Some thoughts on prehistoric land use in NE Oregon

So…first blog, first blog entry. Not quite sure what to write, so might as well start off with what I have been working on recently. Two weeks ago, I attended the Northwest Anthropological Conference (NWAC) in Newport, Oregon. Saw a number of good papers, some not so good. In general, the conference was fun and informative.

I presented on some stuff I have been playing with for a little while. It was a pretty simple review of prehistoric land use in the Blue Mountains of NE Oregon. The data was collected for a recent project, but as the project is probably canceled, I thought I should proceed with using the data and what I learned. May work it up into an article.

I only used excavated data, no site forms, ethnographic or historic information. That's all great stuff, but I wanted to try and get a grip on what we are learning from the archaeology. There were a lot of problems with the data. For some strange reason, no one in that part of the state puts quantitative data in tables, which makes collecting info pretty hard.

Anyway, I ended up with about 18 temporally discreet sites dating to the Middle Holocene (8,000 to 4,000) and Late Holocene (4,000 to 0). I used diversity and cluster analysis to try and define site types, then sees how these types varied over time and space.

There were no significant differences in diversity or elevation between the Middle and Late Holocene assemblages.

The most satisfactory clustering provided two primary clusters:

Cluster 1, which included most of the sites, was richer and more even. Overall, these sites seem to have a generalize suite of tools. The mean elevation for Cluster 1 sites is lower and they occur in both the Middle and Late Holocene.

Cluster 2 included three sites, which were less rich and more uneven. This clusters formation appeared driven by very high densities of used flakes, maybe more cores and cobble choppers, but fewer scrapers. I included a site that had only four points and lots of bison bone into this cluster. Due to the small number of tools, I had excluded it from clustering. The mean elevation for Cluster 2 sites is higher and they occur only in the Late Holocene.

All differences in diversity and elevation were significant (p<0.05).>

Overall, it appears land use diversifies in the Late Holocene. At least two types of specialize sites occur and activities are segregated across the landscape.

What’s really important here is what we don’t have. We don’t have villages (at least no housepits), clear quarry sites or specialized plant processing sites. Their lack is probably a sampling issue. We just have not excavated any of these types of site yet. But, that may also not be the case.

Take the lack of quarry sites. The assemblages from 35UN52, the Stockoff Quarry site, don’t really look like a quarry. Assemblages are large and diverse, with moderate biface densities and only a little early reduction debitage. The site seems to be a generalized residential site, rather than a specialized quarry site. Thus, it’s probably a logistical base. While it was certainly used while quarrying basalt at the nearby outcrops, the lack of a typical quarry signature suggests we may be looking at an embedded raw material procurement system.

I would have to do some more analysis, but I also think projectile point densities increase in the Late Holocene, while groundstone decreases. As noted above, this could be a sampling issue and we just have not excavated specialized plant sites. But it does seem that ground stone is more rare at sites dating after about 4,000 years ago. Maybe we are seeing an increase in the importance of hunting and a decrease in plant processing. Hmmm.

What’s really fascinating about all this is all the possibilities for research. I really like this approach for getting a handle on archaeological work.