Etowah Archaeological Field School: Weeks 5 and 6

If you followed this blog through the summer you know that I left it unfinished. Things got really busy the last two weeks of the field project and I didn’t keep up with the blog. A month later, here I am trying to wrap that up.

One of the most exciting things about the last two weeks is that we got some extra help from members of the Muscogee (Creek) Nation’s Cultural Preservation Office. In Week 5 Emman Spain, David Proctor, Chumona Deere, and Corain Lowe drove from Oklahoma to work with us for a week, while in Week 6 Charles Kelly, Charlie Hicks, Alan King, Robin Soweka, and Gano Perez did the same. We also were lucky enough to have some experienced volunteers in Mark Crawford, Carl Ethridge, Pam Enlow, and Kelly Ledford.

Back when I finished Week 4, Grant and Tim were working in Grid 6 on the southern expansion of the unit producing the whole pot.

In that extension, they exposed more of the smashed vessels and they also found some of those single-set posts I was hoping to find, confirming that the building that I thought was a single-set post building actually was a single-set post building. They also found some kind of strange mineral concretion that seemed to extend into sterile subsoil. We left it in place and I am still shrugging my shoulders.

On the west side of Mound A, James (now joined by Carl Ethridge) excavated a second unit to the west of the first one they started in Unit 16. As I was hoping, they found a couple of single-set posts to the east of the mass of daub. 

Similarly, in Grid 18 (the unit started by the Carter family) Owen (with help from Mark Crawford and Chumona Deere) exposed another nice daub concentration. Immediately to the east of that concentration they found a standing charred post, along with a second post hole. 

In Grid 20 we opened up units along what looked to be the north and south walls of a large structure between Mounds A and C. On the south wall (E763 line), Trey and Julia came upon what looked to be a wall-trench feature, a portion of which they excavated. Because it continued into the southern wall of their unit, I asked them to open a second unit to the south of the first. Eventually they came down on what looked to be the other side of that trench feature. However, after spending a very long time carefully excavated what we thought was a trench, Trey and Julia (now with the help of Corain Lowe) eventually learned that it was not a feature after all. Ultimately we recovered a great sample of artifacts from a rich Etowah period midden in these units but didn’t make any sense of the building identified by the gradiometer.

Unfortunately, the unit placed on the northern wall was equally confusing (N989E757). Tara and Anna found the entire floor of their original unit covered by a daub mass, so I asked them to expand to the north. My hope was that we might find the edge of the wall (whose collapse made the daub surface) and some indication of architectural form. Instead they (with the help of Chumona) found a complex distribution of large rocks, charred wood, and daub. It became clear to me that this wall (or whatever it may be) was too complicated to be figured out with a couple of 1-m units, so I decided to stop excavation and map what we had exposed. Hopefully we will have a chance to come back and expand both sets of excavations on this large anomaly. 

In Grid 15, as I already discussed, we found a small, buried platform.

It appeared from the deep test we excavated on its northern side (N999E705) that it actually was built in two stages but grew laterally rather than vertically. The first stage was made of homogeneous red clay piled to a height of about 50cm. Its flank was exposed long enough for a sandy layer to wash down it.

Eventually the platform was expanded laterally by adding an orange clay layer that overlapped the flank of the red platform. On the south side we exposed that orange expansion (along the E703 line) that extended past the original red platform some 6 meters as it gradually sloped down to the original ground surface. On the summit of this platform Ben and Luis also found a shallow pit very densely filled with charcoal and a set of narrow and deep posts that may have been set in a trench. We will run a radiocarbon date on some of the charcoal from the summit to see when the platform was built.

As we finished units on the west side of Mound A, we moved operations back to the north side of Mound A and focused on exploring more anomalies that we thought represented wall-trench buildings. We opened up units in Grid 9 to explore two suspected wall-trench buildings and also located units on four additional magnetic anomalies along the northern edge of the site. These last anomalies were identified and units positioned using 1-m interval data collected in 2008 rather than .25m-interval data collected this year. As you will see, this created some potential problems.

 

In Grid 9 Will, Emman, and Pam (and eventually a whole bunch of helpers) began excavation on a very promising anomaly (southernmost rectangular anomaly in the grid), while Brenden and David Proctor started on a second (immediately to the north). No one has yet to investigate that tempting circular anomaly defined by low rather than high magnetism. Chet predicts this is an agricultural shrine.

Brenden and David eventually exposed a very nice wall-trench with posts in the bottom. Unfortunately, Will was not as lucky. In fact, even after an expansion to the south Will’s efforts did not definitively locate any architectural features. There were some possible trenches visible in the profiles of his units and, as we will see, those may be the remains of the building visible in the magnetic data.

To the north of Grid 9 (in the vicinity of Grid 12) we set in units on three additional magnetic anomalies thought to represent wall-trench buildings. We did this thanks to the great work of SRARP’s Chris Thornock using 2008 data.

