MEANING IN MOTION
A Personal Walk Through Historical Simulation Modeling at UCLA
Professor Architecture and Urban Design, UCLA
Director, Experiential Technologies Center, UCLA
For the Romans, walking, thinking, and memory were inextricably intertwined. A walk through a city was equivalent to, or even preferred to, reading a text.1 Buildings, statues, inscriptions, and urban occupants all operated as signifiers that elicited potent associations. Even unarticulated spaces sparked memories. The Romans imbued each place with a spirit or genius whose identity was shaped by past and future actions. Describing a peregrination through Athens, Cicero writes, “wherever we walk, we set foot in some history.”2 Certain ancient rhetoricians memorized lengthy speeches by imagining moving through environments filled with content-laden objects encountered in a carefully choreographed, though variable, sequence.3 In the real world, individuals passing through ancient cities themselves participated in content creation as their actions occurred within specific stories and shaped evolving metanarratives. Pedestrians’ speed of movement, costume, associates, and selected paths overtly signified social status, purpose, and mental states. The modulated progress of a senator and entourage moving to the Forum, or the rapid progress of a single freedman on an errand drew from and enriched collective stories and identity. Over time, the readings of movement became prescriptive, informing subsequent developments in the design of buildings and cities.
When colleagues at UCLA began to discuss using digital technologies for the study of Roman architecture and culture in the late 1990s, we agreed that the potential to study ancient movement was one of the most compelling attractions of the new media. Over the last decade and a half, scholars and students at UCLA’s Cultural Virtual Reality Lab (CVRLab) and its successor, the Experiential Technologies Center (ETC), have continuously interrogated this topic. As with all digital explorations, the undertaking has been shaped and enlightened by parallel developments in scholarly approaches and technical capabilities. The elucidating pas de deux between research and digital tools continuously shapes the questions we ask, the methods we use, and the resulting interpretations. The content and digital presentation of Visualizing Statues in the Late Antique Roman Forum are thus informed by their situation within a historiography of phenomenological research relating to the ancient world and to new modes for exploring the meaning in motion.4
The following is a personal walk through the evolving challenges encountered in harnessing digital technologies to investigate kinetic issues in Roman environments. Other scholars working at the CVRLab and ETC have had different agendas.5 My own quest to understand the potent relationship between meaning and motion in the Roman built world was inspired by my mentor Spiro Kostof who often quoted Le Corbusier’s statement that an understanding of architecture requires, “the foot that walks, the head that turns, the eye that sees.”6 The potency of kinetic physical engagement with Roman sites was brought to life by William L. MacDonald, whose own peregrinations throughout the Roman Empire were the basis for his 1988 book, The Architecture of the Roman Empire: An Urban Appraisal which influenced Romanists and architects alike.7 My early attempts to re-engage with kinetic viewing in ancient Rome utilized fictional narratives and diagrams, drew on the empirical work of modern designers and planners, from Kevin Lynch and Gordon Cullen, to Colin Rowe and Rob Krier.8 The multiple perspectives utilized in Roman wall-painting and architectural design analyzed by Lise Bek and Bettina Bergmann, among others, further provoked me to consider the Roman preference for shaping design around kinetic viewing.9 Such a dynamic visual perspective was realistic and engaging, though often downplayed in modern architecture and architectural history. In these fields, the reliance on still photographs and drawings privileged the ideal, fixed, singular view, a quest that was popularized with the “Kodak Picture Spot” campaign of signs across America indicating where to stand to take a good picture.10
At UCLA, the utilization of digital technologies to investigate Roman environments was sparked by the experimental adaptation of computers in contemporary architectural design. In the 1980s, the Department of Architecture and Urban Design offered a graduate degree program in computing. Program head William J. Mitchell assigned students the challenging and instructive task of modeling the classical columnar orders, transitioning from the relatively simple forms of the Doric order to the demanding geometry of the Corinthian. Soon after I arrived to teach architectural history, he invited me to discuss not only the forms and meanings of ancient buildings, but also how they were experienced in their original contexts. At the time, the Postmodernist movement was still in vogue, reacting to the sterility of Modernist designs by incorporating witty symbolism, sensory-rich experiences, and overt historical references. Architect Charles Moore, Dean of the UCLA Graduate School of Architecture and Urban Planning, designed the Piazza d’Italia (1978), and other environments that utilized classical references and directly engaged all the senses. Moore and colleagues adapted and extended earlier scientific work in 3D computer modeling, to create real-time simulation technologies to facilitate the design and analysis of contemporary building and urban projects. In the early 1990s, these efforts and interests resulted in the establishment of the Urban Simulation Team (UST) at UCLA, which utilized a proprietary 3D virtual world system to generate large environments, most notably an ever-expanding real-time digital model of the Los Angeles basin.11
Simulations of driving through Los Angeles in the UST models piqued the appetites of Romanists on campus who longed to be able to simulate movement through similar models of ancient cities. Professor Bernard Frischer from the Department of Classics audited a 3D modeling class in the Architecture and Urban Design Department and we excitedly began to discuss possible projects and strategies for recreating past worlds.12 The most pressing challenge was how. The expansive UST LA models were in part generated using photographic and video captures of extant buildings. Such ease of construction was not available for ancient cities where only wistful fragments and incomplete building footprints remained. Extensive documentation of archaeological remains, secondary images, and comparanda from various sites were needed to reconstruct ancient environments in toto. For greatest accuracy, the same extensive information is needed to create a digital model of a historic building as for a full-scale physical reconstruction. In other words, we needed the equivalent of a full set of working drawings and structural specifications.
