Pixel vs Pigment. The goal of Virtual Reality in Archaeology

Savage Minds welcomes guest blogger Colleen Morgan.

Post by Laia Pujol-Tost.

Archaeology has a long tradition of using visual representations to depict the past. For most of its history, images were done by hand and based on artistic skills and conventions. But the last fifteen years, we have witnessed 3D models take over archaeological visualization. It is interesting to note that while hand-drawn depictions tend to show human figures and seem to be associated with scenes of “daily life”, virtual reconstructions mostly show architectural remains and public spaces, usually devoid of people and objects. Yet, authors state that their intention is to represent the past.


My field of research is what we now call Virtual Archaeology, but I started investigating when we still talked about “VR applications in Archaeology”. I have seen it become mainstream and evolve; and I wonder why after almost twenty years of technological improvements and theoretical debate, virtual reconstructions are still empty. Especially in comparison with drawings. Do the virtual and the physical have implicitly different goals? Are they subject to different perceptions or expectations by researchers and/or audiences? Have they received different historical influences? Maybe technological capacities still play a role?

Let’s put some evidence on the table. I have reviewed a lot of bibliography, and in my early career I conducted a series of studies that took me to Rome, Ename (Belgium) and Athens. The conclusion in relation to the matter at hand was the following: both audiences and experts associated VR with the objective reproduction of reality, while the introduction of human characters (and objects) seemed more speculative. As a result, they considered illustrations were better for wide audiences, especially for children, and associated their production with non-academic professionals (illustrators and educators).

This situation arises from a confluence of causes related to VR and to Archaeology. In the first case, VR allows the representation of and navigation within 3D geometrical spaces. The Cartesian concept of the world is deeply rooted in the Western mind, and is best represented by computers. But VR also comes from the pictorial perspectivist tradition, which is arguably considered the closest to human perception. In addition, we now have the possibility to acquire the model directly from reality, by means of photogrammetry or 3D scanning. Hence the belief that VR represents the world objectively. On the other hand, modelled humans are problematic because they require a lot of computing resources but still seem “fake”. The “uncanny valley” effect contributes to the belief that humans hinder realism.

Technological capacity has been one of the historical justifications for not populating archaeological virtual environments. However, even now that I am building my own VR-mediated experience, I am not persuaded. Video-games display very realistic worlds with human characters and maximum interaction. Certainly, the entertainment industry mobilizes a lot of money and people. But we do have examples of archaeological projects supported by big investments in terms of time and/or human efforts. On the other hand, realism is not limited to visual effects, but is achieved thanks to the interplay of different elements (an overload of visual details, affordances for interaction, sounds, etc.), which generate a general impression of verisimilitude. Besides, there are other solutions, none of them new, such as procedural generation (of textures, buildings, etc.), Non-Photorealistic Rendering, and more recently, the combination of digital and video-recorded content. Yet, I have come to understand that people involved in virtual archaeological reconstructions mostly think in terms of visual accuracy.


To discuss the other set of causes, related to archaeology, let me tell you about {LEAP]. This project aims at building a theoretical and methodological framework for VR mediated experiences, based on the concept of (Cultural) Presence. In order to refine the concept and its translation into user requirements, last August I conducted fieldwork at the Neolithic site of Çatalhöyük (Turkey). VR applications share features with the original site (raw data); an experimental house built close to the site (immersivity); and illustrations of the settlement (presence or absence of humans). But VR applications also have specific capacities (dynamism and simulation) due to their virtual computational component. I wanted to see which ones were spontaneously used or considered important by experts when describing verbally and/or visually a past culture. What did I learn about illustrations and virtual reconstructions? According to the results, virtual reconstructions should: 1) show life and objects (like the illustrations but in a dynamic way); 2) distinguish between evidence and reconstruction (like the site); 3) be immersive and multisensory (like the reconstructed house); 4) allow natural navigation (at eye’s, not bird’s view); and 5) show temporal evolution and unusual perspectives (making the most of virtuality).

There is an obvious conflict between these expectations and virtual reconstructions. Moreover, Virtual Archaeology is now a mature area of research, for which an internationally acknowledged set of guidelines, the Seville Principles, has been established. Yet, several points (related to coherence between aims and methods, transparency, and effectiveness) are still not implemented in many VR projects. In my opinion the causes are to be found in archaeological practice. As debated in a couple of recent seminars, the dominant epistemological paradigms aim for the truth, understood either as explanation or description of the archaeological record. Objectivity and visual realism are one and the same. Nevertheless, there are other, very important paradigms that acknowledge the interpretive nature of Archaeology, and attempt to express it visually. Unfortunately, they are both confronted to the pressure of funding bodies, stakeholders and audiences, who demand the illusion of a (scientifically accurate) trip to the past. The consequence are hyperrealistic environments, where buildings are fully reconstructed (in spite of the availability of data), but human characters and material culture are (sometimes) introduced as a necessary embellishment.

So in the end my question should not be why are virtual reconstructions empty, but if we want to change this and how. How do we do our job, how do we contribute to science and society, against hype, inertia, and economic interests?

Dr. Colleen Morgan (ORCID 0000-0001-6907-5535) is the Centre for Digital Heritage Postdoctoral Research Fellow in the Department of Archaeology at the University of York. She conducts research on digital media and archaeology, with a special focus on embodiment, avatars, genetics and bioarchaeology. She is interested in building archaeological narratives with emerging technology, including photography, video, mobile and locative devices. Through archaeological making she explores past lifeways and our current understanding of heritage, especially regarding issues of authority, authenticity, and identity. She received her PhD in Anthropology at the University of California, Berkeley and her BA in Anthropology/Asian Studies at the University of Texas. Since that time, she has worked both as a professional and academic archaeologist in Turkey, Jordan, Qatar, England, Greece, Texas, Hawaii and California, excavating sites 100 years old and 10,000 years old and anything in-between. She remains deeply interested in excavation methodology, high falutin’ theory, interstitial spaces, skeuomorphs and good bourbon.

One thought on “Pixel vs Pigment. The goal of Virtual Reality in Archaeology

  1. Very nice! You raised some important points. I would like to see someone use translucency or shadow effects to distinguish between what exists and what is predicted or reconstructed. Showing people as shadows might eliminate a number of the reasons our reconstructions are often so empty.

    Also, a chapter written by Hughes, Little, and Ballantyne (in Tourism and Archaeology, 2013) observed that visitors to heritage tourism sites were more concerned with feeling like they had an authentic experience than they were with accuracy of that experience. This suggests to me that the obsession with perfect accuracy comes from our end and may not be shared by our audience. If you have evidence to the contrary, or do not think this applies to digital reconstructions, I would love to hear about it.

Comments are closed.