An important part of any museum’s mission is to ensure that its collections are properly and completely cataloged, both to ease the tracking of objects and to make collections permanently accessible year-round for outside researchers. Museums, especially the Smithsonian, collect objects and specimens not only to exhibit to the public, but to also assist in current and future research. In order to make sure that we meet this important requirement, we first need to know what we have and where it is in order to tell researchers what we have for them to study! For more information on how we assign catalog numbers, see our previous post here:
Part of our work here in Dr. Rogers’ Archaeology Lab is the continuation of the cataloging of collections from the River Basin Surveys (RBS) project, a massive salvage archaeology project that examined a number of historic and prehistoric sites along the Missouri River. The project, which was conducted between the 1940s and 1960s, sought to retrieve as much information about these sites as possible before the construction of a series of dams and reservoirs that would have flooded them forever.
This summer we have been been working on cataloging the collections from the Sommers site (39ST56), located in the Big Bend Reservoir in South Dakota. As we were going through the collections, attaching catalog numbers to each object, we were increasingly intrigued by the wide variety of interesting artifacts we were working with, so we decided to do a little extra research on the site and share our findings!
First, a little background information: excavations at the site took place in 1947 under E.E. Meleen and again in 1964 under Richard Jensen. The village contains more than 100 house depressions and covers an oblong area 1350 feet long and 850 feet wide and encompassing a total of 26.3 acres, making it one of the larger sites in the area (Johnson 1977). The site had a fairly dynamic settlement history, in which its population and size grew and contracted over time (Steinacher 1990). The site began as a small, unfortified settlement in the late 10th century AD but sometime around 1100 AD fortifications and a palisade were constructed at the north end of the site (Johnson 2007). There are 17-18 structures enclosed in a fortification ditch; of the houses at the site, scholars believe these were used the most extensively across time.
While we were cataloging the collection, we kept coming across large carbon and wood samples. During excavations of the site, archaeologists collected a number of wood posts and charred organic material (carbon samples) with the intent of assigning an absolute date to the site through dendrochronology and radiocarbon dating techniques. Because we came across so many out at the Museum Support Center (MSC) in Suitland, Maryland, we decided to take a closer look at what these samples told scholars about the site.
As it turns out, Sommers was not an easy site to assign a date to. A variety of dating techniques were utilized to determine the site’s occupation dates, from stratigraphy and seriation to historic documentation and oral traditions. In preparation for these dating methods, excavators collected as many samples as possible; not only would a larger data sample provide a higher probability of successfully dating the site, it would also help researchers calculate as accurate of a time frame as possible. With regards to dendrochronology, more wood samples would provide a more complete tree ring analysis, while more radiocarbon samples would allow archaeologists to determine a more concise date with a smaller error margin.
However, when used independently, these different kinds of samples were producing some very different dates. For example, the dendrochronology dates claimed Sommers was 500 years younger than the radiocarbon dates suggested (Johnson 2007)! In addition, the radiocarbon dates themselves covered a fairly wide time frame, from 600-1300 AD, making it difficult to pinpoint a more specific and accurate occupation period (Caldwell and Snyder 1983; Johnson 1977). How on earth were archaeologists going to figure out the site’s true age?!
Researchers then turned to seriation in another attempt to date the site. Looking at pottery characteristics such as rim shape and outer decoration in comparison to the ceramics’ location within the stratigraphy of the site, archaeologists developed a relative dating system for the Initial Middle Missouri period. One of the more common ceramic types found at Sommers that helped relatively date the site is an S-shaped rim with cord impressions on the outer edge (see image below).
While this dating method provided scholars with a general idea of the site’s age, they also wanted to know a more definitive, or absolute, date. To achieve this, they decided to compile the information provided by the different dating methods and use them as a system of checks and balances against one another. As a result, the team was able to establish that the Sommers site was generally used between 982 and 1157 AD, placing it squarely in the established Initial Middle Missouri (Johnson 2007).
Having done this extra research on the site, we can really appreciate the diversity of the objects we’re cataloging, both as objects themselves and as tools to further explore the history of the site. Using a variety of dating techniques and collecting numerous samples can come in handy, even when one least expects it. This diversity is what makes the Sommers site an important collection for studying Middle Missouri archaeology -- making it a site for not just for research in summertime but for all seasons!
What other interesting facts will we uncover as we continue cataloging the River Basin Survey sites? Stay tuned to find out!
Authors: Kelly Lindberg and Kendra Young
Caldwell, W. W., and L. M. Snyder. 1983. Dendrochronology in Plains Prehistory. Plains Anthropologist, 28:33–40.
Johnson, Ann M. 1977. “Testing the Modified Initial Middle Missouri Variant.” Plains Anthropologist 22 (78): 14-20.
Johnson, Craig M. 2007. A Chronology of Middle Missouri Plains Village Sites. Smithsonian Contributions to Anthropology 47. Smithsonian Institution Scholarly Press.
Steinacher, T.L. 1990. Settlement and Ceramic Variability at the Sommers Site (39ST56) Stanley County, South Dakota. Ph.D. diss. (unpublished), Department of Anthropology, University of Oklahoma, Norman.