In 1996 an opportunity arrived to test the capabilities of digital reconstruction models. The J. Paul Getty Trust contacted the UST to create a scientifically accurate real time model of Trajan’s Forum in Rome for the exhibition Beyond Beauty: Antiquities as Evidence for the opening of the new Getty Museum in Brentwood. Professor Frischer and I acted as scholarly consultants, and immediately invited Professor James E. Packer of Northwestern, who had meticulously studied Trajan’s Forum for over two decades, to participate.13 This initial foray into the application of digital tools for humanistic study raised numerous challenges regarding research objectives, display, user interface, and proprietary rights. The resulting simulation model allowed viewers to move through the recreated spaces at will, experience changes in lighting, observe rolling sections, and click switches to see different time periods or features. Due to the need for high-powered computing, as well as proprietary issues regarding the software, the complete, high-resolution version of the model with full interactivity and a wide range of user-driven movement could be experienced at UCLA, but not at the Getty Museum. A video featuring animations of the model and expert commentators was displayed next to ancient sculptures found in Trajan’s Forum. The crowds filling the museum responded enthusiastically to the reconstruction, but found the scripted narrative style of the video frustrating; they longed for a less mediated, less passive opportunity to “move” through the digital reconstruction at will. The video showcased the reconstruction model, but did not take into consideration the characteristics and meaning of movement in either the simulation model or the Roman world. For example, to maintain audience attention and save time, the video adopted an unrealistic rate of speed through the model, thus compromising understanding of an ancient pedestrian’s kinetic experience. Modes of transport (by foot, cart, sedan chair), or the different significance the Romans associated with individual and group processing were not discussed.14
In the heady early days of digital modeling, soft- and hardware were evolving at a rapid rate; visually beguiling hyperrealistic reconstructions seemed to proliferate like mushrooms overnight, with scientific accuracy at times overwhelmed by the enthusiastic presentation of gorgeous renderings. I was naturally attracted to the full reconstructions of temples, basilicas, and other ancient buildings, but felt frustrated by their presentation as isolated art objects floating benignly in a black void of space. Models of broad historical environments were few. In addition, the technology of the time offered restricted interactivity and movement. Motion primarily took the form of rotation using simple, accessible QuickTime Virtual Reality (QTVR) image files supported by Apple’s QuickTime. These drawbacks, however, seemed of little concern as the rapidity of technological developments promised unlimited advances.
Phase 1: Formation, data aggregation and interrogation of process
In 1997, Professor Frischer (director) and I (associate director) established the CVRLab at UCLA to explore computer applications in historical research.15 In so doing, we joined other researchers experimenting with new technologies whose efforts were then coalescing around a field generally known today as “digital humanities.”16 Frischer and I naturally began with a Roman project, a model of the Forum Romanum in the late antique period. The choice was obvious. The site was historically significant and well known. In addition, though the Forum had been studied for centuries, there were few scholarly reconstructions in any media.
