This comes from an unannotated, unpublished paper that I presented at Kalamazoo's International Congress for Medieval Studies, "Digital Art History" session, in May of 2010. I'll try to get the ppt up at some point.
[start slide] Thomas Kuhn, in his famous 1962 book ‘The Structure of Scientific Revolutions,’ suggested that new discoveries and technologies every few centuries have the potential to create new paradigms for thought, research, and application – the new and easy accessibility of computer software for the visualization of historical architecture and environments offers just such a shift, right now. We are still assessing (and indeed, living) its impact and potential, but it is clear that…
[props slide #1] 1) technology is changing the way we record and represent material and visual cultures
2) architectural space and surfaces are essentially quantitative, and processes of quantification have been simplified through automation
3) digital applications have allowed the creation of virtual, hyper, and alternate realities.
[props slide #2] 4) technology has created modeling processes that help us to better understand the interaction between people and their environment, through simulations
5) collected or processed data can now be manipulated and disseminated very inexpensively.
6) and the flexibility of software options means that most any theoretical position can be brought to bear on a digital model –theoretical inquiry into the natures of representation, space, materiality, and agency are challenged and defined by complementary advances in graphics, geoinformation and mapping, environmental or artifact laser-scans, and artificial intelligence.
What does this mean for historians interested in visual culture?
[plastico #4] We might start by quickly considering our forebears: in 1931 Mussolini commissioned Italo Gismondi to build a 1:250 scale-model of Rome as it appeared at its greatest size, during the reign of Constantine. How does this model, visible even today at the Museo della Civiltà Romana function for the viewer? Set on a low table, and covering several square meters, one looks at the tiny features of Gismondi’s model from a nearly omniscient bird’s eye perspective, or at an oblique angle, if you bend down and steady yourself to get a line of sight through streets or open spaces, like the Forum Romanum. Close up photography of the model has been a common place in publications, particularly exhibition catalogues and advertising literature. Obviously only exteriors could be represented, at a scale that distorts even the larger details. Locations are wholly approximate. The model is an artistic summation and interpretation of Lanciani’s famous topographical catalogue of the ancient city, finished in 1907 – as such, it was the product and presentation of all that acquired knowledge, rather than an incubator/generator for new questions. The map, in every way, was a situated cultural product rather than a correspondence to some historical actuality.
[rome reborn – video #5 and img #6] More recently, James Packer (UVA) and Bernie Frischer (UCLA) have created the Rome Reborn project, a new digital model of the city available in both public (free) and private (licensed) versions. Rome Reborn is in many ways a new iteration of Gismondi’s project. Indeed, the new Rome was not built from marble, but from a laser scan of Mussolini’s Plastico model, upgraded with consideration of a vast archaeological scholarship produced in the past 70 years. Architectural illustration thus blurs with visualization. For the licensed viewer, the model exists as a hyper-real alternative reality, accessible through a video-game like interface that provides fly-throughs and animations for TV and documentaries. For the end user of the free version online, Rome Reborn exists in Google Earth as 3 dimensional images, which project from the face of Google Earth’s georeferenced map of the city. Both versions potentially deceive the viewer through the metaphors of photography and cinema, giving a false impression of mastery, knowledge, and mimesis. The images or animations are still the products of an artistic process – the artist chooses what she will represent, etc. Is this sufficient if our primary objective is educational, to give the viewer a hypothetical sense of appearance, scale, and spatial relationships in the ancient and contemporary cities? Like the Plastico model before it, Rome Reborn is properly conceived as a visual summation of contemporary research, a vehicle for educating the public rather than for further analysis.
[Byz 1200 slide #7] Projects like Rome Reborn or its less well developed medieval equivalent, Byzantium 1200, are almost already a thing of the past – on such vast scales, they’re extraordinarily cost and labor prohibitive – and like a kid with a trainset, what do you do with the thing after you’ve built it? Since maybe 2005, there has been a justified push back against the use of digital models for visualizing historical cityscapes. Yet further developments with simulations, which optimistically hope to provide a window for “dirty” phenomenological views of the past, are more challenging and theoretically fraught, though they too continue to give us new experimental platforms for studying the past. I’d also like to suggest that the utility of even very analytical, quantitative, and manifestly artificial models is still clear for certain types of applications.
[VISTA slide #8] Historical Preservation is one such application where cheap, fast measurements are a desideratum. Relatively inexpensive, quite portable laser scanners (like LiDar) have enabled conservators to make high resolution three dimensional models directly from the buildings themselves, as seen here in the East window of York Minster Abbey, scanned and rendered by Vince Gaffney, from the University of Birmingham’s VISTA Center for Visualization. These 1:1 digital models are often commissioned by heritage groups in order to guide programs of conservation, repair or restoration; historians are frequently hired as consultants, or can use the models independently after they have are published. For a lot of sites, the collected data is already there if you know to look for it.
[zink slide #9] On the other hand, it’s not that difficult to collect your own data, either directly (with laser scans) or indirectly (with drawn plans or other materials). Stefan Zink, a graduate student from the University of Pennsylvania, surveyed the Temple of Apollo Palatinus in Rome, and was discouraged by the paucity of a nearly barren ruin with a massive temple podium but few column bases and fragments of architectural sculpture. Yet he realized that the careful laser measurement of those fragments, when drafted in AutoCAD and digitally multiplied and reassembled, could yield the dimensions and appearance of the whole temple. This led to the reconstruction of a place that no one had seen for over 1600 years, as well as consideration of the temple’s surprising design, which in some ways is almost anachronistic in its conservatism.
