30 Days of Tennessee Archaeology 2016, Day 12
Washington State University
Some artifacts can be quite rare and even difficult to identify in the archaeological record, especially when compared to the number of and information on projectile points and ceramics. When there are fewer available examples of an artifact type, it can be difficult to understand it’s significance, as well as how it was produced. Turtle shell rattles are one of those difficult to identify artifacts. Turtle shell rattles have been found throughout Tennessee in a variety of prehistoric and historic contexts. For the Mississippian period (ca. AD 1000–1450), there have only been approximately 72 rattles recovered in Tennessee, of which most only consist of one or a few fragments, and are often associated with ceremonial or burial contexts. You can read more about the context surrounding turtle shell rattles in my previous TCPA blog post from 2014.
Turtle shell rattles from the Mississippian period have a much different construction than those from the Historic period (ca. AD 1450–present). How did people in prehistory construct turtle shell rattles, and how can this help us to identify small, broken pieces of turtle shells that may have functioned as rattles? These were some of the questions that I addressed in my research at Middle Tennessee State University.
To answer these questions, I turned to experimental archaeology, which, in this case, was to recreate an artifact to understand the process of the past item or technology. To create a turtle shell rattle, I needed to first decide on a design (e.g., where to drill the holes, type of cordage to use, type of tools to use). I decided to model the reconstruction after one of the Hiwassee Island rattles, which was illustrated by Lewis and Kneberg (Figure 28). Incidentally, this is also the illustration that I based my tattoo on.
Once deciding on a design, the first step I took was to use river cane and sand to try to drill the holes into the turtle shell**, which really did not work at all. In my second attempt, I used a bow and chert drill, which worked quite well and very quickly. The chert drill created a concave, oval shape, and a small ridge in the middle of the hole when the shell was drilled from one side, which is similar to the hole shape of most Mississippian period turtle shell rattles. When the shell broke during the drilling process (an unintended effect), it left very sharp angles coupled with drill indentations, which was also similar to some possible rattle fragments from the Fewkes and Castalian Springs sites.
After the top (carapace) and bottom (plastron) of the shell were drilled, cordage and rattle implements were added to complete the rattle. I used suede cordage although cedar or a similar fibrous material was most likely used to make cordage prehistorically. For rattle implements, I simply used river pebbles, although prehistorically, drum teeth and hard seeds were also used.
By studying and recreating the drilling stages, we can recognize turtle shell rattle manufacturing among artifacts recovered from the archaeological record. The breakage patterns among turtle shells allow us to see and identify this process in the archaeological record. The study of cordage and river pebbles can also allow us to see the unintentional modification upon the turtle shell. From my limited use of the rattle, there was no immediate modification caused by the cordage or the river pebbles, although extensive use may result in some unintended modifications to the rattle. These factors show some of the indicators of how to identify turtle shell rattles within the archaeological record, no matter how small the piece.
Check out this video (also embedded below) for some of the process of creating a turtle shell rattle.
**Disclaimer: No turtles were harmed in the making of the experimental rattles. These were empty shells that were collected under the proper Wildlife permits.