Before beginning, the CVRLab team of faculty and graduate students discussed technological advantages and limitations, as well as new directions of inquiry that opened up with the dynamic capabilities offered by digital technologies and the emergence of new practices in digital humanities. As a result, the CVRLab established five operating principles when creating historical simulations using high-performance computing:
- First, and foremost, to follow the highest scientific standards, with scholarly vetting by international scientific teams of experts throughout the modeling process, full documentation of all source materials, and careful recording of modeling decisions in expansive metadata linked to the reconstructions.
- Second, to conceptualize the models as “knowledge representations” depicting what is currently known about the historical structures and spaces rather than an imagined hyper-realistic simulacra.17
- Third, to include the surrounding environmental context for building models to facilitate experiential and viewshed analyses.
- Fourth, to utilize real-time visual simulation to explore diverse aspects of movement through recreated spaces.
- Fifth, to promote access and utilization by other scholars by advocating the use of open source software and common, accessible platforms.18
These decisions had clear advantages and disadvantages. The distinction between visually representing knowledge about an environment and more hyperrealistic (and thus more hypothetical) depictions was largely lost on students and lay viewers who longed to see sculptures, color, dirt, paintings, ephemeral decorations, and people. Observers found real-time movement engaging, but judged the scholarly renderings generated on-the-fly at 30 frames a second as inferior to the more costly pre-processed animations familiar from contemporary films such as Gladiator (2000) which took days to render a single frame. In effect, the visual quality overshadowed the rigorous documentation, the strict scientific limitation to verifiable data, and the introduction of kinetic capabilities into humanistic research.19 The value of kinetic analysis was further degraded for users by the lack of immediate personal engagement. The large size of the first UCLA models required a complex operating system and expansive computing power with trained navigators for viewing. Showings were held in the UCLA Visualization Portal.20 This theater boasts triple projection, sound surround, and a large semi-spherical screen (24’ wide x 8 ½’ high) that gives the audience a sense of immersion in the projected models. However, the seated observers had limited opportunities for simulated bodily engagement or self-directed action.21 Without control over movement, both viewers familiar with the historical spaces and those with extensive gaming experience became disoriented in the simulated environments, prompting the CVRLab team to include a pop-up map that shows operator position and orientation while in motion.22 Linkage to the metadata through on screen windows interrupted the simulation of movement, creating a cognitive dissonance. The contextualization of buildings also had inherent limitations; though the simulation model of Rome expanded with new buildings added continuously, operators moving through the spaces ultimately came to an unmodeled section, which complicated evaluation of kinetically experienced viewsheds, circulation routes and strategies of way-finding.
The real-time simulations created rich kinetic experiences, but these required mediation to be of research value. Operators gave preference to showing the Roman models at the average eye-level of ancient pedestrians, but often found it hard to resist flying up high to see projects from the sky,an exhilarating effect that diluted the casual evaluation of ancient movement. Time limits imposed on presentations by class-length, room or speaker availability necessitated a quickening of pace; operators guided observers through recreated environments at unrealistic speed, progressing across the 650 feet length of the central Roman Forum in a few seconds. The inclusion of localized sounds, such as the projection of Cicero’s speech from atop the Rostra in the Roman Forum and random Latin conversations, prompted more tempered passage, but only to a degree.23 The full 3D capabilities of the viewing system necessitated the use of glasses, thus further compromising the sense of a realistic encounter with the recreated environments; as a result this capability was only used on occasion.24 Experiments with the insertion of figures proved largely unsatisfying. Not only did human simulations require considerable computing power, they proved distracting as observers fixated on the figures’ overall appearance and unrealistic actions, a reaction that negated their value for enhancing phenomenological analysis in the recreated environments. More effective were assignments in which students imagined the varying pathways through the Forum taken by different ancient occupants (slave, emperor, dowager, senator, etc.), considering where each would have been allowed to go, and at what time of day and year. Since the students could not self-direct the movement through the models, their analyses on paper were often corrected during session in the Portal.
