Paper delivered on April 24, 2014, by Donald H. Sanders, PhD, at the 79th annual meetings of the Society of American Archaeology, Austin Texas.
posted here September 18, 2019
Archaeology demands exactitude, copious documentation, rigorous analyses, and prompt dissemination to peers. That rarely happens. Archaeology is also about understanding the distant past, which is difficult, because we weren't there. What we know for certain is that the lives of the ancient people we are trying desperately to know happened in 3D, in color, and as a continuous set of actions (see, for example, our digital animated reconstruction of the Hellenistic Vari farmhouse above; this is a large video, you might need to restart it once to see the full effect). Therefore, it would seem logical to study the remains of those past lives with similar parameters, rather than as the disconnected, 2D, black and white images that are still too common in our excavation reports, teaching materials, and Websites.
20 years ago, I stood before another SAA audience and began my talk this way: Welcome to the archaeological Holodeck, in the totally immersive virtual environment before you, you have the ability to walk around in any ancient structure you wish, move artifacts around, watch virtual inhabitants, study their behavior, even destroy all or part of the site under various conditions. Too fantastic? Fine for the special effects masters of science fiction, but not for scholarly research? We'll see how close reality is to fiction at the end of this talk.
What if we could understand the nuances of daily activities in an ancient and foreign culture and lifestyle just that specifically? What if, as with a Holodeck, we could have a 3D body scan made as the basis of a personalized computer model? What if you could then don your digisuit--your wearable time machine so that your avatar could explore any excavation, site, or building as if in a first-person experience? You could wander through different time periods, each fully reconstructed and fully populated, in interactive stereo 3D with all the senses activated. With a flick of a switch you could compare the actual remains with the virtual re-creation of the past. Flick another switch and watch the virtual sun move across the sky or hypothesized wall heights change and doorways move, as you test various scenarios against the excavated evidence.
You could take pictures of the virtual place and interact with other timetraveling tourists or scholars (the image at the left shows Donald Sanders being interviewed by Dutch television inside the virtual world of Learning Sites Northwest Palace, Nimrud; hover over to enlarge). Fellow archaeologists could actively debate whether their interpretation of the past matches what is found in the archaeological record. This would not be your typical Oculus rift video game. As I asked two decades ago, how close are we to being able to do just that?
Today I'll discuss some visualizations methods of the past, the sacred triad, alternative visualizations of today, and some not-too-distant future tech.
How archaeological data has traditionally been illustrated has affected what we know about the past. Since the 18th-century beginnings of archaeology, we have relied mostly on drawings, photographs, charts, and graphs. But before then, long before then, there was stone. When the Sumerian prince, Gudea (c. 2125 BCE), wanted a new temple built, he depicted his vision using a simple plan (see the image of the temple plan on Gudea's lap at the left; hover over to enlarge). A plan was chosen because it symbolized the beginning of the construction process and could be reproduced fairly precisely on the ground.
But Gudea was depicting future architecture, not past architecture. Similarly, when the Egyptians, Assyrians, Greeks, and Romans depicted architecture (as room decoration, to record battles, or show foreign lands; see examples at the left; hover over to enlarge), they were usually illustrating their own present--for storytelling, education, or enjoyment--not trying to understand a distant past. Yet, the methods chosen to represent their built environment were consistent--2D plans and elevations (sometimes with a perspective view tossed in to provide a sense of three-dimensionality in an otherwise flat medium).
Specifically documenting the history of architecture began at least by the 13th-century work of Villard de Honnecourt, a French traveler with a keen eye. His drawings both aided contemporary construction and chronicled past styles. Still, the use of plans and elevations continued as the preferred means of representation (see an example of his drawings from c. 1230--of Laon Cathedral--at the left; hover over to enlarge).
Archaeologists traditionally record architecture using precisely the same methods, for documenting space and for delineating the context of other artifacts; thus perpetuating a reliance on plans and elevations. This continuity persisted primarily because (1) most of the early archaeologists were themselves trained initially as architects; and (2) the development of archaeology as a discipline in the late 18th and early 19th centuries coincided with the rise of the Classical revival. Architects needed accurate renditions of ancient Greek and Roman building details in order to construct contemporary structures, and they needed those depictions in the same formats they used to translate their designs for use by builders; archaeologists obliged by drawing their architectural finds in ways they knew best and were easy to use by architects--plans, sections, and elevations, which became known as the Sacred Triad (see the late 18th-century examples of James Stuart & Nicholas Revett drawings at the left; and how those details all find themselves duplicated in the early 19th-century St. Pancras New Church, London, by Henry W. Inwood; hover over the images to enlarge).
Then, during the 19th century, a radically new technology emerged. Photography offered archaeologists a more efficient and effective means of recording their work and, at the same time, providing duplicate images for mass distribution (much like the promise of 3D graphics today).
By the 1840s-50s, travelers were using the new medium to document their discoveries, that is, newly found architecture and foreign cultures. But professionals resisted accepting it into normal archaeological fieldwork and reporting, for many of the same reasons that interactive 3D graphics have been slow to become widely adopted today: the equipment is awkward and expensive and breaks down frequently; the images can't be trusted; and too few people know how to use the technique adequately (compare the 19th and 21st documentation technologies at the left; hover over to enlarge). For a century thereafter, photography was little more than a way of illustrating monuments, enlivening reports, and providing visual aids for fundraising efforts. It wasn't until the mid-20th century, when Mortimer Wheeler insisted that site photographs should reveal every detail of excavations as they proceeded, that archaeological photography finally made the transition from mere snapshots to scientific recording.
