Well now archeologist J.M. Adovasio has written a book entitled The First Americans and chauvinistically subtitled it In pursuit of Archaeology's Greatest Mystery. The greatest mystery is When and Who first populated the Americas. Mr. Adovasio is currently excavating the Meadowcroft Rockshelter site in southwestern Pennsylvania where, to the surprise of everyone, the lower levels recorded human occupancy radiocarbon dates of 16,000 years BP.

Before this dig, and two others in West Virginia and Virginia, the archaeologic smart money was bet on the dictum that Clovis Man, whose distinctive flints were found in North and South America and dated about 11,000 years ago, was the archetype American, the progenitors of the nine hundred language groups of American Indians that were here at the time of the arrival of Columbus.

Now what?

Archaeology, like most scientific pursuits, doesn't like to use the term "maybe". Both sides hurried to line up support from professionals in other fields to make their case.

The question: Are the descendants of Clovis Man the Indians of today?

Mr. Adovasio and those who believed in multiple and early peopling of the New World embraced the possibilities offered by nontraditional linguist Johanna Nichols of the University of California at Berkeley. Among other claims, Ms Nichols says that she can trace languages back beyond the cutoff point of 6,000 years. I quote from the book...

Nichols method has little to do with the traditional linguist's comparison of words and phonemes to find similarities and therefore relationships. Comparative methods have so far left linguists without their holy grail, a single tree from which all languages have branched over time. Instead they are confronted by some two or three hundred separate trees (what Nichols calls "stocks") that are hard to relate one to another. For example, the so-called Indo-European stock has 144 branch languages, all of which can be traced back about 6000 years beyond which is murk. Other stocks have fewer or more member languages; some, such as Korean, Basque, and maybe the language of the Zuni Indians, have only one. How these stocks are related remains enigmatic.

Ms. Nichol's strategy was to use the structure of the languages rather than word comparisons to map the language stocks. The structure of a language ( like verb placement-first, middle or last) tends to last through time, whereas the language built upon it comes and goes. On average, she finds, one new family of languages occurs in a stock every 4,000 years.


I'll explain why Adovasio and Nichol think that this method will help support the idea of man's early entry into the Americas (maybe 30,000 years BP) later, but my question is...

" What could cause a periodic 4,000 year spin off of a new language off an old language stock?"

(yes I know an average is but an average, but so do they. I think she's just plain wrong.)