Who's Really on First?

As the principal investigator of the Meadowcroft/Cross Creek Project since 1973 and as collaborator on the Monte Verde investigations for nearly a decade, I feel compelled to correct some of the misinformation presented in Anna Curtenius Roosevelt's recent review, "Who's on First?" (7/00-8/00), in which she questions the validity of virtually all sites that place humans in the Americas before about 12,000 years ago. Among other things, she confidently cites an article by Stuart Fiedel that questioned some of the artifact sources and documentation at Monte Verde. All of Fiedel's misapprehensions have been thoroughly refuted by Thomas D. Dillehay and his colleagues on their Web site (www.uky.edu/Projects /MonteVerde).

An informed visitor to Monte Verde and/or a careful reader of the final reports would know that the stratigraphy at that site is not discontinuous; that there are many objects of indisputable human manufacture, including such items as cordage and worked wood, some of which have been directly dated; and that the radiocarbon (carbon 14) dates cluster convincingly at ca. 12,750 B.P. (before the present). The dates are not inconsistent with the age of the associated projectile points, as documented at other South American sites, and there is no evidence that the site is contaminated by bitumen or other substances that would distort the radiocarbon dates.

Roosevelt also charges that Meadowcroft Rockshelter has questionable dates as a result of contamination. Previous research by four different radiocarbon laboratories, as well as a recent micromorphological study, conclusively found no evidence of particulate or dissolved contaminants or even of a possible source of such contamination (see the Journal of Field Archaeology 26, Fall 1999).

Contrary to its listing on Roosevelt's map, the date cited for Meadowcroft (19,100 [ or -] 810 B.P.) is not associated with the lanceolate point that was discovered there. This item and dozens of others at the site (notably prismatic blade flakes) fall within layers bounded by radiocarbon dates of 11,300 [ or -] 700 and 12,800 [ or -] 870 B.P. There are seven additional earlier dates, consistently ordered within the stratigraphy, all of which are directly associated with objects of indisputable human manufacture and some of which have a margin of error (standard deviation) of as little as 165 radiocarbon years. Even if the two earliest dates are discounted, the remaining five indicate that humans were present at this site sometime between 14,555 and 13,955 B.P.

The essential facts are that the Clovis hunting culture is no longer first and that the majority consensus in the profession is very different from the reviewer's.

J. M. Adovasio Director Mercyhurst Archaeological Institute Erie, Pennsylvania

ANNA CURTENIUS ROOSEVELT REPLIES:

Contrary to what J. M. Adovasio implies, by the time Monte Verde was visited by an independent team of archaeologists, the crucial early deposits no longer could be inspected. The complex, discontinuous stratigraphy is documented in the final, two-volume site report, which also explicitly records, as possible contaminants, numerous occurrences of bitumen and volcanic ash, as well as modern pollutants (detergents, pesticides, and other chemicals derived from geological carbon). The researchers avoided dating objects with likely contaminants. The possible cordage and food plants and the indisputable artifacts were not directly dated. And there is no clear connection between the two bipoints and their reputed dates, which are 2,000 to 7,000 years older than dates for similar bipoints elsewhere. My published comments about the dating problems at the site preceded Fiedel's critique. (Incidentally, the response to Fiedel--and Adovasio is one of the coauthors of the response--mentions bitumen.)

Regarding Meadowcroft, as a visitor to the site, I've observed that coal is present and thus--in the form of particles or solutes from hearth ash or groundwater--is a potential contaminant. The soil micromorphology study found weak evidence for groundwater influence but did not test for coal chemistry or contamination of samples. One test for contamination is to date solids and solutes separately; a study by Vance Haynes, which Adovasio does not acknowledge, found that Meadowcroft carbon solutes were earlier than the solids, so contamination was likely. As for the date of 19,100 [ or -] 810 h.p. listed on my map (which included just the earliest possible dates for each site), according to Adovasio's site reports, it is from the deepest occupational levels, in which biface fragments and possible blades were found.

While some scholars do accept Monte Verde and Meadowcroft, my point is that our reconstructions of the past should rely on sites that satisfy the standard criteria of validity.

CONTRIBUTORS

Biologist Craig Packer ("When Lions Ruled France," page 52), who teaches in the Department of Ecology, Evolution, and Behavior at the University of Minnesota, has been coordinating field studies of lions in six different reserves in South Africa for more than two decades. Since 1978 he has headed the Serengeti Lion Project, which began in 1966 and is one of the longest continuous field studies of animals. His book Into Africa (University of Chicago Press, 1994) includes an account of his work with lions. Jean Clottes, archaeologist and expert on prehistoric art, officially retired last year after a distinguished career as a conservator of France's decorated caves. He has served as chairman of the International Committee for Rock Art and now heads the team that is conducting scientific and conservation work at Chauvet Cave in southern France. Clottes is a contributing author of Dawn of Art: The Chauvet Cave (Abrams, 1996) and wrote Le Musee des Roches: L'art rupestre dans le monde (Seuil, 2000).