not a measured result.

The flood thesis. Reed's own explanation is the best original idea in the literature. Salmon Pueblo sits about 200 metres from the San Juan and was built during the low-flow drought of the late 1080s and early 1090s. When the river returned to full discharge, the Chacoans there discovered their mistake: flood deposits were later found in rooms at both corners of the site and inside the great kiva, which was reroofed and given a berm perhaps two metres high in the 1260s. Aztec West was built more than 400 metres from the Animas, a much smaller river. On Reed's reading, the Salmon Chacoans moved upriver and helped build Aztec East.

Factions. Ruth Van Dyke argues that environmental and social instability at the end of the eleventh century weakened faith in Chacoan leaders and that rival factions emerged; some leaders left to found new centres in the Middle San Juan. Aztec, in this reading, was "an overt attempt to replace Chaco as the center of the Puebloan world", and the new McElmo great houses at Chaco were a rival faction's attempt to hold followers.

The Chaco Meridian. Stephen Lekson's argument is that Chaco, Aztec and Paquimé were deliberately placed on the same line of longitude and that a small elite moved the capital north around 1125. The geometry is real: Aztec is about 89 km from Pueblo Bonito at 357 degrees, roughly three degrees off true north. The rebuttals are also real. The Great North Road, the supposed route, has been traced only as far as upper Kutz Canyon; efforts to follow it further have failed, and Lekson's case requires projecting it onward over eroded ground. David Phillips of the University of New Mexico called the argument "novel, provocative, and wrong"; his critique is a detailed circulated paper rather than a peer-reviewed refutation, and should be described as such. Treat the Meridian as the most stimulating and least accepted idea in the field.

What was Aztec West?

The largest known great house outside Chaco Canyon: at least 400 rooms on three storeys, with the park service's own answers ranging from four hundred to five hundred, and about 30 kivas, of which 18 had been identified by 1921. It is not D-shaped in the classic Chacoan sense — the plan is rooms around a plaza with a long straight back wall — but it is unmistakably Chacoan in its masonry, its core-and-veneer walls, its great kiva and its orientation to the south.

It was built fast. Cutting dates on its beams run from 1106 to 1131, with the heaviest cluster in 1111–1115; the park service divides the work into three stages — Kiva L and the first core about 1100–1109, the north and east wings about 1110–1120, the west wing about 1120–1130 — and notes that most of Aztec West was complete in thirty years, against nearly 300 for Pueblo Bonito. Two renovation phases followed: a McElmo phase in the late 1100s, when rooms were demolished and large first-storey kivas blocked in, and a Mesa Verde phase in the mid-1200s, when rooms were subdivided and kivas were sunk into second-storey rooms.

Two details make the building legible. The long straight back wall is a solstice instrument: from its west corner the summer solstice sun rises directly up the east corner, and from the east corner the winter solstice sun sets down the west corner. The park service keeps the necessary hedge — no one can prove the alignments were intentional — while noting that such alignments recur at major sites across the region. The masonry carries decorative bands of green stone, which may mimic local rock strata; T-shaped doorways face the plazas, and their purpose is still argued.

The wider community was more than the monument. Aztec North sits on the mesa above; Aztec East lies 150 yards to the north-east, with two mounds and a great kiva larger than the west one's; there are three tri-walled structures, including the Hubbard Site, 64 feet across with 22 rooms around a central kiva, and Mound F, the largest such structure in the Southwest. The Earl Morris Ruin, unexcavated on the monument's edge, may hold another 63 rooms and seven kivas. About 52 contemporaneous sites surround the great houses.

What was inside the Great Kiva?

The great kiva at Aztec West is the best-known ceremonial room in the Pueblo world, and the strangest thing about it is that most of what you see is 1930s work. Earl Morris excavated it in six weeks in February and March 1921, and described it as "an enormous wheel": a circular kiva proper, 12.6 metres across at floor level and 14.7 metres at bench height, surrounded by a concentric ring of arc-shaped rooms. Two stairways entered on the north–south axis. Four masonry pillars, 2 feet 8½ inches square and built of alternating layers of stone and poles, rose from the floor; Morris calculated that the ceiling was at least 16 feet high and that the pillars were at least that tall.

The roof was his reconstruction, and it is a construction of genius and guesswork. He found charred fragments of great logs, saplings and cedar splints, worked out five pairs of girders resting on the pillars, 23 radial timbers and three parallel logs, and rebuilt a flat circular roof of about 90 tons on four columns three feet square and an estimated 18 feet tall, each standing on a stack of four stone slabs over a base of compacted lignite. The north alcove, Room 160, opened directly into the kiva seven feet above the floor and held a plastered, whitewashed and red-tinted masonry block with a white circle on top — the ritual focus of the room.

