About 4,000 B.P. (around 2,000 B.C.) the environment began to change along the South Carolina coast. Sea levels rose from a low of about 12 feet below their current levels to almost modern levels. This increase in sea level created the marshes, inlets, and tidal areas that today make the coast such an environmentally rich and productive area. This ecological change also brought about significant changes in the way prehistoric South Carolinians, or Native Americans, interacted with their environment. 

The Native American groups that had been seasonally moving from location to location, hunting deer and collecting nuts, began to understand that there was a great diversity and abundance of foods on the coast. So many foods to choose from in one place provided another option – nearly permanent settlements. 

These people began to collect shellfish and trap fish using nets. They supplemented their diet by collecting occasional marsh visitors, such as marsh rabbits, turtles, opossums, snakes, and even alligator. They still hunted deer and collected hickory nuts.  However, by adding a variety of other foods, their diet improved and they were able to live in one place longer.  This reduced the amount of time they had to move from one location to another. This is what archaeologists call “sedentism.”    

Shell Rings and Shell Middens 
An unusual type of archaeological site developed during this period – doughnut-shaped piles of shell that archaeologists call “shell rings.” Although these rings are found in different sizes, they average about 160 feet in diameter, 4 to 5 feet in height, and have a width at their base of about 45 feet. There are several theories about the function of these sites, but what we know is that they contain a huge assortment of floral and faunal remains – showing us the varied diet of the people who lived on or around them.  

Because these people existed so long ago, archaeologists cannot associate them with any modern Native American tribal groups.  Instead, these Native Americans are known as the “Thom’s Creek” people.  Thom’s Creek is also the name of the type of pottery these people made.  Thom’s Creek pottery has a sandy texture and is often decorated with various punctuations.  Found along the South Carolina coast, it is associated with the shell rings, and dates from about 2,000 B.C. to 500 B.C.  It is one of the earliest types of pottery made by the Native Americans in South Carolina. While the Thom’s Creek people are best known for their shell rings, these were not the only sites they left behind. 

In addition to shell rings, there are also huge shell mounds or middens such as Spanish Mount on Edisto Island (Charleston County) which several USC archaeological field schools have studied. Probably the most common Thom’s Creek sites are actually small accumulations of shells and refuse (the garbage thrown away by the Thom’s Creek people), and they have received very little study. Recently, however, with the cooperation of Carolina Park Associates, archaeologists have been given a unique opportunity to examine one such site.
 

Carolina Park Associates and 38CH1693
Known as 38CH1693,  this site was discovered in 2003, at the corner of Airport Road and US 17, during routine archaeological surveys.  The site was slated for additional investigation prior to development.  Over several weeks during the summer of 2006, archaeologists with the Columbia, South Carolina based Chicora Foundation, conducted excavations, made maps, took samples, and collected artifacts from this site.

Research indicates that the site was situated on a relic Pleistocene dune ridge. Located nearby were not only upland resources, such as deer and hickory nuts (a favorite – and nutritious -- food source for Native Americans), but also springs and marshes.  The springs  supplied fresh drinking water, while the  saltwater marshes provided fish and shellfish to eat. Within just an hour’s walk from 38CH1693, the Thom’s Creek people could find a variety of foods – making this an ideal area to live. 

Archaeologists found that the site was not a shell midden; instead, it consisted of several areas where the inhabitants had dug pits in which to steam oysters and other shellfish. Thus, while there were no layers of shell, there were individual pits (what archaeologists call “features”). These features were especially important because they helped preserve the food remains. Recovered from these features were the shells of hickory nuts eaten and discarded into the fire, as well as animal bones, and fragments of shellfish. The features even preserved pollen and phytoliths – all of which were studied by the Chicora team.
 

What Was Learned?
One of the most important findings was the age of the site. Using radiocarbon dating of the hickory nutshells, Chicora determined that 38CH1693 dates from about 1,700 B.C. to 2,000 B.C. In fact, two of the four dates place occupation at almost precisely the same time – 2,135 B.C.  

This is important since it reveals that the site, occupied during the middle of the Thom’s Creek period, was used only briefly. 38CH1693 is giving us a brief snapshot of life at a Thom’s Creek site. 

