Watkins SIR Spring 2006

Two classes for me this term, both of which are only tangentially related to Archaeology.

Agent-based Modeling in the Social Sciences

What is agent-based modeling you may ask? At the end of an entire semester, you'd think that I'd be able to answer that question. Tragically, I'm not sure that I'm going to be able to do that.

It may be best to begin with an example. Have you ever wondered how they do those epidemic disease predictions? You know on the news when they say, "If the Bird Flu broke out in Hong Kong, 10,000 people would be infected in the first 6 hours, Tokyo and Beijing would be hit within the first 12 hours, yada yada yada..." Well, these predictions are made with agent-based modeling. They basically program a bunch of little people, aka agents, and let them walk around a city. They program stuff like infection rates and the proximity of infection and then watch what happens. They run it like 1,000 times and then interpret the results.

We were supposedly trying to apply this technique to problems in the social sciences. Good luck. This is simply the newest fad in archaeology. Most of the models are based on so-called prisoner's dilemma exchanges, where agents are assumed to be selfish. I could get into a lot of reasons why ABM is problematic, but it just makes me remember the horror of last semester.

Archaeometry

This class was really cool. I worked with turquoise, a geological sample from Kingman, AZ and a bead Joel and Cady loaned me from the Parowan Valley. We learned a materials science approach to analysis, where you characterize the texture (surface characteristics), structure (phases), and composition (elemental makeup) of a substance. Identifying texture is done primarily with scanning electron microscopes. There are a variety of techniques available to determine elemental composition (INAA, PIXE, Electron Microprobe, ICP-MS, etc). These sort of data are most often used in provenance studies. What is a phase you ask? This is kind of a tough question. I think it's best explained with an example. If you sent a piece of chert through a compositional analysis, the results would be mostly silicon with traces of whatever else is in the material (i.e., Ti, Fe, Ca, whatever). Running the same sample through a structural analysis would identify it as chert. Essentially, instead is identifying elements, compositional analysis identifies minerals and various molecules.

Anyway, these folks believe that to do an analysis right, you need to address texture, structure, and composition. There are a lot of advantages to this approach, and if you have the $$ and the means, I would highly recommend it.

Other News

Knowing what Mike and Dave are currently going through, I feel a little guilty mentioning this, but it is going to come out in the end. My dissertation has just been funded by the NSF, through no effort on my part. At the beginning of the semester, I was attached to an NSF proposal submitted by my advisor, Dave Abbott, and some other ASU faculty. A few weeks ago, we learned it had been funded, and I'm going to be paid for 2 years to work on the project. We'll be investigating the Southern Sinagua (the so-called Verde Confederacy consisting of the pueblo sites along the Verde River, like Tuzigoot, Montezuma's Castle, etc), and my part of it is going to be creating a ceramic typology, chronology, and a mechanism for sourcing the pottery. I think that it's going to be kind of a big deal, we're essentially taking a culture from 0-60 in a few short years. Anyway, I'm pretty excited about it, while at the same time, I ackowledge that I have fallen into this without really doing anything much to particularly earn it.

Hope to see some other SIRs soon. I feel bad that mine was late too.