Many claimed pre-Clovis sites had genuine problems with their evidence — contaminated radiocarbon dates, ambiguous stone tool assemblages that might be naturally produced rather than human-made, and stratigraphic issues that made it difficult to establish the age of the human presence claimed.

The field had been burned too many times by spectacular claims that collapsed under scrutiny, and the appropriate response to early claims was rigorous methodological scrutiny.

But the Clovis first model itself eventually came under pressure from a growing body of evidence that was increasingly difficult to dismiss.

Sites from multiple locations — Monte Verde in Chile, Meadowcroft Rockshelter in Pennsylvania, Buttermilk Creek in Texas, among others — produced evidence of human presence that predated the Clovis horizon by varying amounts, and the methods used to document that evidence were increasingly robust.

The most geographically striking of these pre-Clovis sites is Monte Verde in southern Chile, where evidence of human presence dating to approximately 14,500 to 15,000 years ago has been documented.