Other information, however, is not so simple to interpret. For example, the animal bones suggest the site was occupied seasonally, perhaps during the spring through late summer or fall. This is suggested by some of the fish species present. Not a lot of deer bone was found, suggesting that the site was not used during the fall when deer rut and are more common in the coastal area. 

On the other hand, the ethnobotanical remains are suggesting of a fall or early winter occupation. Grape seeds and hickory nutshell were two of the plant foods recovered. Grape pollen was also found in the pollen record at the site. Weedy seeds like greenbrier, knotweed, and bedstraw were also present in the features. 

Another common food trash at the site was shell:  oyster, clam, periwinkle, and stout tagulus. Although these occur in different marsh areas, all are species common to the area. These shellfish can be gathered at any time of year, so they are not especially good seasonal indicators. However, a small parasite of the oyster, known as Boonea impressa, is a good seasonal indicator. At 38CH1693 these almost microscopic snails indicate collection during summer and fall, as well as early winter. 

The shellfish do tell us about procurement strategies and the importance of shellfish to the prehistoric diet. Although periwinkle are small, they are easily to  collect and might be a food gathered by older people or small children. Moreover, shellfish provide good quality proteins and would have been invaluable in the prehistoric diet. 

Since remains from different seasons are found in the same refuse – the garbage thrown away by the Thom’s Creek people – 38CH1693 may provide evidence for people living at the site off and on during the entire year. 

And speaking of garbage, archaeologists found plenty of evidence of other things discarded by these early occupants of Carolina Park. The most common item is broken pottery. The Thom’s Creek pottery at this site was primarily undecorated. The most common surface treatment, however, was finger pinched. Made by pinching the wet clay between the thumb and forefinger, the artisan typically produced linear arrangements of this design around the vessel. Other decorations included finger smoothed, and punctuated. The pots were about 13-14 inches in diameter, although a few were as small as 6-inches. Even when these pots were broken, they were not necessarily discarded. Some were actually repaired and reused. By drilling holes on each side of the crack, and lacing the pot together with sinew, the pieces of the pot were held together. To make the pot watertight, or leak proof, pitch might be smeared over the break. Examples of these mending holes are found in the 38CH1693 collection. 

When sherds (bits of broken pottery) were thrown away they might still be picked up and used, or recycled, – most commonly as abraders or hones. The Thom’s Creek people used them to shape and smooth bone tools. These bones are thought to have been used to weave the nets for collecting small fish. Other bones were sharpened to make tips for spears (the bow and arrow was not invented for another 2,000 or so years).

Although a few stone flakes were found at 38CH1693, no finished tools were recovered by archaeologists. This is probably because the coast has few sources of stone, making it a highly prized and carefully tended object. Stone points might be resharpened, but they wouldn’t often be lost. 

Another interesting find is what archaeologists call a coprolites – a fancy word for fossilized fecal material. In a shell-rich environment the organic matter can be replaced by calcium, helping to preserve these remains. Identified as human based on size, shape, and contents, they can be used to better study exactly what Thom’s Creek people were eating.  

Unlike the condensed for television accounts of archaeology, the research at 38CH1693 reflects the reality of archaeology. After several weeks of careful but exhausting excavation, followed by months of equally painstaking analysis, we know a little more about how the Thom’s Creek people lived. There is no major breakthrough or amazing discovery – just hard work that gradually builds on itself to allow us, bit by bit, to better understand our ancestry and the history of South Carolina.

For More Information 

• Read the entire Chicora report on excavations at 38CH1693
• Another pdf download is the 2002 National Park Service test excavations at the Fig Island Shell Ring, available here.
• A brief overview of South Carolina archaeology is available  here [link to Chicora’s SC Archaeology page].
• If your local public library doesn’t have these books, you can ask that they be obtained for you through interlibrary loan, a free service:
• Structure and Process in Southeastern Archaeology. It contains an article, “The Form and Function of South Carolina’s Early Woodland Shell Rings” by Michael Trinkley.
• Early Pottery in the Southeast