Anna and Tara started the easternmost unit and they eventually also were joined by a whole host of people. Unfortunately, their initial unit did not uncover any clear architectural evidence. However, in the north wall of that unit a possible trench was visible, leading to the opening of another unit immediately to the north. This was a complex unit made messy by a later building with a nice puddled clay hearth and a fairly continuous but sparse layer of mica. It was also made messy by the intrusion of many tree roots.

Eventually we were able to uncover a very nice wall-trench that appears to have been rebuilt and expanded, creating two parallel rows of posts. It is noteworthy that Brenden ended up doing much of the work on this wall-trench. In fact, Brenden was involved in the excavation of every wall-trench segment we found this summer. As often happens, we did not expose this feature until late in the afternoon of the last day in the field and did not finish mapping it until long after most of the crew had left the site for the day.

In the unit to the west of this, Grant and James (with help from others like Corain and CJ) opened up a 1-m unit. Within 30 cm of the surface they had found single-set posts. Excavating deeper did not produce any other architectural features, especially wall-trenches. However, like in Anna and Tara’s unit, a trench feature was visible in one of the unit’s profiles. Because we had run out of time, we did not expand this unit to explore the possible trench.

To the south of this unit was a third opened by Tim (with lots of help including from Cameron Howell on loan from the University of South Carolina). It too produced no clear architectural features but had a trench in one of its profiles. Again, because of time constraints we did not expand this unit to locate the wall-trench.

Further to the south and west (what was to be Grid 12), Ben and Luis excavated yet another 1-m unit on a suspected wall-trench building. Their experience was only slightly different from the others in the area. They did find a possible trench segment in the floor of the unit and also some possible trench features in the profile. They did not, however, find a well-preserved wall-trench with posts in the bottom.

Our experience with these four units, along with Will’s in Grid 9, left me puzzled and frustrated. The magnetic anomalies are clearly there and are so consistent and convincing that they can only represent Mississippian buildings. So why was it so hard to find the actual wall-trenches in the ground? I think there are a couple of reasons. The first has to do with the magnetic data we used to position the units. In each case, we used gradiometer data collected in 2008 at 1-m intervals. Because those data are essentially one quarter as dense as the .25m-interval data collected by Chet this year, the anomalies aren’t identified and represented as precisely. For these specific anomalies the lines of magnetic high that I considered to be the walls was considerably wider than the same lines on the .25m-interval data. That means there was a greater chance that we might actually miss the anomaly when we chose the location of a unit.

I think that explains our results in part. However, in all five areas in question we can see trench features in the profile even though we did not see them on the floors of the units. I don’t think we missed those trenches because of inexperience. I had a very well trained crew, several experienced volunteers, excellent graduate assistants—and I looked over everything. I think we missed them because they were pretty indistinct and difficult to differentiate from other soil changes apparent in the unit floors. I think the visibility of these features has something to do how long they were used, whether they were taken apart or left to decay in place, etc. We don’t really understand the details yet, but there’s a lot more to it. Most importantly, if I am right then the gradiometer may be more effective at find these kinds of buildings than traditional archaeology. Now that is something to think about.

The last area we attempted to explore was something I had called Fort de Soto. The anomalies are located just south of Mound F, which is a mound built during the time when de Soto visited Etowah. It appears to be some kind of large enclosure or possibly building. Outside of its association with a de Soto period mound, I had no other real reason to call this anomaly Fort de Soto. I will admit that I had hoped to start a friendly rivalry with Chris Rodning, Rob Beck, and Dave Moore who announced finding the earliest interior European fort at Fort San Juan.

We positioned one unit along the western wall and a second on a circular anomaly within the larger enclosure (on the N1219 line). I am disappointed to report that we found no architectural remains whatsoever, nor did we find any 16^th century European artifacts. So my hopes of scooping Rodning, Beck, and Moore will await another day. I am afraid that these excavations suffered from the same problem as those on the northern part of the site. We positioned the units using 1-m interval gradiometer data and did not have time to expand those units in hopes of finding the features that created the magnetic anomaly.

We wrapped up our excavations on August 1 and the next day cleaned out the field house and departed Cartersville. However, we were not done in Cartersville. We had left artifacts and flotation samples drying in the garage of the field house and all the units were still open at the site. About a week later, Johann, Amy and I returned to backfill and collect the artifacts and flotation samples. I rented a cargo van and the three of us spent a very long two days driving, backfilling, packing, driving and unloading. Matt somehow got out of the fun using the weakest excuse out there—his first day of graduate classes.

Now we have some 400 bags of artifacts to wash, sort, weigh, count, and rebag. We’ve also got about 20 more bags of soil to process and then some 50 flotation samples to sort and weigh. As we come across fun things or lean something new in the lab, I will do another blog post.