The use of interactive 3D graphics in archaeology has followed a similar, but compressed, trajectory. Too many archaeologists used CG visuals as merely eye candy, just as was the case with early the use of photographs, opting to illustrate their publications, teaching materials, and research with the same 2D sacred triad that has depicted architecture and artifacts for millennia. There is an inherent oddity about this persistence, especially when archaeologists are trying to record settlement features that are by their very nature 3-dimensional and can only be understood through the action of moving through them (the fact that I'm presenting here using 2D PowerPoint slides enhances that incongruity).
Today, however, the interactive worlds of virtual heritage practitioners have become more widely used for hypothesis testing, fieldwork documentation, classroom exercises, museum displays, and augmented-reality-based tourism (see our digital reconstruction of the mid-9th century BCE Northwest Palace throneroom, Nimrud, Assyria at the left; hover over to enlarge).
Further, the methods of human geography, environmental psychology, agent-based modeling, and semiotics have shown us that, despite the inclination to suppose that archaeologists can't do "ancient anthropology," we can indeed extract from ancient architectural remains the residuals of ancient human behavior. We can then quantify certain behaviors in those ancient spaces with some degree of confidence. Thus, today we have at our disposal viewshed maps, privacy spheres, visibility graphs, sensory overlap zones, height maps, circulation diagrams, shape and material distribution diagrams, and the discipline of virtual heritage help us integrate these analytical tools into our fieldwork, teaching, and publications. By integrating these new image types with interactive 3D worlds, virtual heritage has evolved into a global community of practitioners, university labs, and professional organizations whose methods invariably lead to new insight into the past unobtainable through reliance on the Sacred Triad alone.
For example, the citadel at Nimrud, Iraq, an ancient Assyrian capital city for several hundred years from the 9th century BCE onward, has a complex architectural history. Traditional site plans have been used for over 170 years to try to make sense of this development without much success at providing a useful visualization mechanism to present intrasite changes (see the plans of the citadel at Nimrud, Assyria, modern-day Iraq at the left created from the 1850s [center], to 1970s [left] to the 2010s [right]; hover over the enlarge).
Our digital publication will move well beyond the Sacred Triad by integrating virtual reality, rudimentary AI, and the Unity game engine (see the video snippet extracted from our virtual world at the left; this is a large video, you might need to restart it once to see the full effect). We can then allow researchers to walk through the site and dynamically see the different eras of construction and how the citadel changed. Our mostly intelligent avatar will act as an intermediary between the user and the complex datasets used to correlate the excavated evidence and support our reconstructions.
Another example: we're all familiar with how certain bits of evidence about the ancient world exist now only in 2D images since the actual objects no longer survive. Often the depictions don't match one another causing the historian some difficulty in determining what the original objects looked like or how they functioned.
Such was the case for these vessels that played a critical role in ancient Dionysian rituals (see the video above; this is a large video, you might need to restart it once to see the full effect). The techniques of virtual heritage once again provide helpful insight. Hypotheses can be tested quickly, mockups can be devised and tested for structural integrity and design appropriateness. Multiple interpretations can be shown and the new visualizations introduce researchers to more appropriate tools for studying their data.
The huge Nubian city-sanctuary of Jebel Barkal has a building history of 1000 years, as indicated by the nearly 100 temples, palaces, pyramids, and kiosks so far identified through 100 years of excavation.
So, after nearly 2 decades of VH, we know it works because there is:
But what's really important is that VH helps us do our best work. And anything less than our best work would be unfair to colleagues and the general public. Interactive virtual worlds and accompanying behavioral science theory are not all we have at our disposal these days. We can add computer vision, machine learning, social media, and cloud-sourcing to envision a not-so-distant future for archaeology that is very different from one dependent on the sacred triad and static photography.
For example, the free and open-source REVEAL all-digital excavation documentation and recording package may be the only software that fieldworkers will ever need. The software can automatically build georeferenced virtual worlds of sites, architecture, and trenches as the work proceeds and all data about the excavation and all finds are integrated with the models and each other. Uploading new data, taking notes, doing hypothesis testing, generating interactive 3D models, querying the evidence so that you can readily adjust your dig strategies, and preparing publishable output ALL happen in real time, as the excavation unfolds. No laser scanners, no survey equipment, no architects, no waiting until the end of the season for the results; and all performed with your hand-held devices. The future is now, as REVEAL is in use at sites all around the world.
REVEAL and other real-time photo-based field recording and documentation systems allow unprecedented access to automatically generated and fully textured interactive models. Online social networks, linked virtual worlds, and unprecedented methods for innovative hypothesis testing can extend those capabilities to create a globally connected real-time research community.
At the start of my talk, I mentioned the digitsuit, the Holodeck, and science-fiction-sounding scenarios that archaeologists might adopt to help not simply understand the past, but actually simulate reliving the past. I trust that these forecasts no longer seem so far fetched. The past happened in 3D, in vibrant colors, with lots of people engaged in all kinds of messy and complex activities that have left residual echoes in the shapes, patterns, and circulation routes of the ancient built environment. We have the ability and the tools to capture those echoes and learn all about the actions that initiated them.
We can create near first-person experiences of past places, cultures, and events. Virtual environments can become portals to distant pasts complete with AI-based avatars playing and replaying various scenarios until the results of their actions match the excavated record. Our understanding of the past is now limited not by the questions we ask, but by the means we choose to visualize the answers.