After over five years of work on the project, the CVRLab formally launched the Digital Roman Forum in 2003.25 This website displayed an interactive time-map, textual data on all buildings, and classical texts relating to the site, as well as images and short movie clips of the models. All the information was organized geo-spatially and temporally as well as chronologically.26 The project reflected the team’s interest in spatialized movement, but only in a restricted way. At the time, the narrow bandwidth of the Internet thwarted attempts to deliver the dense 3D models with full interactivity. As a compromise, the site presented static images of building models in various resolutions to maximize access by users with different computing capacities. Some short QTVR clips allowed users to rotate individual structures, but contrary to the team’s guiding principle, presented the structures as if floating in a void; to compensate, other clips presented views of the Forum from fixed locations. The website was (and remains) a useful repository of architectural, textual, and extensive metadata about the digital reconstructions, but does not allow full immersion or kineticism. Motion was most directly interrogated in the synchronized “then” and “now” QTVR clips. These engaging synced views offered simulations of lighting effects at different times of day and year, but had technological limitations which restricted movement to rotation around a fixed viewing position, a posturing that privileged appearance over polisensory kinesthetic experience.27 Furthermore, the site presented catalogued information, placing the burden on users to narrate the expansive and diverse data about space, place, architecture, and occupation.
This brief overview of the CVRLab indicates some of the benefits and challenges facing early adapters of digital simulation technologies for the study of historical movement. The team found that the greatest production of new knowledge occurred during the construction process of both the reconstruction models and the website. These realizations were in line with an emerging shift in the humanities and arts from data aggregation and interpretive analysis (centered in architectural studies on style, form, proportions, and dating), to generative analysis associating transformational meanings with creation, evolution, and use.28 The modeling of the entire Roman Forum, the spaces between buildings became understood as palpable, identifiable and designed aspects of the environment. The meticulous location of individual building models in relation to each other and to the sloping ground plane prompted consideration of speeds of transit and accessibility to heavy transports. After simulating movement through the digital 3D model, scholars began to revise ideas about processing in the Forum based on the size and nature of various interstitial areas.29
During creation of the website, access to data through the geo-temporalized time map raised important issues. The team pondered how to define the spatial bounding boxes of the individual buildings (according to building footprint, platforms, or street edges) and their temporal limits (from date of conception, initiation of construction, or dedication to destruction or alteration). The very existence of the time map underscored a little-discussed issue in relation to reconstruction models: the aging and adaptation of structures over time. In the digital simulations, the structures appear at the correct date as the timeslider is moved, but subsequently show no sign of aging or alterations. The lack of physical aging, seasonal ephemera, artwork, and changes in human use prompted discussions about the city as a living thing.30 Although familiar and expedient, the god’s-eye view presented in the time map was also acknowledged as problematic; ancient occupants rarely had such a perspective.31 Further discussions led to the questioning of sensory universalism. The notion that bodily reactions are physiological and thus consistent overtime was countered by debates on the cultural constitution of responses to heat, cold, smells, danger, and so on, issues only just beginning to be explored in relation to the ancient world.32
Phase 2: Geo-temporal interrogation and increased kinetic experimentation
Significant developments in technologies occurred early in the new millennium. In 2004, the UCLA School of the Arts and Architecture established the Experiential Technologies Center (ETC) as successor to the CVRLab after Professor Frischer moved to the University of Virginia.33 I assumed the role of ETC director; Christopher Johanson, Lisa M. Snyder, and Dean Abernathy became associate directors; faculty from a wide range of departments, graduate students and staff members completed the ETC team.34 The ETC maintained the same operating principles as the CVRLab while promoting the use of new technologies for experiential research in diverse disciplines. The center’s research-driven projects explore a wide range of issues, including movement, sequencing, sonification, and visualization. Each modeling project addresses a specific research or pedagogical agenda. As a result, the type of simulations range from simplified architectural massing models for the consideration of pan-urban sightlines, to more realistic renderings testing representational strategies for lighting analyses.
The new ETC evolved amidst exciting advances. 3D modeling software was becoming increasingly accessible and affordable. Included among the offerings was Sketchup, first released in 2000, a general purpose, easy-to-use 3D content creation tool offered with the tagline “3D for everyone.” Improved computer capabilities allowed many of the capabilities of Phase I to migrate from high computing and the Portal to personal computers accessible at all hours. Simultaneously, web-based geographic computing systems proliferated, further increasing accuracy of model placement. Along with several Virtual Globes or Digital Earths, was Google Earth (GE), formally launched in 2005. The ability to sequence informational points and formulate tours on freely accessed geo-browsers promoted general and academic interest in movement through space. Further stimulus came from the release of GE4 a year later which added a time-slider and free version of Sketchup software. Suddenly people around the world could easily build 3D models and situate them in time and space.35 The UCLA team immediately exploited the new web-delivered environments, decimating extant high-polygon-count models to make smaller, more efficient versions for web delivery. In 2006, the ETC hosted the NEH Summer Institute Models of Ancient Rome (PIs Sander Goldberg and Diane Favro, with Chris Johanson) and presented reconstruction models of the Forum, houses and other buildings on GoogleEarth.
