30 Days of Tennessee Archaeology 2015, Day 19
Mississippi State University
Fort Lewis College
Like most archaeologists in North America, we both have fond memories of walking freshly plowed fields looking for “arrowheads” during our childhood. It’s not a stretch to say that this a common pastime in the southeastern United States, and it’s probably the gateway by which most people were exposed to archaeology prior to the proliferation of documentaries on television and the Internet.
As undergrads, we both took introductory Anthropology courses were fascinated by Ice Age Colonization of North America. The thought of small groups of hunter-gatherers traversing an unknown continent inhabited by mammoths, mastodons, giant sloths, and other extinct megafauna species was fascinating. That feeling has never gone away.
Even though we both left Tennessee to pursue our doctorates in Anthropology, and are now professors in different states, we both continue to do research in Tennessee. The reason? Tennessee has a spectacular archaeological record, especially if you’re interested in studying the early colonization of the continent. In fact, the Tennessee River Valley has the highest concentration of early stone points in North America.
This brings us to one of the questions archaeologists frequently get when they ask farmers or collectors to show them the artifacts they’ve found: “Why do you care so much about the arrowheads? Sure, some of them are cool to look at, but what does that tell you about anything?”
Our usual response is, “Well, a lot!” Most collectors select artifacts based on their appearance or rarity. For us, however, it’s not about the individual artifact, but what we can interpret about its’ entire life history. We want to know how these items are made, used, and discarded, because that gives us all kinds of information about people in prehistory. It’s the archaeological equivalent of studying commodity chains and the stock market to understand world’s economy today.
By studying locations where people made stones tools and piecing together the waste flakes they left behind, archaeologists can deduce the sequence of steps it takes to make particular artifacts. It’s the equivalent of reverse-engineering a recipe or owner’s manual. However, unlike today where such things are written down, in prehistoric North America the knowledge of how to construct a stone point was communicated verbally. At the Topper site in South Carolina, Derek Anderson found evidence for what this process may have looked like by reconstructing the production of a stone tool by refitting all of the individual pieces (i.e., artifacts) back together. He argues that it looks like several people were sitting around a fire (based on the distribution of burnt flakes) and the tool was passed around the fire with more difficult, expert removals happening over and over in the same spot. In other words, less experienced knappers would work on the tool until they reached a point where they couldn’t proceed any further, and then they would hand it to the more experienced knapper. Today, we would call this scaffolded learning.
Despite not having “how to” manuals, the shape of points remained fairly stable over time, with some styles lasting several centuries. However, as aesthetic preference or technology changes, so do the size and shape of the points. We see the same thing happening today with clothing, cars, and furniture, although at a much faster rate. Also, sometimes things just change – since the knowledge of making a stone tool is transmitted verbally, it’s prone to the same tendency to make communication errors like the classic telephone game. Yet, all of these factors combined have the effect of making the size and shape fluctuate over time in a predictable manner. If we can find them in good stratigraphic context and/or associated with organic materials that can be radiocarbon dated, we can place these styles in chronological order.
Once we know the time ranges for the specific shapes and sizes of points (i.e. types), we can begin reconstructing how people used them, and get a sense of the economic decisions behind prehistoric hunting. Imagine a modern bow hunter – if they shoot at a deer and miss, do they worry about retrieving that arrow? There is a lot that goes into decision – the number of arrows still in the quiver, the likelihood the arrow broke, the cost of replacement, the time spent searching for it, and the time lost that could be spent hunting. By examining the frequency of resharpening and re-use in a large sample of spear points we get an idea of how prehistoric hunters perceived their chance of success and failure. That perception influenced the size and shape of prehistoric toolkits, and how people decided to organize their technology. This approach is especially powerful if you have faunal data for comparison.
As an example, the Tennessee Fluted Point Survey includes information on over 5,000 of the earliest spear points reported from across the site. By closely examining this dataset, we see a strong relationship between the lengths and widths of the earliest point types (Clovis and Cumberland), whereas later point types (Quad, Beaver Lake, and Dalton) show more variation. Why? From the faunal remains at Dust Cave in northern Alabama, we see that Quad, Beaver Lake, and Dalton points are associated with species that are fast and particularly hard to catch (like birds). This pattern suggests that people making Quad, Beaver Lake, and Dalton points were hunting species that were harder to catch, which likely resulted in more firing attempts and more effort in re-shaping the tip so they could fire it multiple times. These re-sharpening episodes whittle away the length of the point versus the width, which is why we see variations in the data. In other words, their stone-tipped spears were taking a beating. By looking at the points, we can also see that with Clovis and Cumberland show very little re-sharpening. To us, this indicates that Clovis people weren’t putting much emphasis on gathering and re-using spear points they’ve already fired.
Finally, we can use the distribution of points to begin reconstructing where people were (and where they weren’t) in prehistory. For example, using data for the Duck River, it looks like the earliest recorded groups (Clovis and Cumberland) stayed close to the confluence with the Tennessee River. Later groups appear to have expanded to higher and higher elevations on the Highland Rim and Cumberland Plateau, possibly related to increasing temperatures at the end of the last Ice Age.
So, when a person asks us, “What does that tell you?,” we can honestly respond with “A lot!” and mean it. If we have enough of a sample size, we can tell you how old the points are, how they were made, the structure of prehistoric learning networks, the economic decisions related to hunting, and where people were spending their time (and where they weren’t).