Cahokia: The Lost City on the Mississippi

Cahokia sits near present-day Collinsville, Illinois, adjacent to East St. Louis. Our three files agree, without qualification, that it was the largest pre-Columbian settlement north of Mexico: approximately 6 square miles, about 16 square kilometers, with a core ceremonial precinct surrounded by residential neighborhoods. Cahokia Mounds was designated a UNESCO World Heritage Site in 1982. Everything else in this article rests on that paragraph, and it is worth being clear about what kind of paragraph it is. The site is not a candidate location, not a rumor, and not a reading of an ambiguous text. It is a mapped, excavated, protected archaeological landscape. The arguing starts only once you ask what happened on it.

Population estimates at Cahokia's zenith, roughly 1050 to 1200 CE, run from 10,000 to 20,000 within the city proper, with some scholars arguing for up to about 40,000 once satellite communities within a roughly 50 km radius are counted in. That range is what makes the comparison everyone reaches for defensible: Cahokia was contemporary with medieval London and comparable to it in population. Checked independently during this article's research pass, London's own population circa 1100 CE comes out at roughly 10,000 to 15,000, with some estimates running 14,000 to 18,000. So the careful form of the claim is comparable to, not definitively larger than. Both figures carry real uncertainty, and stacking one uncertain number on another to declare a winner would be a weaker claim than the one the evidence already supports. Our second file adds a comparison that needs no uncertainty at all: Cahokia was the largest city in what is now the United States until Philadelphia surpassed it in population in the late 18th century, more than 500 years later.

Monks Mound is the largest pre-Columbian earthwork north of Mexico. It stands approximately 100 feet high, 30 to 30.5 meters, in four terraces, on a base of roughly 955 feet by 775 feet, 291 by 236 meters, covering about 14 to 15 acres, 5.7 to 6.1 hectares. It was built in stages between approximately 900 and 1200 CE, and the earth moved to raise it is estimated at 22 million cubic feet, roughly 622,000 to 623,000 cubic meters, carried entirely by hand in woven baskets, without draft animals and without wheeled vehicles. Now the correction, stated where a reader can watch it happen. Our primary file for this article gives the base as over 950 feet long by 650 feet wide. The 650 foot width does not hold. Both of our secondary files independently give the base as 291 by 236 meters, and 236 meters is approximately 775 feet, not 650, and independent checking during this article's research pass converges on 955 by 775 feet as the standardized 1988 measurement. This article uses 775 feet. The primary file's own height and volume figures for the mound are compatible with either width, since the mound's real footprint is irregular rather than a clean rectangle; the correction rests on the two secondary files' independent agreement plus the standardized 1988 survey, not on the primary file contradicting itself elsewhere, which is what makes the 650 look like a single transcription slip rather than a rival survey.

Soil analyses of Monks Mound show that different clay types were deliberately layered to manage drainage and structural stability, which contradicts an earlier assumption that the mound was simple, undifferentiated earthen piling. This article is not a construction study and will not become one here, because what that finding contributes to a story about a lost city is not really about technique. Somebody understood the material well enough to sequence it. Somebody organized the movement of 22 million cubic feet of soil, one basket at a time, across roughly three centuries of staged work. That is a population fact, an authority fact, and a duration fact, and it is the first hard corroboration that the numbers in section 01 are not fantasy.

South of Monks Mound lies the Grand Plaza, a leveled public space covering approximately 19 hectares, roughly 47 acres, which our file records as one of the largest public spaces in pre-Columbian North America. It is not a natural clearing. Geophysical surveys and excavations led by Rinita Dalan, published in 2003, showed that the plaza was deliberately engineered, with layers of colored clay fill leveling the original terrain. The mounds are the labor everyone sees. The flat ground beside them is the labor nobody looks at, and our file reads it as evidence of massive coordinated work beyond the mounds themselves.

West of Monks Mound, Warren Wittry identified a series of large post pits during excavations in the 1960s, dated to 1961 in one of our files, and the pits formed timber circles. They were named Woodhenge, by analogy with Stonehenge. At least five successive circles, Woodhenge I to V, were built between approximately 900 and 1100 CE, using red cedar posts arranged to mark solstice and equinox sunrise alignments as observed from a central observation post. The largest of the reconstructed circles contained 48 posts at a diameter of approximately 128 meters. One circle was physically reconstructed in 1985 using the original excavated postholes, and it hosts public solstice and equinox sunrise observation events to this day. Note where the tiers fall here, because it is easy to get backwards. The solar alignments at Woodhenge sit at Tier 1 in our files. The stellar alignments proposed elsewhere on the site sit at Tier 3, and section 09 keeps them there.

