first post!

I haven't written a blog before, so this should be interesting. Let's hope this one isn't as self-absorbed as the blog concept would imply (:

I'll start with the class. Anthro 128 is an anthropology class regarding human origins and prehistory, but it is saturated with archaeology. Indeed, pre-history -by definition- leaves us with no first-hand accounts or witnesses aside from physical evidence. That's not to say other disciplines don't provide other pieces of raw clay which we can weave into an elegant tapestry describing human history.

Recently, genetics have come into play to investigate the divergence of modern human location-groups from a common ancestor, and the contributions of particular remains to the current species (given that the DNA is salvageable) This study  is prototypical of such investigations of archaeological DNA extraction, a cursory glance (search ancient DNA)will find many such studies, here's some examples. Incidentally, that last link also describes some problems with this method. DNA from researchers can contaminate the results, because tiny samples are vastly multiplied, which also multiplies error. Structural mechanisms which favor the decay of certain aspects of DNA can also bias what remains of ancient DNA.


The other genetic approach is to analyze the degree of similarity between modern human groups to estimate how 'distant' they are from in each other, if we assume mutations occur at a fairly common rate. This has the advantage of using very available DNA from living humans, as opposed to rare and chance-based ancient DNA extraction, as well as the advantage of DNA not selected by the decay profile of the specimen.

If we look at an event which was long enough ago to not be patently obvious, but which happened recent enough to provide a large amount of evidence, we can use both approaches. An example of this is a study which looks at Aleutian populations ~1-2kya and thus has much surviving DNA evidence. In this manner, or even without genetics, (for instance observing how modern-day hunter gatherers live in their environment) we can extrapolate backwards to understand what we find.

source: xkcd.com

The problem with extrapolation, of course, is the old cliché of "past performance is not an indicator of future returns", or, in this case, "past performance is not an indicator of even further past performance which was so far ago we're not sure what really happened." Regardless, some information is usually better than none at all, and the models created are often tested and either confirmed or refined by newer research.