Humans are in many ways defined by their social interactions with others. Think about it: much of what we do hinges on peoples’ use of social networks to facilitate solutions to collective problems like resource shortfalls, information acquisition and dissemination, and political or conflict resolution. I don't think it's a reach to say that humans live and die within the context of social networks. Material culture—think clothing, artwork, jewelry—is often used to advertise important information about one’s personal and group identity within a social network. One way that people do this is to adopt or manipulate the style of an object to distinguish themselves from, or more closely identify themselves with, other people. (If you've ever put on the merchandise of your favorite sports team, you know what I'm talking about.) Archaeology is uniquely equipped, through the study of artifact style, to track this most fundamental aspect of human behavior across vast expanses of space and time.
Way back in 2016, the Archaeology Program at UNCG organized an Ashby Dialogue around social networks generally, and social network analysis, or SNA, specifically. SNA is a methodology based in graph theory that allows us to visualize and quantitatively describe relationships between social entities (people, organizations, tweets, etc.) and the features that bind them together (kinship, shared identity, common interests, etc.). This approach has gained significant traction in archaeology, as artifacts can be used as representations of social entities and the stylistic attributes of those artifacts can be used as proxies for social proximity. The idea here is that if two people create or use a pot with the same or similar decorations, they must share, at least at some level, aspects of their social identity with each other, whereas people that create or use pots with very different decorations probably share little or no part of their social identity with each other. As part of our Ashby Dialogue, I was asked to lead a discussion on the use of SNA in my period of interest, the Paleolithic. I quickly realized, however, that while there was a ton of research on Paleolithic social networks, only one study had used SNA. How, then, might we apply SNA to the Paleolithic more broadly?
Some time soon afterwards, I attended a luncheon organized by UNCG's RISE Network. The purpose was to connect STEM researchers in order to facilitate collaboration across academic fields. At one point, I found myself sitting at a table with Dr. Jing Deng, a faculty member in the Department of Computer Science. The attendee list provided to us listed his area of expertise as—you guessed it—social network analysis. Jing showed a great deal of interest when I described the sort of data we Paleolithic archaeologists work with. A small exploratory analysis supported by UNCG's Undergraduate Research, Scholarship and Creativity Office (and the hard work of then-anthropology and computer science students Amanda Chase and Nathaniel Arnold) convinced us that SNA could be a valuable tool for reconstructing Paleolithic social networks. While any time period is ripe for SNA, we decided for two reasons to focus on the Magdalenian of western and central Europe (ca. 20,000 to 14,000 calibrated years before present). First, Magdalenian cultures are famous for the creation of a staggering array of art, including rock paintings, carved and engraved bones and stones, and perforated bones, minerals, and fossils—all of which encode potentially important information about social identity. Second, Magdalenian peoples were confronted with rapidly changing environments as Europe, and the rest of the world, emerged from the last Ice Age. These volatile circumstances no doubt placed stress on the social networks of folks moving into landscapes only recently vacated by ice sheets. To estimate social "distance" between Magdalenian sites, our initial analysis used frequencies of shared artifact (stone tools, beads, engraved bones) and raw material (stone, bone, fossil, ivory, shell) types. As we worked through this, I began to wonder if it was possible to measure social distance not only by the frequencies of artifact types, but by the stylistic motifs of the engravings on individual artifacts. Could we figure out a way to compare the engravings objectively in order to determine which artifacts were stylistically more similar?
I spoke with Jing about this, and he introduced me (at yet another RISE event) to Dr. Minjeong Kim, another faculty member in the Department of Computer Science. Minjeong's research focuses on image analysis, particularly medical MRIs and CT scans. She was optimistic that a computer could be trained to extract the engraved patterns from 2D digital images of artifacts and overlay and compare the shapes of the extracted patterns to gauge their stylistic similarity. Among the most common engraved Magdalenian artifacts is called a "perforated disk." These small, flat disks were crafted from stone, bone, or ivory and often include engravings of geometric designs and/or animals. Just like humans today make choices about what to wear in order to (sometimes unconsciously) signal something about themselves, Magdalenian people likely created and used perforated disks to signal aspects of their own identity to others. It just so happened that a good friend of mine, Dr. Rebecca Schwendler, had looked at more perforated disks than just about anyone else in the world. (In fact, it was Rebecca's paper on Magdalenian art and social networks that I discussed way back in 2016 as part of the Ashby Dialogue.) We also needed someone to help reconstruct the ancient environments of post-glacial Europe. Enter Dr. Chris Nicholson, a GIS whiz with tons of experience working with paleoclimatological data. With Chris and Rebecca on board, we had an extremely interesting project (at least to us) and a talented research team that was fully capable of pulling it off.
In 2019, we put together a research proposal to the Archaeology Program at the National Science Foundation. After two years, three submissions, and tons of valuable input from NSF's hard-working review panels, we were awarded ~$230,000 in 2021 for a three-year project entitled "A network approach to Magdalenian social landscapes." As of this writing, we are a little less than a year into the project, and I am very excited to see where things go from here. If you are interested in more information, including updates on our progress, please visit the project's homepage.