A city that size on that floodplain needed a crop, and the timing of the crop is specific. Mississippian societies depended on intensive maize agriculture. Maize reached the eastern woodlands from Mesoamerica by way of the Southwest by approximately 200 CE, but it was not intensively cultivated until cold-adapted, short-growing-season Northern Flint varieties emerged after 800 CE. Stable isotope analysis, delta-13C, of Mississippian burials shows maize supplying 40 to 65 percent of dietary calories, a figure measured in the bones of the people who ate it. Gayle Fritz documented in 2019 what came before it: the Eastern Agricultural Complex, the cultivation of goosefoot, sunflower, sumpweed and squash, native plants grown long before maize was farmed intensively, gradually supplanted by it and continuing alongside it as supplementary agriculture. Read the two dates together and the sequence is hard to miss. The plant is present from around 200 CE and does very little. A cold-tolerant variety appears after 800 CE, and within roughly two and a half centuries the largest settlement north of Mexico is going up on the bottomland.

Around 1050 CE, Cahokia went through a rapid transformation that Timothy Pauketat has named the city's Big Bang. Population surged, monumental construction accelerated, and the site plan was reorganized around a new cosmological template, all inside roughly a generation. Pauketat has further proposed that this may be linked to a supernova observation, SN 1054, the event that produced the Crab Nebula. Our file is explicit about the standing of that second step: the chronological correlation is suggestive rather than demonstrated causation. That is the right way to hold it. A sky that changes and a city that reorganizes inside the same decade is a striking coincidence and an untested hypothesis at once, and nothing about the first claim, the transformation itself, depends on the second one being true.

Cahokia's trade connections span over 1,500 km, and they are demonstrable because the materials themselves are traceable. Marine shell from the Gulf of Mexico, copper from the Great Lakes, particularly Isle Royale and the Keweenaw Peninsula, galena from Missouri, and chert from various Midwest sources have all been recovered at the site. The flow ran outward as well: John Kelly and colleagues documented the wide distribution of Ramey Incised pottery across the Mississippian world as further evidence of the same exchange network. A settlement that imports Gulf shell and Great Lakes copper while its own pottery style turns up elsewhere is not a place having a local moment. It is a hub.

Chemical residue analysis of Cahokian beaker-style ceramics, published by Crown, Emerson, Gu, Hurst, Pauketat and Ward in the Proceedings of the National Academy of Sciences in 2012, confirmed the presence of caffeine from Ilex vomitoria, yaupon holly, the source of the ceremonial Black Drink. That beverage, a purification and stimulant drink, was used in rituals across the Southeast. Yaupon holly does not grow at Cahokia; the site lies far north of the plant's natural coastal range, so the residue demonstrates a long-distance trade network and a shared ritual practice reaching well outside the plant's own growing region at the same time. Our file places this at Tier 2, and the reason is worth naming rather than leaving implied: the detection chemistry itself is solid, and the tier reflects the ritual-interpretation layer built on top of it. The caffeine is not in doubt. What people did with it is read from context. This is also one of the four identifiers in this cluster of files that were checked and found correct, and it is linked at the foot of this article.

Marine shell gorgets, copper repousse plates, and engraved ceramics found across Mississippian sites share a recognizable iconographic vocabulary: the Birdman, a falcon warrior; the bi-lobed arrow; the forked-eye motif; the cross-in-circle; the chunkey player. This body of imagery was once called the Southern Cult. It is now understood as the Southeastern Ceremonial Complex, a widespread religious and symbolic system rather than a single organized religion, and that renaming is a change of claim rather than a change of fashion. F. Kent Reilly III and James Garber reinterpreted the complex in 2007 as an integrated cosmological system expressing a three-part universe: an Upper World, This World, and a Lower World. It flourished from approximately 1200 to 1450 CE, with major centers producing distinctive local styles while sharing the core iconographic elements.

Chunkey, or chungke, was a stone disc game, and our files carry it as far more than recreation. Ethnographic and archaeological evidence, including thousands of chunkey stones documented by Thomas Emerson at Cahokia and at Mississippian-influenced settlements, suggests it functioned as a mechanism for diplomacy, alliance-building, redistribution of wealth, and cosmological enactment. Chunkey stones have been found hundreds of miles from Cahokia, which suggests the game spread along with Cahokian influence. Hold the tier honestly here. That the stones exist in that quantity and travel that far is not the debatable part. What the game did socially is an interpretation of them, and Tier 2 is where our file puts that interpretation.