In 2008, UCLA models were featured in Google Earth Ancient Rome 3D created in collaboration with the University of Virginia and Past Perfect Productions.36 One of the first 3D models of an entire city at a particular moment in history, the layer included 6700 3D building models, 250 with informative placemarks, and 11 with viewable interiors.37 The promos encouraged visitors to “Fly into Rome as it looked in 320 A.D.” and “Tour the interior of famous buildings.” Nuanced study of movement was, however, compromised by several factors. The large data set hindered ease of movement through the models on some users’ computers. The GE decision to “float” the Ancient Rome 3D layer several meters above the modern Rome inverted the physical layers as the ancient city lies several meters below the current urban surface. As a result, the project is most useful for panoramic viewing rather than the simulation of a person moving through recreated spaces.38 This example underscores the challenges of relying on external platforms, as well as the difficulty of accurately reconstructing and situating architectural components along the vertical (Z) axis when data about original foundation levels and building heights is either incomplete or unreliable.39
Early open access versions of Sketchup and similar software lacked the sophisticated capabilities of more complex modeling programs, but offered an ease and rapidity of creation that promoted hypothesis testing. This scientific approach increased comfort with idea-generating schematic representations (such as spatial diagrams, unarticulated massing models, and negative space models) commonly used in the field of architecture, though relatively unused in history, archaeology, and classical studies.40 Conceptualizing the digital reconstructions of ancient Rome as laboratories, the ETC team began to conduct experiments on processing and other rituals. While much of the analysis was done on PCs, the team found projection of the models in the Portal to near full-scale greatly facilitated the interrogation of sightlines, soundscapes, and sequences experienced in motion.41
Chris Johanson and I collaborated on an investigation of Roman funeral processions. Previous researchers had studied numerous aspects of ancient funerals, but none had contextualized movement of the associated corteges. Thus the important events were neither grounded in cityscape of Rome, nor interrogated in relation to the intricately constructed Roman sensory experience. Drawing on descriptions in ancient sources, we conducted experiments simulating moving through period-specific topographical and architectural contexts. When examining funerals in the Roman Republic, a period for which limited architectural information is available, we used simple massing and terrain models graphically distinct from those depicting the more documented Imperial period.42 Expanding on Professor Johanson’s previous research, we investigated processional movement in the early Republic considering such issues as vantage points and timing of the parades.43 For the Imperial period we analyzed how the choreography of the route created a legible propagandistic narrative for participants and audience alike. We presented our findings in the first digital issue of the Journal of the Society of Architectural Historians (JSAH).44 The digital layout directly associated the argumentation with zoomable still images and an interactive kmz file of the Roman Forum that opens in Google Earth. In the geo-browser readers can move through the reconstruction models, change periods using a timeslider, and link to the Digital Roman Forum archive. With access to the experimental spaces, they thus have the ability to assess our conclusions and conduct their own experiential analyses of the kinetic experiences.
Simulated movement through a digitally reconstructed historical environment is appealing and informative, but also complex and uncertain. Many questions arise. Are modern physiological reactions and conceptual valuations of experience similar to those of historic occupants? Does the experiential interpretation of built forms and urban spaces lose credibility when only select sensorial capabilities are included? How does the type of human interaction with the digital simulation shape interpretations?45 After all, the static, experientially passive perusal of a text is distinctly different from the dual engagement of eyes and hand using an input device (such as a mouse) while moving through a model. The choreography of a reader/visitor’s interaction is new for most scholars. We crafted the print-based experience to mimic some of the real-time kinetic experiences, but this parallelism was in some cases lost in practice. Visitors to the digital JSAH article frequently begin by reading the associated text rather than deal initially with opening windows, downloading files, and experimenting with movement through the associated models. As a result, they peruse the models separately from the written argument. This order of engagement situates the 3D models in the traditional position of “illustration,” implying they were derived after the formation of ideas, rather than serving as integral components in the creation of knowledge and assessment.