The reconstruction matters because it is usually credited to the wrong people. The Civilian Conservation Corps did not rebuild the great kiva. In 1933 it was "an offensive tumble of stones and a pond of standing water after every rain"; Morris asked for $10,000 to restore it, was appointed a collaborator-at-large in the National Park Service with his salary paid by the Carnegie Institution, and began work in late May 1934 under Public Works Administration project F.P. 503.14, at a cost of $12,450.76. The digging changed the design: two benches were found, a Mesa Verdean remodelling over "a more carefully crafted, earlier Chaco bench", and Morris decided to restore the Chaco-phase room rather than the later one. He could not find a Chacoan great kiva roof to copy — "the result was largely negative, hence I shall have to rebuild guided entirely by local evidence" — and the four columns and 90-ton roof are his inference from charred beams. The red-and-white paint is modern too, mixed to a formula by the Santa Fe architect John Gaw Meem. The CCC's actual contribution was separate: grounds work in 1935 and wall stabilisation by a Native American mobile unit between 1938 and 1942.

The park service now rejects total rebuilding of this kind, and says so. The great kiva is nevertheless the reason most visitors come, and Morris's justification was explicit: to make available "an example of the most intricate sort of sanctuary that was ever developed by the Pueblo people".

Who lived in the great houses?

Not kings, and probably not a palace. The old picture of Aztec West as an elite residence has faded, and the current reading is closer to a planned ceremonial and administrative centre with a substantial resident population around it. What the rooms show is use changing over time: large Chacoan chambers in the early 1100s; adobe partitions splitting them in the 1200s; small round kivas built inside former square rooms and under the plaza; and Chacoan rooms ending up as middens, burial places and turkey pens. The park service's own sequence describes a McElmo renovation, then a Mesa Verde renovation, after which the building may have been more heavily lived in.

The dead are informative too. There is no cemetery. Some burials were placed inside walls and under floors at West Ruin; many more were laid in the middens around the settlement. That pattern fits a community that treated the spaces around and inside its buildings as part of the same social world rather than zoning a graveyard away from the living.

Hierarchy is harder to read than at Chaco. There is no Room 33 equivalent, no crypt with a 330-year matriline, no single burial with 11,000 turquoise beads. What there is, is a planned community of nearly 900 great-house rooms plus 600 in small pueblos, arranged around plazas and great kivas, with tri-walled structures whose function is genuinely unsettled — Vivian suggested "the turf of an incipient priestly class", and the park service keeps the hedge that ceremonial and domestic use are both visible. The honest answer is that Aztec shows leadership without a documented dynasty, and a ceremonial order without a documented priesthood.

What did they eat?

Maize, on land that could actually grow it. Both Salmon and Aztec sit on alluvium beside permanent rivers, and irrigation is documented rather than assumed: at least two ancient ditches near Aztec were recorded by Newberry in 1859, and the park describes flood irrigation through small ditches, plus basin-and-waffle grids around individual plants. Crops were corn, beans and squash, with tolerated greens — goosefoot, amaranth, purslane, beeweed, sunflower.

The archaeobotany gives a sharp two-phase contrast. The Chacoan occupants relied heavily on corn and little on wild plants, had limited access to gourds and cotton cloth though they acquired some by trade, and organised work teams that travelled long distances for beams. The later, local occupants gathered a much wider range of wild plants, apparently ate less corn, had no nonlocal plant foods in their trash, reused Chacoan roof timbers and cut local juniper when they had to, and used juniper bark and yucca for bedding, baskets and sandals. And at the end: as the 1200s closed and the climate changed, they ate more wild plants and resorted to famine foods — juniper bark, yucca leaves and leftover corn cobs.

That sentence is the best summary of the collapse in the literature, and it contains no warfare at all. Turkey was kept, deer and antelope hunted, and rabbits trapped; the middens also preserve corn pollen on room floors, which may mark prayers rather than meals.

What did they trade?

The same exotics that circulated at Chaco, in smaller quantities: turquoise, marine shell, obsidian and pottery. Three macaw skulls and a skeleton were found at Aztec, and nine macaw burials at Salmon, some ochred. Cacao residues have been identified in mugs from Aztec — the same Crown and Hurst work that found chocolate at Chaco — which means the Chacoan drinking complex travelled north with the Chacoans. Copper bells and Pacific shell complete the Mesoamerican set.

The pottery tells a political story. Before the late 1000s the Middle San Juan imported little nonlocal pottery, mostly from Mesa Verde; when local leaders began building great houses the focus shifted south, and Chuska corrugated cooking pots arrived in quantity. After the mid-1100s the Chuska share collapses, along with the Chuskan timber supply — the material signature of a system losing its southern connections.

The exchange system had a geography. Turquoise came from the Cerrillos Hills and other source areas far to the south-east; marine shell from the Gulf of California and the Pacific coast; obsidian from sources in the Jemez and other volcanic ranges; macaws and copper bells from Mexico, at the end of a chain of middlemen rather than a single voyage. Chuska pottery arrived from the south-west in quantity until the mid-1100s, then stopped. Aztec itself made no redware and no turquoise ornaments on the Chaco scale; it imported finished goods and raw materials and exported, so far as the evidence goes, maize, hides and possibly salt.

One negative is worth stating cleanly: there is no kachina religion here. The kachina cult is a Pueblo IV development, after 1300, after the Totah empties. What the kivas show is the older Chacoan suite — sipapu, benches, floor vaults, niches, foot drums. The "Aztec" name has encouraged generations of readers to see Mexican religion in a Puebloan building; the archaeology does not support it.

