Arrival routes of first Americans uncertain

See allHide authors and affiliations

Science  16 Mar 2018:
Vol. 359, Issue 6381, pp. 1224-1225
DOI: 10.1126/science.aar8233

In this watercolor, Clovis Paleoindians journey eastward toward the Americas.


In their Perspective “Finding the first Americans” (3 November 2017, p. 592), T. J. Braje et al. argue that people first entered the Americas about 25,000 to 15,000 years ago by way of the Pacific coast. We believe that current evidence yields far less certainty than Braje et al. suggest—and more likely supports a later arrival by inland and/or coastal routes.

Contrary to Braje et al., genetic evidence indicates that Native American ancestors diverged from Siberian populations ∼24,900 to ∼18,400 years ago, followed by population expansion into the Americas ∼16,000 to ∼13,500 years ago (1). Only two Native American lineages have been identified south of Beringia: a northern lineage constrained to northern North America, and a southern lineage directly linked with Clovis—the earliest unequivocal widespread cultural manifestation south of the ice sheets (2). Thus, a Native American lineage in the Americas between 25,000 and 15,000 years ago is inconsistent with current data. There is no consensus on the validity of purported pre–16,000-year-old sites, which vary in accurate dating, unambiguous artifacts, and clear association between them (3). Moreover, there are few technological connections among 16,000- to 13,500-year-old pre-Clovis sites, or with later Paleoindian artifacts; thus, the relationships between these sites and later Native Americans remain ambiguous.

The coastal colonization route Braje et al. advocate remains a viable hypothesis for a later arrival date, but several issues should be addressed (4). Despite the rise of sea levels in the Holocene, much of the late Pleistocene coast from Puget Sound to Alaska remains above sea level (5), yet surveys have failed to discover coastal sites securely dated older than 12,500 years ago (4). Furthermore, widespread empirical patterns are inconsistent with the coastal hypothesis: All known populations in Siberia, Russian Far East, and Beringia had terrestrial-oriented economies and technologies, as did widespread Paleoindian groups (6, 7), with limited evidence of coastal exploitation in lower latitudes (8). This empirical patterning suggests that they were more likely to follow a land route through Siberia, Beringia, and the Americas south of the ice sheets. Current studies indicate that deglaciation began 19,000 years ago and that an ice-free corridor, largely vegetated and free of proglacial lakes, existed by at least 15,000 to 14,000 years ago (9, 10).

Braje et al. suggest that stemmed projectile points in different contexts provide evidence for a coastal expansion before 16,000 years ago, but this is not a consensus view. Stemming is a widespread form of weapon design that was innovated numerous times and thus cannot be used to argue for cultural affiliation. No technological analysis has established a valid connection between these disparate assemblages, and there remains debate on the dating of North American stemmed points (11) because most securely dated sites are younger than Clovis.

Braje et al. assert that there is near complete agreement among archaeologists on these issues. However, the most recent survey (12) showed that archaeologists are divided, with many thinking that both interior and coastal routes were used, and expressing skepticism about several proposed pre-Clovis sites. Genetic and archaeological data suggest expansion from Siberia into the Americas around 16,000 to 13,500 years ago, consistent with terrestrial and/or coastal migrations. This evidence base explains the absence of consensus among scientists regarding both routes and timing of the peopling of the Americas.


Stay Connected to Science

Navigate This Article