[DIY #10] Similar reconstructive approaches have been used to piece back together fragmented sculptures scanned with photogrammetry, with equipment as simple as a laser-line pen (starting at about $30), a webcam, and free imaging software from david-laserscanner.com.
[pickett #11] Some of my own work has also taken advantage of the quantitative potential of easy to learn programs like Sketchup and AutoCAD, not for reconstruction, but for enhanced economic understanding of the circumstances of constructoin. Using photographs, plans, and site visits, I made three 1:1 models of well studied and preserved Late Antique and Medieval buildings in Turkish Cappadocia, in order to facilitate the calculation of the building’s material volumes (like cubic meters of brick or stone, for instance). This volumetric data was then interpreted using rates for labor tasks (like building-days per cubic meter of wall or quarried stone) found in anthropological data or artisanal manuals published before the Industrial Revolution. The construction of the model itself was an iterative rather than a summarizing process: each step required that new and unpredicted sorts of data or measurements be collected from the original site. Eventually, this produced an estimate of the total person-days required for building the biggest things on Cappadocia’s landscape. With this kind of information spread out for monuments distributed throughout space and time, we give ourselves a new data set to think about bigger issues like the interaction of people with their landscape, human or material resource capacities of regions, demographics, labor organization, agricultural production, or transportation – these topics are otherwise very much underrepresented in the textual record. Also, you probably noticed that my model isn’t shiny and realistic like the others – it calls attention to its facture. This is the diverse, egalitarian aspect of visualization that I think is fascinating. For my project’s objectives, an immersive, highly imitative environment just isn’t necessary, or even desirable.
[holod vid #12 img #13] But immersive environments can be quite helpful, if the question suits the approach. A group of students working with Prof. Renata Holod at UPenn were interested in the effects of medieval lamps and lighting in the Mosque at Cordoba. Whereas Rome Reborn was designed to represent ideal conditions on one particular date (June 21, 320), the Penn Cordoba model was designed to be flexible, and to handle diverse external and internal variables. First, an AutoCAD mockup was made and textured with Maya (to make surfaces like rugs and marble appropriately opaque or reflective). A variety of lamp-types were laser scanned and then digitally reproduced to outfit the interior of the building. The designers took care to assess the changing effects of different weather, types and quantities of lamps, their arrangements, luminosities, and even wick-types or positions (which were all found to contribute to dramatically different interior environments). Because these lamps hang over spaces for congregational prayer, it was also important that the effects of lighting be considered from the perspective of worshippers in the motions of prayer. Patterns of circulation and vision within the building were also taken into account, producing a range of scenarios for the use and vision in the interior. Though one can easily imagine the possibilities for the application of these ideas to churches on the pilgrimage roads, with questions about lighting, privileged positions for viewing art or reliquaries or sculpture, or pilgrims’ access to crypts, this type of approach remains largely untested on western medieval architecture to my knowledge.
[Manzikert #14] Simulations can handle increasingly large amounts of data and sophisticated situations and behavior – we as historians should feel free to think big with these types of applications, with the knowledge that the problems arising from complexity in simulation are the same problems puzzling physical and computer scientists. Google Earth and Microsoft’s Bing, for instance, are engaged in a fight as of June 2010 to make historical time and imagery a component of their mapping services – modeling time in a GIS is, as it happens, a prime desideratum of contemporary archaeologists working with these technologies. On the other hand, Princeton and Birmingham’s Manzikert Project is an effort to model the impact of 100,000 man armies moving across the eleventh century plateau of Anatolia – its April 2010 workshop attracted not only historians and archaeologists, but specialists in palaeo-ecology, game theory, chaos theory, artificial intelligence, information technologies, and so on. As historians, we’re still figuring out how to make good use of all these options – from the highly quantitative and analytical to the interpretative and perceptual - to understand both their potential and their limitations.
[final slide] One could make the fair criticism that all of the models I’ve discussed are sleek and seductive - enhanced by ever more sophisticated software packages for texturing and the addition of environmental details - but they are still ultimately sterile, with no intrinsic or substantial relationship to the past that they seek to represent. Images are indeed deceptive, and concern for their public and commercial consumption is justifiable. But I would argue that applications for rendering space and surfaces or behavior and perspectives are not confounded by theoretical critiques, of space or representation or agency, but in fact can work as complements, to challenge and redefine the conceptions we have of our own discipline and objects of study. What’s important is that we learn to allow technology to complement our research questions: lucky for us, the declining cost and learning curves of software packages, the growth of university computing and visualization centers, and the common problems of complexity and scale faced by researchers across disciplines, all means that the use of new technologies for the study of architecture is far easier than it was even five years ago. Three dimensional places and simulations are not substitutes for historical reality, or paths to easy answers: this is the mythos of technology that we have to forget and replace. But at best, the fabrication and use of visualized environments provide novel frameworks for testing hypotheses, and can produce new questions and new perspectives for the study of medieval architecture.
