Chaco Canyon · United States · 36.0611, -107.9614 · PYR record in the living atlas
A Chacoan roof beam is about 4.5 metres long, 20 to 25 centimetres thick, and weighs somewhere between 85 and 140 kilograms. There are no wheels in the ancient Southwest, no draught animals and no navigable river. Every one of those beams was carried in on people's backs, most likely with tumplines across the forehead, from mountains 70 kilometres away or more. Between roughly 850 and 1150 AD the builders of Chaco Canyon brought in at least 240,000 trees that way, and stacked sandstone into great houses whose scale would not be matched in North America until the 19th century. Treat that total as an estimate rather than a count: published figures start around 200,000 and run upwards, the most cautious sources say only "many thousands of beams", and nobody has counted them.
The most precisely dated ruins on the continent
Almost every date in this article comes from wood. The roof beams at Chaco carry tree-ring sequences that can be matched against regional chronologies, which dates the felling of individual timbers, which in turn dates construction episodes at individual great houses to the decade and in some cases to the year. That is an extraordinary level of resolution for a prehistoric site anywhere on Earth, and it is the reason Chaco can be argued about in a way that most places cannot. Ceramic seriation, archaeomagnetic dating and radiocarbon fill in around it.
Where the timber came from was worked out slowly and by several methods, which is why the answer is trustworthy. Julio Betancourt and colleagues identified species by wood anatomy in 1986. Nathan English's team applied strontium isotope signatures in 2001, and Amanda Reynolds extended that with over 200 living trees in 2005. In 2015 Chris Guiterman matched the tree-ring patterns of 170 timbers from seven great houses against regional chronologies and found that after about AD 1020 roughly 70 per cent of the wood came from the Chuska Mountains to the west. Earlier and later, beams came from Mount Taylor and the San Mateos, and from the La Plata and San Juan ranges to the north. The sourcing shifted; it did not drift.
Pueblo Bonito, the largest great house, grew over roughly three centuries and many separate building episodes into a D-shaped complex with two great kivas and about forty smaller ones, standing at least four storeys and possibly five in places, with beams cut between 1033 and 1092 during the peak of building. Its room count is usually given as 800 and sometimes as 600, and the honest answer is between about 600 and 800 depending on whether later-phase and reconstructed rooms are counted in. Around 200,000 pieces of turquoise have come out of the canyon, and more than 15,000 beads and pendants accompanied two burials at Pueblo Bonito alone. Marine shell and copper bells reached the canyon too, and the remains of scarlet macaws, birds from far to the south. Cylinder jars from Pueblo Bonito tested positive for cacao residue in work Patricia Crown and W. Jeffrey Hurst published in PNAS in 2009, which is the single most concrete piece of evidence that this canyon sat at the northern end of a network reaching into Mesoamerica.
How many people, and how they ate
Two questions that sound basic are not settled. The first is how many people actually lived here year round. Published estimates range enormously and there is no accepted consensus figure, which is worth knowing before reading any argument that turns on Chaco being a populous capital or a thinly inhabited ceremonial centre.
The second is how a centre in a low-rainfall basin fed itself. Chaco is usually described as a capital with almost no farmland of its own, which frames the problem as one of rainfall and stops there. What the Chacoans did about it is the more interesting question, and in 2024 Wills, Williams and Dorshow published work in KIVA on the Bonito paleochannel cycle, examining water management in the canyon and how far it represented communal investment as against household-scale effort. That is a live line of research and it bears directly on what kind of society this was.
The dagger is not what it was
In 1977 the artist Anna Sofaer, volunteering to record rock art, climbed Fajada Butte at the canyon mouth and watched a blade of light cut through a spiral pecked into the cliff. Three large stone slabs lean against the rock face near the summit; they channel light and shadow onto two spirals. At about 11:15 in the morning on the summer solstice a dagger of light bisects the larger spiral. Other patterns fall at the equinoxes and at the winter solstice, and Sofaer's Solstice Project has argued that shadow positions on the same spiral track the moon across its 18.6 year standstill cycle.
Sofaer is often introduced as the amateur who found the Sun Dagger and left it there, which undersells her. The work is peer-reviewed and it goes well beyond the butte: Sofaer, Zinser and Sinclair published in Science in 1979, and the Solstice Project has since built a substantial published corpus on the orientation of great-house walls and road segments to solstice and lunar standstill azimuths. Specific alignments in that corpus are accepted. The general claim is not: not every researcher agrees that every proposed wall or road alignment at Chaco is intentional rather than a coincidence of a landscape with a lot of straight lines in it. Take the individual measured cases seriously and the blanket statement that the Chacoans encoded astronomy into everything with rather more care.
