Who discovered the Americas?

Four questions that make more sense than the one usually asked

The question in the title is badly formed, and the way it is badly formed has shaped four centuries of argument. A continent inhabited by tens of millions of people cannot be discovered in any ordinary sense of the word, and the phrase survives only because it was written by Europeans describing their own arrival. What can sensibly be asked is a set of narrower questions: when human beings first entered the Americas, by what route, how often, and what happened when populations that had been separated for millennia met again.

For most of the twentieth century the answer to the first question was fixed and confident. Distinctive fluted stone points, first identified near a town in New Mexico and dated to around thirteen thousand years ago, were found across North America, and the accompanying account held that a single population had crossed the exposed land connecting Siberia and Alaska, passed south through a corridor between two ice sheets as they retreated, and spread rapidly across two continents. The model was coherent, it fitted the evidence then available, and it hardened into orthodoxy to a degree that made contrary claims professionally dangerous. Sites purporting to be older were examined with a severity not applied to sites that agreed with the consensus, and several careers were damaged.

The consensus broke on a site in southern Chile, excavated over many years and eventually accepted, after an inspection by a delegation of prominent sceptics, as older than the supposed first arrivals in North America by a considerable margin. The implication was inescapable: if people were living fourteen and a half thousand years ago at the southern end of South America, they cannot have arrived through an ice-free corridor that geological evidence suggests was not passable until later, and they cannot have come from a population that had only just entered Alaska.

The reaction to that acceptance was instructive. Once the orthodoxy gave way, sites that had been dismissed for decades were re-examined, and several proved to have been rejected on grounds that would not have been applied to a conventional claim. The episode is cited in discussions of scientific conservatism as an example of a consensus doing useful work — filtering out weak claims — for a period, and then continuing to filter after the evidence had changed. The route now generally favoured runs along the coast. A shoreline supports a predictable diet of fish, shellfish, seabirds and marine mammals, requires no adaptation to unfamiliar inland environments, and can be travelled by boat at a speed that makes rapid colonisation plausible. The difficulty with the hypothesis is that the coastline of that period is now under a hundred metres of water, which means the evidence that would confirm it is largely submerged. Underwater survey has begun to recover material, and the pattern of finds is consistent with the idea, but the absence of a decisive site is a genuine weakness rather than a technicality.

Claims of much greater antiquity continue to appear and are treated, reasonably, with caution. Human footprints preserved in the sediments of a dry lake bed in the American Southwest have been dated to more than twenty thousand years ago, which would place people in the interior of the continent at the height of the last glaciation, before any corridor or coastal route is thought to have been available. The dating was originally based on seeds of an aquatic plant, which can absorb old carbon from water and therefore appear older than they are; subsequent work using different materials has supported the early dates. The site is not yet settled and is being argued about in exactly the way such a claim should be.

A quite separate arrival is documented beyond dispute. Norse settlers from Greenland established a small base on the northern tip of Newfoundland, and the remains — turf buildings, iron rivets, a smithy — were excavated in the 1960s. Wood from the site has since been dated with unusual precision by identifying, in the tree rings, the signature of a known burst of cosmic radiation, which fixes one year of felling at 1021. The settlement lasted perhaps a decade and left no lasting consequence, which is precisely what makes it useful: contact without transformation demonstrates that the transformation which followed later was not inevitable.

What made the fifteenth-century arrival different was not seamanship but epidemiology and continuity. The populations of the Americas had been separated from the Eurasian disease pool for millennia and had no exposure to smallpox, measles or influenza; mortality following contact was catastrophic and preceded, in many regions, the arrival of any European. Estimates of the decline vary widely and the methods used to produce them are contested, but no serious account puts the losses at less than a very large fraction of the population, and in several regions the societies that Europeans described had already been reshaped by an epidemic that arrived before they did. The Norse came, traded, fought and left. The later arrivals kept coming, and brought with them microorganisms that did more to determine the outcome than any weapon. The question of who arrived first, in the end, matters much less than the question of who arrived repeatedly.