The broader Mississippian cultural tradition, roughly 800 to 1600 CE, is defined by flat-topped earthen platform mounds supporting timber structures: temples, chiefs' residences, charnel houses. It spread across the southeastern and midwestern United States, and major sites beyond Cahokia include Moundville in Alabama, Etowah in Georgia, Spiro in Oklahoma, Angel Mounds in Indiana, and Winterville in Mississippi. Stratigraphic analysis confirms repeated cycles of mound construction, use, deconstruction, and rebuilding, which suggests the mounds embodied cosmological renewal cycles rather than being built once and left static. That is worth pausing on, because it changes what a mound is. In this tradition a mound is not a finished monument. It is a practice, repeated, and the repetition is legible in the layers.

Mound 72 is a small ridgetop mound about 850 meters south of Monks Mound, excavated by Melvin Fowler and colleagues from the University of Wisconsin-Milwaukee, with fieldwork running from 1967 to 1971 according to independent checking during this article's research pass. It contained over 270 burials. Among them was a central high-status figure interred on a bed of approximately 20,000 marine shell disc beads arranged in the shape of a raptor, and a nearby mass burial including over 50 young women who appear to have been sacrificed. One of our files gives a more granular count for a specific mass grave inside the mound: 53 sacrificed women and 4 headless and handless men. Other burials held caches of finely worked arrowheads, some with material or stylistic ties to Oklahoma, Tennessee and Wisconsin, along with mica and copper artifacts. Those are the excavated facts, and all three of our files report them consistently at Tier 1. What they mean about power at Cahokia is the next claim, and it sits one tier down, on purpose.

Whether Cahokia constituted a true political state or a complex chiefdom is still argued. Timothy Pauketat argues for centralized political authority approaching statehood, citing the rapid, planned construction of the city center, craft specialization, and the Mound 72 mass burial as evidence of coercive power. George Milner, of Pennsylvania State University, counters that Cahokia lacks the hallmarks of statehood: no writing system, no standing army, no bureaucratic administration, no evidence of taxation. He classifies it as a complex chiefdom, and he separately argues that Cahokia's population has been inflated, proposing a lower peak of 5,000 to 8,000 residents based on residential-structure density analysis, well below the 10,000 to 20,000 and higher figures most other cited scholars use. Susan Alt has argued something different again, that Cahokia's influence over outlying communities was negotiated and less coercive than Pauketat's centralized model implies. Our files record the debate as genuinely unresolved, and note that it turns partly on a definitional question about what counts as a state at all. This article does not settle it. A file that handed you a winner here would be telling you something the evidence has not established, and the honest position is that a bed of 20,000 shell beads and 53 dead women in one grave is compatible with more than one political arrangement.

By some point in the 1300s, Cahokia was largely abandoned. The proposed causes are several, and our files do not rank them: environmental degradation, including the deforestation and stream sedimentation documented by William Woods and George Holley; the onset of the Little Ice Age; political instability; warfare, with a palisade wall enclosing roughly 120 hectares of the central precinct constructed around 1175 CE; and social disintegration. The best-sourced claim in the whole decline debate is a 2019 study. A.J. White and colleagues, publishing in the Proceedings of the National Academy of Sciences, used fecal stanol analysis of Horseshoe Lake sediments to track population fluctuations, and found that depopulation was gradual rather than sudden, and that it began before a major Mississippi River flood in 1200 CE. That points toward internal political causes contributing ahead of, rather than solely in reaction to, a single environmental catastrophe. No single cause has been definitively established. Our files' own conclusion is that the decline was almost certainly multi-causal, and this article's conclusion is the same.

The first European eyewitness descriptions of functioning Mississippian polities come from Hernando de Soto's expedition, which traversed the southeastern United States from 1539 to 1543, well after Cahokia itself had emptied. Measured against the abandonment dates in section 07, de Soto's party arrives roughly 140 to 190 years too late to have seen the city working. What the chronicles record instead is the wider tradition still running. Garcilaso de la Vega, writing in 1605, Rodrigo Rangel in the 1540s, and the anonymous Gentleman of Elvas in 1557 describe densely populated town centers, paramount chiefs carried on litters, harvest-storage granaries, and large-scale warfare. Charles Hudson reconstructed the expedition's route in 1997, identifying contact with the chiefdoms of Cofitachequi, Coosa and Tascaluza, among others. By the time of the next sustained European contact, French and English in the late 17th century, most Mississippian polities had collapsed, likely due to epidemic disease introduced by de Soto's party. So the tradition Cahokia belonged to outlived the city itself, and when it ended, it ended under a cause that appears nowhere in the list proposed for Cahokia's own decline.