Here is the part the documentaries leave out. In 1989 one of the three slabs shifted, the movement attributed to erosion around its base caused by visitor foot traffic. The slabs were stabilised in 1990, but the displaced one was not returned to its original orientation, and the assembly has lost some of its former precision. The National Park Service had already closed the butte to the public in the 1980s. The most celebrated archaeoastronomical device in North America was degraded by the attention that discovering it produced, within twelve years of anyone noticing it existed.
The meridian, and why we are careful with it
Chaco has one named alternative theory with real academic standing behind it, and it happens to be about exactly the kind of thing our engines measure. Stephen Lekson's The Chaco Meridian, first published in 1999 and revised in 2015, argues that Chaco Canyon, Aztec Ruins about 88 kilometres north, and Paquime some 630 kilometres south in Chihuahua were deliberately placed on a single line of longitude, near 107 degrees 57 minutes, as an elite moved north and then south over several centuries. In its strongest form the case is not silly at all: the three sites share architectural signatures including T-shaped doorways, colonnades and room-wide platforms, and the Chacoans demonstrably could lay out long straight lines, since they built roads that do exactly that.
David Phillips has set out the case against, and it is worth reading in full because it is a model of how to test this kind of claim. Aztec sits about 3.7 kilometres west of Chaco's meridian over a distance of 86 kilometres, a deviation of roughly 2.5 degrees, which exceeds the two degree tolerance Lekson himself allows for Chacoan surveying error. The Great North Road, usually cited as proof of the surveying skill required, does not help: of the 50.3 kilometres between Pueblo Alto and Kutz Canyon, Phillips calculates that only about 15.5 kilometres actually runs north to south. The chronology has its own problem, since on Lekson's dates construction at Aztec begins around 1110, about fifteen years before the elite are supposed to have left Chaco. And underneath all of it sits the methodological objection: the alignment was noticed first and the theory built afterwards to explain it, so the improbability of the alignment is not evidence, because it was never a prediction.
One clarification on that road, because two true-sounding statements about it are in circulation. Survey descriptions have it running approximately due north from Pueblo Alto, and that is how it is usually summarised. Phillips's figure is about how much of its actual measured length holds a north to south bearing. Both can be right at once, and the disagreement is really about what "runs north" is being asked to mean. It matters here only because the road is being used as evidence of surveying precision, and for that purpose the measured length is the relevant number.
That is the trap our own engine exists to avoid, which is why its settings are fixed before the run and identical everywhere.
Room 33, and who gets to ask
In 2017 Douglas Kennett and colleagues published in Nature Communications the results of sequencing nine individuals from Room 33 at Pueblo Bonito, a burial crypt in use from roughly 800 to 1130 AD, about 330 years. All nine share an identical mitochondrial genome. Nuclear data from the six best-preserved individuals identified a probable mother and daughter one generation apart, and a probable grandmother and grandson two generations apart. The authors argue this is direct evidence of institutionalised hereditary leadership passed down the female line, in a society with no writing to record it.
That study lands inside an older argument it did not start. How Chacoan society was organised has been contested for decades on the material record alone. Stephen Lekson argues for pronounced hierarchy, with the great houses as elite residences and the region as something close to a polity. Others read exactly the same buildings, burials and deposits as evidence of corporate, communal decision-making without a ruling class. Both positions are held by working archaeologists reading the same evidence, and naming one genetic paper as though it closed the question would misrepresent where the field is.
The part that belongs in any honest account is what happened around the study. The American Museum of Natural History authorised the destructive sampling without consulting tribes, relying on an earlier determination that the remains were culturally unaffiliated. Leigh Kuwanwisiwma of the Hopi Cultural Preservation Office and Kurt Dongoske of the Zuni Tribal Historic Preservation Office both objected, Dongoske arguing that leaving cultural affiliation to be decided by an institution of the settler state is itself the problem. A later genomic study carried out at the request of Picuris Pueblo, on the pueblo's own initiative, linked its people to Chaco-era ancestors. The contrast between those two projects is the most useful thing in this whole section.
Who dug it, and who is allowed to now
Spanish expeditions noted the ruins in the eighteenth century, and the US Army Corps of Topographical Engineers documented them in 1849. Excavation proper starts in the 1890s with Richard Wetherill and the Hyde Exploring Expedition, whose Pueblo Bonito material is largely held at the American Museum of Natural History, which is how the collection at the centre of the Room 33 argument came to be in New York. The systematic modern work is the National Park Service's Chaco Project, which ran from 1971 to 1986 and published twenty technical monographs.
