The pre-Clovis archaeological record
Sites purportedly pre-dating Clovis are received skeptically and subject to considerable, perhaps inordinate, validation criteria. Numerous pre-Clovis sites have emerged as contenders but few Paleoindian archaeologists agree as to which sites and artifacts they accept as definitive evidence of a pre-Clovis colonization event.
Currently, a human coprolite from the Paisley Cave site (Map 19.ι) is widely accepted as the earliest evidence of human occupation. Dated to approximately 14,000 ybp it is not associated with diagnostic artifacts, so it could be representative of Clovis or an earlier population of migrants or non-colonizing explorers. Since Clovis was identified and defined, claims of pre-Clovis aged sites became ubiquitous throughout North and South America. Often lacking chronological control, found in disturbed contexts, and containing equivocal artifacts most of these sites have struggled to gain widespread professional acceptance. With the reporting of excavations at the Monte Verde site in Chile in the 1990s, claims for pre-Clovis occupation gained considerable support if not outright acceptance by many archaeologists.[629] Cultural deposits dated to between 14,650 and 15,600 ybp remain controversial, but Monte Verde, located over 1,500 kilometers from Beringia, provided the first pre-Clovis aged site with a distinctly non-Clovis artifact assemblage to challenge the Clovis-first model.Sites in North America such as Meadowcroft, Buttermilk Creek, and Topper all contain potentially pre-Clovis aged artifacts (Map 19.1).[630] Meadowcroft, a multicomponent rockshelter in Pennsylvania, is a long-standing contender for evidence of a pre-Clovis occupation. Artifacts in its lowest levels pre-date 14,000 ybp and if verified provide a second instance of early cave use. The Buttermilk Creek site in Texas (Map 19.ι) contains a lithic assemblage below diagnostic Clovis artifacts.
The pre-Clovis level contains a mix of blade, flake, and bifacial tools from deposits potentially dated to 15,000 ybp. The Topper site in Georgia produced an assemblage of over 1,000 small flakes and microblades. Like Buttermilk Creek, the early material was found below a Clovis occupation level and may exceed 15,000 ybp. Concerns about stratigraphic disturbance and inaccurate dates have plagued interpretation of these sites.In addition to Monte Verde, possible South American pre-Clovis aged sites include Taima-Taima in Venezuela and Pedra Furada in Brazil. Both of these sites have lithic artifacts derived from deposits dated in excess of 15,000 ybp, but like North American sites have been accused of lacking stratigraphic integrity resulting in erroneous dates and artifact identifications. The Toca da Tira Peia in Brazil has produced a small assemblage of quartzite cobble and flake tools that reportedly date to 20,000 ybp.[631] Unfortunately, or tellingly, pre-Clovis deposits often lack material suitable for radiocarbon dating and researchers rely instead on optically stimulated luminescence dates, a less preferred dating method for Late Pleistocene aged materials. Which, if any, of these sites provide indisputable evidence of a distinct pre-Clovis population of colonizers remains unresolved.[632] Many archaeologists accept one or more of these sites as evidence of a pre-Clovis presence in the Americas but it is unclear at what point in the colonization process human populations obtained a sufficient size or duration of occupation to attain archaeological visibility.