Thursday, February 23, 2012

Dating Methods: The Woodlands

            Scholars agree that, for most woodland regions, corn did not play a vital part in Native cultures for most of their history.  Instead, they used locally domesticated crops to help sustain their partially agricultural societies.  The date when corn became a significant crop has been variably dated, but often a date of 900 AD is given to the middle and northern woodland regions.  The date that it was initially introduced is even further shrowded in shadows, because of the limited evidence, and even less dated evidence for corn, either in seed or pollen form.  This is obviously a significant problem when trying to come up with a relatively precise number.
            Using previously published information, Crawford et al. wanted to compile well dated evidence to try to trace the spread of corn into woodland regions, and give a less equivocable date for the first introduction of corn into woodland cultures.  A significant portion of this essay is a table, and then the explanations for the entries on the tables.  By doing this, the authors hoped to minimize any counter-arguments based in lack of data.  The data that the team puts most faith in are those dated using AMS (accelerated mass spectrometry).  They concluded that corn may have been around in the Lower Great Lakes Region as early as 200 AD.
            Although their results apply to a very specific region, the region for which they had the most data, the ideas they used to show the earliest confirmation can be applied for other geographical regions of the woodlands, to continue the determination for the entire woodlands.  By doing this, it may be possible if enough evidence can be gathered, to create a virtual pathway by which corn spread into the woodlands.  Perhaps by examining such pathways, one would be able to also look at the spread of cultures tied with trade and/or migration.

            As Shott states in the article, “The signifance of radiocarbon dating is almost impossible to exaggerate (…) but it has often failed to meet the high expectations we developed for it.” (202)  Shott uses the Childers Site as one such example to show where it has failed in dating the woodlands.  He begins the essay by discussing several inherent problems in using radiocarbon dating, including such familiars as inconsistent C-14 ratios, the ‘old wood’ problem, and the problem of conflicting dates from alternate dating methods, or from different locations on the same site.  The final problem in the previous list is the major one that shows up at the Childers site, and is discussed in depth in the essay.
            Radiocarbon analysis was carried out on a large number of samples found near the Childers site, by multiple radiocarbon labs.  Although the arrangement of the artifacts, as well as the faunal, ceramic, lithic, and botanical artifacts found at the site all suggest a single occupation in the Late Woodland period of between 100 and 150 years, the radiocarbon evidence would seem to suggest multiple occupations of the site spanning between 1650-1050, as well as 950-750 BP.  Although these type of results are not uncommon for the approximate time frame, it is still curious why the two types of methods, both carefully conducted, would produce such different results.  Using statistical techniques, Shott was able to throw out several of the points as outliers, and using dispersion techniques, further eliminate the spread of several of the dates.  It is not clear to me how he would be able to throw out so many of the points, as the statistical jargon is quite dense, but in his conclusion he states that many similar problems that have been found in other woodland sites can be reduced by similar methods to a more concise and apparent result.  I would imagine the author has taken some hits for eliminating up to 10 of his 20 data points in his conclusion that a short, single occupation took place at Childers, but I do not have enough expertise to make such a claim definitely.

1 comment:

  1. It is interesting that it was a shortcoming for the Shott article to use two different dating methods, due to wildly different results. One of the articles I read for the Woodlands, which was written by James Stoltman, makes the argument that it is actually best to use more than one method in analysis, although he was not specifically talking about dating. In his analysis of various sites throughout the Northwest Woodlands, he found that combining both visual comparison and point counting with his ceramic paste analysis allowed for better results. Again, however, this was not actually dating, but I would figure that the more methods, the better the results. This might indicate something interesting about the Childers site.

    Stoltman, James B.
    1989 A Quantitative Approach to the Petrographic Analysis of Ceramic Thin Sections American Antiquity 54(1):147-160