Practice has changed. New excavation at Chaco is now strongly discouraged in favour of non-invasive survey and consultation with descendant communities, which means ground-penetrating radar rather than trowels. That radar has found buried features whose function is not understood, and under current norms they will stay in the ground. Whether that is a loss or a correction depends on who you think the canyon belongs to, and the profession has largely settled on an answer.
Straight roads, and a canyon nobody lost
Pueblo Bonito's principal interior wall runs close to north to south and part of its facade close to east to west, and the same east to west line has been reported running between Pueblo Bonito and Chetro Ketl. Roads up to nine metres wide, scraped down to hard caliche, run out of the canyon toward outliers across the San Juan Basin, some of them for tens of kilometres. The network reached more than 150 outlier great houses across the Four Corners region, which is the scale that makes the question worth asking. Why a society without wheels or pack animals built highways is genuinely unresolved: pilgrimage routes, formal approaches, and statements written into the landscape are all on the table, and nobody has closed it.
The strongest recent evidence leans toward the landscape reading. A Dartmouth College team using lidar near the Gasco site, south of the park, documented road segments aligned on winter solstice sunrise over Mount Taylor, a mountain that is sacred to Pueblo and Navajo peoples and that also supplied some of the canyon's timber. That is a road pointing at something, and what it points at is not a destination.
After about 1130 a long drought settled over the region, monumental construction stopped, and the centre of the Puebloan world moved elsewhere. It is worth separating the two endings that often get run together: great-house building and Chaco's regional primacy end around 1150, while occupation of the canyon itself tails off more slowly, toward about 1250. The descendant communities, from Hopi to Zuni to the Rio Grande pueblos, are still here, still hold Chaco as an ancestral place, and are still arguing with federal agencies over drilling near it. Nothing about this canyon was ever lost, which is worth saying plainly, because the word gets used about it constantly by people who have not asked anyone who lives there.
What our engines found
We ran the same test here as everywhere else, with the settings fixed in advance and identical at every one of the fifty sites we cover: a 200 km search radius, at least 12 km between sampled sites, 15 sites, a 2 per cent tolerance, and 2,000 Monte Carlo runs against a null model resampled from real catalogued sites in the same region. The settings are frozen because the answer moves if you tune them, and tuning them after seeing the answer is exactly the trick we object to elsewhere. Chaco Canyon returned 1 golden-ratio pairs where chance predicts 1.8, giving p = 0.7666.
That is fewer than chance would predict, and it is not significant. The control points the same way: the arbitrary ratio pi/2 scored 1.05 times its expected count against phi's 0.56, so nothing here singles out the golden ratio.
Worth flagging: the engine could reach only 6 catalogued sites here, so this result carries little weight in either direction. That is a limit on our data, not a verdict on the region.
We publish the misses because a tool that only reports its hits is not measuring anything. Every figure above is reproducible from the same coordinates, the same settings and the same seed.
One limit worth stating plainly, because it is the one most often glossed over. This test compares positions between monuments. It says nothing about any building's own proportions, its height, base, slope or perimeter. That is a separate question, and at several sites a genuinely open one.
Explore it yourself
Open the Chaco Canyon record in the living atlas to see the great houses in 3D among their neighbours and aim the sky engine at a Chacoan summer solstice. The Chaco Canyon site-emblem tee carries the sun dagger spiral in the Pyramason Chronicle style, with the site name and the exact coordinates.
Sources worth reading
Sofaer, Zinser and Sinclair, Science (1979), the first publication of the Fajada Butte light markings and the start of the Solstice Project's peer-reviewed corpus. Crown and Hurst, PNAS (2009), on the cacao residues. Crown and Wills, "The Complex History of Pueblo Bonito and Its Interpretation", Antiquity (2018), which is the source for reading Pueblo Bonito as an accretion of building episodes rather than a single design. Kennett et al., Nature Communications (2017), for the Room 33 genomes. Wills, Williams and Dorshow, "The Archaeology of the Bonito Paleochannel Cycle at Chaco Canyon, New Mexico", KIVA (2024), on water management. Guiterman et al. (2015) on the Chuska timber sourcing. Lekson (ed.), The Archaeology of Chaco Canyon (2006), and his The Chaco Meridian (1999, revised 2015), read alongside David Phillips's rebuttal. And the National Park Service's Chaco Project monographs, 1971 to 1986, which remain the foundation under all of it.