Did the Chacoans stay?

This is the live argument at Aztec, and it has two sides.

Morris's model, which dominated for most of the twentieth century, was two occupations: Chacoans from about 1110 to 1140, then a break, then Mesa Verdeans from about 1200 to 1290 who reoccupied the great houses. His evidence was stratigraphic and specific: in the great kiva, Chacoan use, then disrepair and drifts of trash, then refurbishment, then a fire that consumed the roof; in Kiva P, a Mesa Verde chamber built directly over a Chaco chamber, with Mesa Verde pottery in the upper debris and Chaco black-on-white below; and the same sequence of floors in six more rooms. The Mesa Verde bench over the Chaco bench in the great kiva is the cleanest single marker.

The revisionist model, now the park service's public position, is continuous occupation from about 1070 to 1290. The ceramic sequence runs through without a break, from local Mancos through McElmo to Mesa Verde Black-on-white; radiocarbon dates on corn cobs come from the same deposits; and the park's own text says: "Contrary to previous thinking, archeologists do not necessarily attribute each phase to population migration. This structure was inhabited continuously, and construction styles evolved over time."

Both sides use good evidence, and the difference is partly what counts as a "people" — a migrating population, or a change of habit inside one community. What is not disputed is the material sequence: the Chuska ponderosa of West Ruin gives way to local juniper at East Ruin; the Chuska pottery declines; rooms are subdivided; kivas are sunk into upper floors; the great houses end up used for storage, burials and turkey pens. Whether that is replacement or adaptation is the argument.

When did it end?

The Chacoan system at Aztec ended in the 1140s, and the place did not.

Construction stops in the early 1130s — beams run 1106–1131, with a break at 1111–1118 — and the Chacoan supply system behind the buildings disappears by about 1140–1150. The drought that did it is usually misdated. It runs from 1130 to 1180, with its worst phase in 1140–1162: 23 consecutive years of negative Palmer Drought Severity Index values, the greatest megadrought recorded for North America, and Colorado River flows at Lees Ferry that were the lowest in 1,200 years. Reed and Brown run it to 1190; different reconstructions draw the edges differently.

Then the great houses were reworked rather than abandoned. Aztec East was built and remodelled across 170 years, into the 1260s; the west great house was subdivided and reused; and the community carried on as a Middle San Juan pueblo. Salmon tells the same story: a Chacoan foundation about 1090 with 275–325 rooms, remade as a San Juan-style pueblo from about 1125, with more than 20 small kivas inserted into rooms and plazas, occupied until the 1280s.

The end came with the Great Drought of 1276–1299, but the drought did not cause the migration so much as finish it. In the central Mesa Verde region the population had already begun to fall before 1260; the drought finalised a movement already under way. At Aztec, Reed and Brown's chronology describes a catastrophic, intentionally set fire sweeping Salmon and much of the western great house in the 1280s; the deposits are real, the intention is their reading. By about 1300 the greater San Juan Basin was empty of Pueblo farmers, and the people had gone to the Rio Grande, the Zuni–Acoma region and the Hopi mesas.

What is left?

Aztec Ruins National Monument was proclaimed by Warren Harding on 24 January 1923 — 4.6 acres around a ruin "of great antiquity and historical interest" — and renamed and enlarged by Coolidge in 1928. It now covers 318 acres inside the town of Aztec, in the United States. The monument's archaeology is largely the work of Earl Morris and the American Museum of Natural History, whose five-part report appeared between 1919 and 1928; most of the finds are still in New York, with others in Tucson and a few handsome pieces at the park.

The UNESCO listing has an awkward history. Aztec was not in the original 1985 nomination, and ICOMOS regretted its exclusion; the revised nomination of April 1987 added "Pueblo Aztec" and five other sites, and the property was inscribed under criterion (iii) only. The listing has no buffer zone, which is one reason the landscape around the monument matters so much. Aztec Ruins is about 55 miles north of Chaco and outside the 2023 mineral withdrawal; its own exposure is local, with active oil and gas wells reached by a dirt road beside a great house site, as the 2015 foundation document records.

Twenty-six tribes are traditionally associated with the monument. The 2014 NAGPRA notice names the Hopi and Zuni as culturally affiliated and notified 27 tribes. The park's own interpretation now leads with the point this entry opened with: the builders were Ancestral Puebloans, the name is a misnomer, and "most Pueblo people today prefer that we use the term 'Ancestral Puebloans' to refer to their ancestors".

Common myths about the Aztec regional system

Timeline of the Aztec regional system

What the map shows next

At 1001 on the map the story is still Chaco's: the Chaco regional system is at its height, and the Totah great houses have not yet been begun.

At 1400 on the map the Totah is empty, the Norse Greenland colony is in its last decades far to the north-east, and the Pueblo world has reorganised along the Rio Grande.

At 1616 on the map the Haudenosaunee confederacy dominates the north-east, and the descendants of the Aztec builders are the Pueblo communities of New Mexico and Arizona.