Several people wrote in with news about a possible new genus of pre-humans, based on fossils found in Kenya. Check out the less-technical articles at
CNN or
MSNBC, or the hardcore paper in
Nature (which has some very nice pictures of the actual fossils).
Slashdot Mirror
Radiocarbon dating assumes only three things:
1: that the rate of radioactive decay of the carbon-14 isotope has remained constant.
2: that the carbon-14/carbon-12 ratio in the atmosphere was the same 'back then' as it is now (or was before we started setting off nuclear weapons blasts...),
3: that the carbon in the sampe being dated was derived exclusively from atmospheric carbon dioxide.
and, 4: that c-14 and c-12 are absorbed into plant tissues (and subsequently into animal tissues) at the same rate. That is, that plant A does not absorb proportionately more c-14 than does plant B, and so on up the food chain.
The first assumption is a pretty safe one, since to doubt that would be to call into question very fundamental physical principals. I think the halflife of c-14 is about 5580 years, but that's just off the top of my head.
The second assumption is obviously false, since cycles in the sun's radiation output affect the amount of carbon 14 produced in the upper atmosphere. For the recent past (4-5000 years) this has been corrected for by testing the c-14/c-12 ratio in precisely-dated (annual) tree rings in very old trees. So this is not a problem for recent times, and the difference even as much as 20000 years ago is probably not more than 1000 years or so. Again, not too much of a problem.
The third and fourth assumptions are also problems, but the problems are understood and are being explored as we speak. Every year the correction methods for c-14 dating get more sophisticated. The important thing to remember, though, is that we are down to pretty fine points now. Any error introduced by failures of assumptions three and four would be in the range of a few centuries, max. Significant, but not enough to invalidate conclusions about the antiquity of human cultures, say.
Sorry for the long post. I just wanted to point out tha while it is true that radiometric dating methods do make certain assumptions about conditions in the past, the limitation which these impose on the accuracy of the technique are well-understood and are taken into consideration by archaeologists and paleontologists.
I also want to say that, like any scientific tool, radiometric dating is imperfect. But it is a hell of a lot better than simply guessing, or throwing up our hands. Do we want to study the past systematically and rigorously? If so, we must use these tools. If not, then, heck, the Bible was good enough for my grand-pappy.
Claim your namespace.
For those wanting something in between the hardcore Nature article and the mostly fluff CNN and MSNBC articles, here's a layman's version prepared by Nature itself. Check out Nature Science Update.