Did Pre-Clovis People Inhabit the Paisley Caves (and Why Does It Matter)?

The date and processes of initial human colonization of the Americas are crucial issues for the understanding of human biological and cultural development. For example, Soares et al. (2009) cited the American archaeological record to validate their proposed revision of the human mitochondrial molecular clock. Their suggested mutation rate puts the date of rapid expansion of Native American clades at around 13,500–15,000 cal yr BP. Similarly, Poznik et al. (2013) have used the “high-confidence archaeological dating” of the initial peopling of the Americas to calibrate the rates of both Y-chromosome and mitochondrial DNA (mtDNA) mutation and thereby to reconcile the ages of the common ancestors of human males and females. They use a date of ~15,000 cal yr BP, based on purported archaeological evidence from Monte Verde, Chile, dated to 14,600 cal yr BP.

Until recently, the consensus of archaeologists was that the Americas were populated after 13,400 cal yr BP (11,500 ¹⁴C yr BP) by a rapid expansion of people bearing the Clovis culture. Clovis ancestors, originating in southern Siberia, had crossed Beringia around 14,500 cal yr BP and subsequently migrated through a corridor between the receding Laurentide and Cordilleran ice sheets. This “Clovisfirst” model has been challenged continually by advocates for various allegedly pre-Clovis sites. As the evidence for a given set of ostensibly promising sites fails to withstand close critical scrutiny (e.g., Tule Springs, Pikimachay, Pendejo Cave, Pedra Furada, Shriver, etc.), pre-Clovis enthusiasts pin their hopes on the latest contenders. Currently, their favorites are Monte Verde in southern Chile (Dillehay 1997; Dillehay et al. 2008), the Debra L. Friedkin site in Texas (Waters et al. 2011), and the Paisley Caves in southern Oregon. The evidence from both Monte Verde and Paisley Caves has been deployed both by archaeologists (e.g., Beck and Jones 2010; Erlandson and Braje 2011) and by geneticists (e.g., Bodner et al. 2012) in support of a model of pre-Clovis migration along the Pacific coast—this despite the continuing absence of any sites older than 13,200 cal yr BP along the entire coast from Alaska to Tierra del Fuego (Dickinson 2011).

The Paisley Caves pre-Clovis evidence is unique in that it consists of both artifacts and directly dated ancient feces that have yielded human DNA. Gilbert et al. (2008) reported coprolites containing human (but also canid) mtDNA dating to ~14,300 cal yr BP. Poinar et al. (2009) noted chronological and stratigraphic problems and suggested that contamination by leaching might account for possibly exogenous human DNA in the coprolites, while Goldberg et al. (2009) (who had actually examined and sampled the coprolites) asserted that their morphology was inconsistent with human derivation.

Jenkins et al. (2012a, 2014) report results of meticulous excavations and analyses designed to verify their previous claims. This research focused particularly on demonstrating that the radiocarbon dates are accurate and that the coprolites have not been contaminated by leaching. Jenkins and colleagues now assert that diagnostic Western Stemmed Tradition (WST) lithic artifacts and associated human coprolites date between 11,070 and 11,340 ¹⁴C yr BP (~13,000–13,300 cal yr BP). The artifact-associated radiocarbon dates overlap with dates for Clovis elsewhere (11,550–10,800 ¹⁴C yr BP), while additional purported human coprolites are of pre-Clovis age (older than 11,600 ¹⁴C yr BP). This entire assemblage is interpreted as conclusive evidence that “the colonization of the Americas involved multiple technologically divergent, and possibly genetically divergent, founding groups” (Jenkins et al. 2012a: 223).

Close examination of the reported data shows that three stratigraphically and chronologically discrete components are present in the deep strata (Jenkins et al. 2014). The deepest deposit attributed to human activity consists entirely of coprolites, with no associated unambiguous lithic or organic artifacts. These purported “human” coprolites (dated to ~14,000–14,400 cal yr BP) are intermixed with others attributed, based on DNA, to late Pleistocene animals, that is, camelids and lion (Panthera...