One more thing belongs in a file about a lost city, because it is the part the phrase lost city quietly gets wrong. Many southeastern tribal nations, including the Chickasaw, Choctaw, Muscogee (Creek), Cherokee and Osage, have ancestral connections to Mississippian sites, and were forcibly removed from the region during the 1830s Trail of Tears. The Native American Graves Protection and Repatriation Act, NAGPRA, passed in 1990, has prompted repatriation of some Mississippian-era human remains and sacred objects, and our file records that tension persists between archaeological research interests and tribal sovereignty over ancestral cultural property. That tension is documented in the corpus file's own counter-arguments section, which is where a research file is supposed to keep the thing that complicates it.

Certain Cahokian features, platform mounds, human sacrifice, a planned city layout around a central plaza, and the possible presence of cacao, have led some researchers to propose direct Mesoamerican influence. One strand of that is a real residue finding: Patricia Crown and colleagues reported chocolate, cacao, Theobroma cacao, residues in Cahokian pottery in 2012, and cacao does not grow north of the Gulf Coast, so its presence supports a trade connection reaching that far. Cahokia flint-clay figurines also show stylistic parallels with Mesoamerican traditions. What is not established is the step from contact to influence. No unambiguous Mesoamerican artifacts have been found at Cahokia itself, and the transmission pathway, direct contact against indirect diffusion through intermediaries, remains undemonstrated. John Kelly, in one of our files, considers indirect diffusion possible but unproven. All three of our files place this at Tier 3, which is a piece of unanimity in a cluster that disagrees about a good deal else.

Some researchers propose that Cahokia's layout was a deliberate map of the cosmos: an Upper World in Monks Mound, a Middle World in the Grand Plaza, and a Lower World in the Mound 72 burials, reflecting widespread Mississippian cosmology. It is an elegant reading, and it is not baseless, since a three-part universe is precisely what Reilly and Garber read out of the Southeastern Ceremonial Complex in section 05. Our file's own limit on it is the honest one: direct evidence of intentional cosmographic planning at the full city scale is limited. That is where this article leaves it.

Several Indigenous oral traditions from the Southeast and Midwest contain references to a great city, a powerful ancient alliance, or a catastrophic dispersal event, and some researchers propose that these may encode memories of Cahokia's rise and fall. The Dhegihan Siouan migration narratives and certain Osage traditions are the named candidates in our file. A direct, confirmed connection to Cahokia specifically is unestablished. Tier 3 here means proposed and unconfirmed, which is not the same as dismissed, and the distinction matters more than usual in a case where the alternative source of testimony, a written record, does not exist at all.

Researchers have proposed that Monks Mound and other Cahokian structures align with significant stellar risings, particularly stars of the constellation Orion and the Milky Way, beyond the well-documented Woodhenge solar alignments. Our file is blunt about the state of that work: these claims have not been rigorously tested with the same precision applied to Woodhenge, and they remain unverified in peer-reviewed literature. This is the place to read the difference between a Tier 1 alignment and a Tier 3 alignment on one site carefully. Woodhenge's solar alignments were excavated, reconstructed from the original postholes, and are observed publicly twice a year. The stellar claims are thinly studied. Same site, same sky, very different evidence.

No. The mounds were not built by a lost, culturally advanced non-Indigenous race. That claim, the 19th-century Mound Builder myth, variously identified those builders as Israelites, Vikings, Welsh or Atlanteans, supposedly displaced or exterminated by the ancestors of contemporary Native peoples. Cyrus Thomas of the Smithsonian's Bureau of Ethnology definitively refuted it in his 1894 report, the Bureau's 12th Annual Report, demonstrating through systematic excavation that the mounds were built by the ancestors of historically known Indigenous peoples. And the myth's function is part of the record rather than a modern gloss on it: our files state that it served a specific political function, justifying Indian removal policy by denying Indigenous peoples credit for their own monumental heritage. All three of our files carry this claim and all three place it at Tier 4, which our research pass records as the strongest possible cross-file agreement on a debunking in this cluster.

No. Claims that Cahokia's scale or alignment required extraterrestrial assistance have no evidentiary basis. The engineering is genuinely impressive at scale and it is fully explicable through organized human labor, as experimental archaeology projects have demonstrated. It is worth setting the two Tier 4 claims side by side, because they are the same move a century apart. Both start from a monument, both conclude that the people whose ancestors built it could not have, and both then supply a more acceptable builder. The 1894 answer to the first one works on the second one too: excavate, and the builders are in the ground.