Dating the middle to late woodland transition in the Illinois Valley: Radiocarbon and thermoluminescence dates from the Baehr-Gust site

Midcontinental Journal of Archaeology, Spring 2003 by Holt, Julie Zimmermann, Feathers, James K

The uncalibrated radiocarbon age of bone collagen from the AMNH bone awl is 1750 or - 50 B.P. (Stafford sample SRLA-1059), or ca. A.D. 200. This date also offers support for the second hypothesis, since it indicates that at least one of the mounds at Baehr-Gust was of Havana construction. Unfortunately, it is not known which mound was dated, since Snyder did not record specific provenience on the artifacts he sent to the AMNH.

TL Dating

TL results are presented in Table 3. The ages reflect external dose rates as determined from the six sediment samples collected post-excavation by the senior author. These samples did not greatly differ from each other in radioactivity, even though they were collected from different parts of the site. The gamma dose rate from the sediments, uncorrected for moisture content, ranged from 1.058 or - .054 to 1.197 or - .066 Gy/ka. These values were not significantly different at one sigma, except between the one sample from the bluff top (the highest value) and two of the floodplain samples, and these were not significantly different at two sigma. Error in the ages (at one sigma) is calculated at both interval and ratio scales, showing an average improvement of about 25 percent with interval scale dating. However, since the external dose rate is not known, only something close to interval scale dating is possible here. How close the pairs of sherds are in age can be assessed by considering different external dose rates.

Using the weighted average of the external dose rates from the collected sediments (Table 3), all three pairs of sherds with classification problems have dates that overlap within margins of error. But would the ages be within the range of errors if the external dose rate was dramatically different? Age differences were recalculated using Equation 2 assuming external gamma dose rate ranging from 0 to 3 Gy/ka (Table 4). A value of 3 is probably higher than is likely, since the University of Washington laboratory maximum (for some 200 sherds across North America) is about 2. For the sherds from Features 4(1988) and 89-XVIII, the difference in age within each pair is not significant over this entire range. For the sherd pair from Feature 1(1987), the difference is significant only for external dose rates less than 1.1 Gy/ka. Dose rates in this range are likely (the external dose rate for the collected sediments is about .9 Gy/ka after moisture correction), but the difference is just barely significant at one-sigma and not at all significant throughout the range at two sigma, so it is not likely these sherds differ in age by much. At a dose rate as low as .5 Gy/ka, for example, the difference in age (at one sigma) could be as low as 60 years.

The one good test-that is, the pair of sherds from Feature 91-XVIII which all ceramicists agreed consisted of one White Hall sherd and one Havana-Hopewell sherd-produced ages that indicated the Havana-Hopewell sherd is hundreds of years older than the White Hall sherd (Table 3). The difference is highly significant at all values for external dose rate (Table 4). In the range from .5 to 1.5 Gy/ka (typical for most samples), the difference changes from 830 to 720 years, with an error of about 130-170 years. Most of the pottery from Feature 91-XVIII was classified as White Hall, with only a few sherds classified as Havana. The TL dating makes it quite clear that this was a Eate Woodland pit with some older Middle Woodland materials.