Demographic Collapse

Class: RESEARCH. Compiled documented record. Pre-contact populations cannot be counted from censuses that do not exist; every figure here is a scholarly estimate, labelled with its author, year, and range. Never present any single figure below as a settled count.

1. The range of estimates for the Americas, 1492

Published hemispheric estimates have ranged from about 8 million to 112 million. That spread is the honest starting point.

Denevan's 1992 regional breakdown of the 53.9 million: Mexico 17.2m; Andes 15.7m; lowland South America 8.6m; Central America 5.6m; United States and Canada 3.8m; Caribbean 3.0m. Denevan's Caribbean figure is now challenged by genetic evidence for a much smaller Caribbean population (see caribbean-and-conquest.md) — an example of a consensus figure being revised by new evidence, recorded as such.

For North America north of Mexico alone, estimates likewise diverge: James Mooney (1928) ~1.15 million; Henry Dobyns (1983) up to 18 million; Thornton (1987) more than 5 million in the coterminous US plus ~2 million in Canada/Alaska/Greenland; Stannard (1992) 8–12 million north of Mexico. No consensus exists for this region; Thornton's is the most methodically derived of the widely cited figures.

2. Koch et al. 2019 — the Earth-system study

Alexander Koch et al., "Earth system impacts of the European arrival and Great Dying in the Americas after 1492," Quaternary Science Reviews 207 (2019): 13–36. DOI 10.1016/j.quascirev.2018.12.004.

The study's findings, stated as the study's findings:

• From 119 published regional population estimates, a pre-1492 population of 60.5 million (IQR 44.8–78.2m), using ~1.04 ha of land use per capita.
• "European epidemics removed 90% (IQR 87–92%) of the indigenous population over the next century" — a decline to roughly 5–6 million by ~1600.
• Abandonment of ~55.8 million hectares of anthropogenic land was followed by secondary vegetation succession sequestering ~7.4 Pg carbon, equivalent to ~3.5 ppm of atmospheric CO₂; with carbon-cycle feedbacks, a total ~5 ppm uptake — 47–67% of the observed 7–10 ppm decline in atmospheric CO₂ in the late 1500s/early 1600s, contributing ~0.15°C of cooling associated with the Little Ice Age.
• Conclusion in the authors' words: "The Great Dying of the Indigenous Peoples of the Americas resulted in a human-driven global impact on the Earth System in the two centuries prior to the Industrial Revolution."

Caution: Koch et al. synthesise other scholars' regional estimates; the 60.5m and 90% figures inherit those estimates' uncertainties (hence the wide IQRs, which must be quoted with the headline figures). Later work has disputed parts of the CO₂ attribution. The demographic order of magnitude — tens of millions before, a decline of roughly nine-tenths within a century — is consistent with Denevan and Thornton; the precise percentages are the study's model, not a count.

3. Regional figures with sources

• Mexico: Denevan (1992): 17.2m in 1492. Classic higher estimates for central Mexico (Cook & Borah) run to ~25m; the population of present-day Mexico fell to roughly 1 million by the early 1600s in the standard demographic series — a decline driven principally by successive epidemics (1520 smallpox; 1545 and 1576 cocoliztli), compounded by war, enslavement, famine, and social disruption.
• Andes: Denevan (1992): 15.7m. Inca imperial population estimates commonly range ~10–14m on the eve of Pizarro's invasion (1532), after a civil war and an epidemic (likely smallpox, killing the emperor Huayna Capac ~1527) that preceded the Spanish arrival.
• Present-day United States: Thornton (1987): >5m in 1492 → ~600,000 by 1800 → ~250,000 by the last decade of the 19th century, "some 4 to 5 percent of its former size." The US Census-trough figure of ~250,000 around 1890–1900 is documentary, not estimated.
• Canada: scholarly estimates for the territory of present-day Canada at contact range from ~200,000 to 2 million; 500,000 was the figure accepted by Canada's Royal Commission on Aboriginal Peoples / Royal Commission on Aboriginal Health in the literature summarised in the population histories. The 1871 enumeration counted 102,358 Indigenous people within Canada's then borders — again, a documented trough far below any contact-era estimate.
• Caribbean: see caribbean-and-conquest.md — the widest proportional dispute of any region, from a few tens of thousands (2020 genetic study) to Las Casas's 3 million-plus (primary witness, doubted).

4. Virgin-soil epidemics — the largest documented killer

• The term is Alfred W. Crosby's: "Virgin soil epidemics are those in which the populations at risk have had no previous contact with the diseases that strike them and are therefore immunologically almost defenseless" (Crosby, "Virgin Soil Epidemics as a Factor in the Aboriginal Depopulation in America," William and Mary Quarterly, 1976).
• The diseases were principally smallpox, measles, influenza, typhus, and later cholera and others, arriving in repeated waves over three centuries — not a single event. Epidemic mortality is, on the scholarly consensus, the largest single cause of the collapse in numbers, ahead of direct killing. Thornton (1987) and the demographic literature are explicit on this, as are critics of Stannard.
• The distinction this record must keep: epidemic mortality is not automatically policy, and policy is not reducible to epidemics. Three documented qualifications prevent the "disease did it, no one is responsible" simplification, and each is evidenced elsewhere in this record:
• Colonial conditions magnified epidemic death: forced concentration in missions, reducciones, reservations and schools; enslavement and mine/plantation labour; malnutrition from dispossession and destroyed food systems; and flight and warfare all raised mortality and spread infection. Lemkin made exactly this point about settlement concentration (see framing file).
• Deliberate infection is documented in at least one case at the level of correspondence and a trader's journal (Fort Pitt, 1763 — see disease-and-intent.md), while other famous deliberate-infection claims fail at evidence level and are labelled there as unsupported.
• Direct killing, removal, enslavement, child transfer, and birth prevention are independently documented in the other files of this record, in numbers that do not depend on any epidemic estimate.

5. What cannot be held to standard

• Any claim of an exact pre-contact population ("there were X million") — no source supports an exact figure; only ranges with named authors.
• Stannard's ~100 million presented as a documented death toll rather than his contested estimate derived from a high pre-contact figure.
• Attributing the entire ~90% decline to deliberate killing, or, conversely, attributing the entire decline to unintentional disease and none of it to documented policy — both exceed what the sources carry.

Sources for this file

Denevan (1976; 1992) • Thornton (1987) • Koch, Brierley, Maslin & Lewis, Quaternary Science Reviews 207 (2019) • Crosby (1976) • Cook & Borah (central Mexico series, as summarised in the standard population histories) • Reich et al., Nature (2020), Caribbean genomics (see Caribbean file) • Population history syntheses citing Mooney (1928), Dobyns (1983), Stannard (1992).