For the past six weeks or so, I've been attempting to get up to speed on the current status of what scholars believe about the timing of the first human occupation of the Americas. There are at least four scholarly theories about the event or events, and each of those theories is complicated by problems of archaeological dating. Trying to be both precise and clear about dates has not proven easy, partly because radiocarbon dates are sometimes reported calibrated and sometimes not. That's not the fault of the researchers, necessarily; calibration for dates older than 11,500 BP was only available beginning in 2004. And it turns out that c14 dates are not the only ones that need calibration. But, mostly I'm not getting the article written because I keep getting distracted by interesting side issues. Case in point: Ushki Lake.
The Ushki Lake sites are five sites on the Kamchatka Peninsula of Siberia--part of Beringia and thus important for understanding the original peopling of America. The earliest dates on the lowest components were reported as 16,800 cal years BP; but recent tests have revised that date to about 13,500 cal years BP. While I was compiling notes for the piece on Ushki, I found a new article coming out this year in the Journal of Archaeological Science, about sourcing the obsidian at Ushki.
In Obsidian use at the Ushki Lake complex, Kamchatka Peninsula (Northeastern Siberia): Implications for terminal Pleistocene and early Holocene human migrations in Beringia, Yaroslav Kuzmin and colleagues report there are between four and six different sources for the obsidian discovered from the earliest levels at Ushki Lakes--and the sources are between 200 and 300 kilometers away from the sites. Now that's interesting--because it illustrates long range resource procurement strategies. That's something you would expect for Terminal Upper Paleolithic/Late Pleistocene hunter-gatherers. But I wondered why the scholars didn't use the obsidian for dating purposes--obsidian hydration is a fairly inexpensive dating technique that's readily available: why not, I thought, back up the dates with a hydration date?
I hadn't gotten around to asking anybody when I ran across Obsidian hydration dating on the South Coast of Peru, an article by Jelmer Eerkens and colleagues, again soon to be published in the JAS. Eerkens et al. report that obsidian dates are affected by temperature (aha!) and obsidian artifacts at higher elevations hydrate more slowly than those at lower elevations. Since hydration works by measuring the growth of rind, and cold temperatures inhibit that growth, the hydration dates from cold climates would be too young. Eerkens et al. figured out a way to use altitude as a substitute variable for temperature data, something that probably won't work for the Kamchatka Peninsula.
Interesting, though, isn't it? No wonder I can't get anything done.
Sources
Updated 2 Jun 08 to add full citations:
Eerkens, Jelmer W., et al. 2008 Obsidian hydration dating on the South Coast of Peru. Journal of Archaeological Science 35(8):2231-2239.
Kuzmin, Yaroslav V., et al. 2008 Obsidian use at the Ushki Lake complex, Kamchatka Peninsula (Northeastern Siberia): implications for terminal Pleistocene and early Holocene human migrations in Beringia. Journal of Archaeological Science 35(8):2179-2187.
Comments
What a vexing problem obdisian hydration has been over the years!. Joseph Michels, one of the pioneers of the technique, was a professor of mine at Penn State back in the 1960s. I had problems with the results he was coming out with back then. More recently, the question of when the Maya center of Copan was actually abandoned has been addressed using the technique, with results that many Mayanists question. I suppose at this point I would say it still has a great deal of potential that has yet to be realized.
Dick Diehl
Kris,
I think obsidian hydration would work in Kamchatka. However, a few issues. 1. You’d have to generate a calibration curve first (for example, using radiocarbon dates paired/associated with obsidian artifacts that you also measure for hydration). Alexander K. Rogers has published some ways to calibrate for temperature, once you do that (see Journal of Archaeological Science). However, more problematical for dating old things, is that the error terms on hydration dates go up as the square of the estimated date. This means that your ability to resolve time goes down with age. As well, it does this faster in colder climates (i.e., the ability to resolve time is worse in colder climates where obsidian hydrates more slowly).
Overall, I think the main problem with hydration is that people have expected too much from it, and their sample sizes have been too small. It’s not as accurate or precise as radiocarbon dating, and getting 1 or 2 hydration readings to estimate when an event took place is not enough – you need samples of 4 or more, preferably in the neighborhood of 10 (i.e., 10 or more artifacts that date a cultural “event”). Also, you can’t resolve things that are too close in time – for example, in most situations, something that happened 1800 vs. 2000 years ago is not distinguishable using obsidian hydration. But something that happened 1500 vs. 2000 years ago usually will be. If your research questions require the former resolution, hydration may not be able to help you. If the latter, it might.
Hmm, good to hear that it would in Kamchatka. Wonder if they’ll do that. I take your point about multiple samples–I still run across articles with dates based on one or two c14 dates, so it’s an important issue for sure.
Here’s the article that Dr. Eerkens is talking about:
htRogers, Alexander K. 2008 Obsidian hydration dating: accuracy and resolution limitations imposed by intrinsic water variability. Journal of Archaeological Science 35(7):2009-2016.
http://dx.doi.org/10.1016/j.jas.2008.01.006
Thanks! Very interesting…