Augmented reality, artificial intelligence, and... classical studies? They don’t appear to go together. And it’s not just because these disciplines are eras apart; it has much more to do with public perception. An archaeologist continues to fit the role of either the lovable, eccentric bookworm or the Indiana Jones type, with a Fedora and whip, but definitely not that of a programmer with a laptop and smartphone. Because my second academic degree is in the Humanities, and more specifically in the area of Egyptology as well as of the philology and archeology of ancient Greece and Rome, I follow the issues of Virtual Heritage and the Digital Humanities with great interest. Furthermore, I’m quite sure that augmented reality and AI in archeology and the research of ancient languages will make new findings possible in the near future — or at least make sure that a much broader audience can get involved with these disciplines.
To the extent that human thought can be considered a tool for information processing, it is possible to transfer this thinking onto devices and model it. Through its specialized applications, artificial intelligence can (still) only be employed in areas for which we conceptualize its use. For modern chess computers, opponents with an Elo rating of more than 2500 are just a warm up, but a simple Google search is impossible. And AI that is meant to analyze an “ancient language”, will do just that. It can sometimes be confusing to already have artificial intelligence all around us, as in most cases nobody would think to name everything “artificial Intelligence". Or as John McCarthy said, “As soon as it works, no one calls it AI anymore.” This is certainly no different in Egyptology or similar sciences.
Now imagine an entire warehouse full of Egyptian papyri that were found either in larger contexts or only in fragments, without further context. Naturally, academic staff (which is often non-existent due to lack of funding) could invest countless hours or use crowdsourcing projects to attempt to organize, catalog, read, interpret, and linguistically classify the existing text fragments. Or an AI program could be designed beforehand for this task, which, based on a predetermined ontology and appropriate subjects of comparison, can achieve the desired results much more quickly and at just a fraction of the cost. Because once digitized, a corpus can be used for further research and conduct linguistic studies that, due to their complexity, can connect entire groups of researchers.
Let’s come back to our example of the warehouse full of papyri. After an AI program has established which excerpts are useful and fit together with the others, we can determine how often a certain case occurs, such as the Middle Egyptian genitive, based on the linguistic classification of characters (such as hieroglyphics). By expanding the extent of the passages being researched, we could get a picture of how the Middle Egyptian genitive originated and how it evolved. So AI can significantly help us to collect, translate, and analyze the literary corpus, and not only Egyptian, but even the Akkadian. And these are not just lofty academic aspirations. If you think one step ahead, it may one day be possible that tourists to Egypt can take pictures of preserved hieroglyphics when visiting historic sites—and have the translations on their smartphones immediately. The scope of such a development carries enormous implications, because it would ensure that the barrier between us and these very ancient cultures would melt away.
I am convinced that technologies such as AI, big data, or machine learning will make it possible for us to interact with the heritage of ancient cultures in different and new ways. We will be in a position to experience the way of life of earlier civilizations. One feasible way already available to us is the visualization of ancient temples and tombs by means of virtual reality. I have myself worked on a number of projects in this area, and I am convinced that VR images not only encourage you to visit Egypt yourself in order to see the historical sites, but can also help scientists to better visualize the placement and design of concrete inscriptions and other features.
Archaeology is still being taught by means of somewhat unclear black-and-white photographs But if you compare these to the VR images of the famous Tutankhamun burial chamber or of Ramses VI, then one realizes how vivid and realistic (even without VR glasses) the impressions are, and how clearly even narrow hieroglyphics are displayed. Of course, VR images could never be a replacement for a visit to the historical sites and archaeological digs. But you can arouse curiosity in everyday people to travel to Egypt themselves, and help scientists with the availability of tangible representation of places that would otherwise be represented as hazy photographs.
I’m not claiming that "artificial archaeologists" (computer programs) can operate like human brains, nor that such computer representations should be isomorphic to "mental" states. I’m not pretending to simulate myself when I undertake archeology, but rather to create something else; in a sense, to expand human possibility to reach predefined targets (faster). I find it a rather exciting task to understand how intelligent behavior in archeology is (or will be) possible. The goal is not to simulate intelligence, but to pose real (natural or artificial) archaeological questions. And I believe that we are already very close on their heels, as evidenced by the many current projects and startups that are currently active in this exciting field.