Regarding the (A?)Symmetry of Coil-Built Ceramics

In an effort to further capitalize on our 3D data, I am beginning to explore the topic of symmetry. During a recent brainstorming session with some talented colleagues, we began toying with the notion of creating a fully symmetrical surface using a single section (mesh sketch) from one of our 3D scans, subsequently revolving it 360-degrees around a central axis (vector) to create a solid surface. This surface (nominal data – silver) could then be contrast with the mesh (scan data – blue). The plane in the image below intersects the vessel at its widest point (click here to learn more about how we identify the widest point).


With regard to our methodological toolkit, understanding that a section (or profile) of a coil-built ceramic vessel cannot–in most cases–be mirrored or revolved to accurately represent the entirety of the artifact is important. This is an observation that may well extend to vessels produced on a wheel, although in a comparison of the two, those built on a wheel are (assumption) more likely to have a higher degree of rotational symmetry – however, this is not to say that coil-built ceramics are not symmetrical (or is it?).


I find myself curious whether it might be possible to assess varying degrees of skill (or expediency?) in coil-built ceramics based upon quantifiable measures of rotational symmetry. Issues of symmetry also have implications for our study of 3D morphometrics, where we continue to learn which tools and approaches are most appropriate for data from a cultural (rather than biological) system.


The deviation between the surface and the scan data were calculated, further illustrating the variability in the vessel profile where the maximum/minimum gap distance between the surface and the scan was found to be 4.9+mm in each direction.


Further, when looking only at the single profile along the plane, the mirrored surface was compared with the scan data. This has important implications for our study of morphometrics, and helps us to better illustrate that each section (profile) is unique. The upper (green) area of this illustration sits along the planar surface, and is the section that we used to create the nominal data (top right).


It is also possible to rotate the nominal data (green vessel radius above) around the central axis (vector) to calculate the deviation of the scan data from this singular (widest) profile and create an overlay of the deviation from the nominal data around the 360-degree rotational axis. The results (keeping in mind that this is a single vessel) point to a large amount of deviation in the vessel profile, further demonstrating that the (assumed) rotational symmetry of wheel-built ceramics may not extend to coil-built ceramics.


While this instance is representative of the symmetrical variation in only one vessel, I look forward to incorporating more samples, and am beginning to explore how these data might be couched within more theoretical discussions.

From a methodological perspective, our decisions regarding landmark/semi-landmark placement might shift during and after a study of vessel symmetry, depending upon what questions that we are endeavoring to ask of the data. It may also be an interesting exercise to calculate the deviation between the final landmark/semi-landmark coordinates extracted from our scan data with a rotationally-symmetrical (nominal) surface produced using data from that same vessel.

Perhaps a suitable test of theory might be an assessment of perceived value (derived from the idea that symmetry = attractiveness?), where a comparison of fine wares and utility wares may yield some noteworthy results. Additionally, it might be profitable to contrast the rotational symmetry and–potentially–fluctuating asymmetry of ceramics found in elite/commoner burials.

If nothing else, I’m finding that the topic of (a?)symmetry in the shape/form of ceramics, while complex, may have something meaningful to contribute to archaeological dialogues. This has certainly given us plenty to think about. More to come on this topic.

Written by zselden

Selden (PhD, Texas A&M University, 2013) is a husband, father, US Marine Corps veteran, cyclist, kayaker, backpacker, hiker, climber, fisherman and general all-around outdoor enthusiast. His research is focused at the confluence of archaeological methods and digital technology, and he is particularly interested in the application of 3D technologies to archaeological problems, geometric morphometrics, network analyses, predictive modeling, archaeological theory, and archaeological science.


  1. This is interesting research, and it looks like you are developing the tools to actually be able to measure the relevant attributes. Having analyzed and made both coil-built and wheel-thrown pottery, I think you are right that as a general rule wheel-thrown pottery had greater rotational symmetry, but possibly only while still on the wheel head. The act of lifting the piece from the head often resulted in fairly significant warpage, such that the rim moves from circular to ovoid. Especially in cases of wheel-thrown production potting, the potters were turning a lot of pieces in a hurry, and they often were not too concerned with perfection. I’m just talking off the top of my head.

    For coil-built pots, it might be interesting to compare rotation symmetry (as a measure of skill) against other possible measures, such as vessel weight per volume capacity. The expectation is that the more accomplished potter would make thinner, more uniform pots. Again, I am just thinking aloud.

    Nice work. Thanks for sharing. Chris


Comments are closed.