Currently, there are no accepted scholarly guidelines for interaction with digital reconstructions or, for that matter, for their academic assessment. Each model must be judged in relation to its scholarly purpose, keeping in mind not only the validity of conclusions and scientific rigor, but also the limitations imposed by the availability of data, the technology used, the time allotted for creation, costs, and the external delivery interfaces that impact functionality.46 Evaluation at all levels is critical. Lisa Snyder has consistently gathered data on user experiences of ETC models and is exploring how to enhance legibility and provide consistent assessment of digital reconstruction models, including movement.47
For almost two decades UCLA scholars have been working together to advance digital humanities by conceptualizing and promoting new hybrid methodologies, disciplinary paradigms, evaluation standards, sustainability structures, publication media and models, and tools.48 Especially relevant for movement studies and publication is the development of HyperCities created in collaboration with USC, CUNY, and community partners.49 The generative premise behind this multi-modal project is that meaning resides in the interaction and intersection of stories situated in time and space. HyperCities evolved from a flash-based mapping project into a robust participatory, multimodal platform that brings together the analytic tools of GIS, the geo-markup language KML, and traditional methods of humanistic inquiry utilizing Google’s Map and Earth Application Programming Interface (API) released in 2005-06. Believing that space and movement are essential components of memory and narratives, ETC team members pushed for the incorporation of interactive 3D models to promote studies on the diverse ways in which space and movement enrich narratives and research.
Visualizing Statues is the first publication project to fully exploit the 3D capabilities of HyperCities. The project weaves interactive 3D reconstruction models, data sets, maps, photos, text-markup, and interpretive essays into a seamless experience centered on narration. Increasingly, researchers in diverse fields are acknowledging the value of stories to make sense of ever-expanding data. To cite one example, the new startup Qwiki has set as its goal, “to deliver information in a format that’s quintessentially human – via storytelling instead of search.”50 The Visualizing Statues project examines the story of the emperor Honorius’ ritual procession through Rome in 404 CE. The goal was not literally to recreate the event, but to present a curated research narrative interrogating the meanings generated by experiencing the environments and sculptures in a memory-laden urban space of late antiquity. Through the analytical essays entitled “Statues and Memory,” “Ritual Experience,” and “Spatial Context,” Gregor Kalas maintains an overarching continuity of voice and interpretation.
The meaning of Honorius’ ritual parade derived in large part from the sequential experience of buildings, spaces, and especially of objects in the late antique Forum. The study focuses on the original impact of sculptures and their related inscriptions. Though well-documented, these highly legible features have been ripped from their original contexts. Some have been relocated within the Forum; others have been dispersed among various collections or lost, thus forestalling attempts to understand their original experiential impact. The Visualizing Statues project recontextualizes the statues and inscriptions, placing them in situ within a 3D reconstruction model of the Forum.51 While readers can peruse the essay from beginning to end in the “Ritual Experience” section, a crucial part of this research is the integration of text with images and modeled environments. The same essay is presented in conjunction with the models to provide a richer understanding. After launching HyperCities, readers progress through the argument and the models in sync, with each consistently situated in time and place. In the “Ritual Experience,” and “Spatial Context” sections, a click on the “next” button simultaneously takes them to the succeeding written segment and to the associated position in the model. Readers can then adjust the size of the views, maximizing or reducing the left and right panels at will. They can also control their pathways, either progressing in a linear fashion through the text and associated simulations, or moving “off script” to create personal storylines by passing from one section of the website to another, or by plotting personal trails through the 3D models. Such interactivity reinforces the dual roles of digital reconstructions as visualizations of complex data and potent sites of knowledge creation.52 By controlling their own simulated passage through the website, readers reaffirm the underlying premises of the project:
- knowledge and understanding is forged and strengthened through movement
- study of ancient sculptures and inscriptions must consider observers’ kinetic, experientially rich engagement within the original physical context
The Visualizing Statues collaborative project belongs in a continuum of evolving studies and technologies applied at UCLA to the study of motion in ancient environments. Of course, pragmatic restrictions — evidence availability, time limits, funding — constrained the team’s aspirations. On the plus side, the project clarified priorities for future advancements in the study of movement and 3D analysis in recreated historical environments. Four crucial, interrelated areas call out for development.53 First is the essential expansion of polysensory evaluation. Navigating using a computer mouse while in a sedentary position limits and distances bodily engagement. Movement is comprehended and remembered in concert with all sensory receptors, necessitating further research on lighting, sounds, smells, taste, haptic responses, and also mood.54 The kineticism and 3D simulation of the Visualizing Statue project hint at embodiment of the recreated spaces, but still privilege ocularcentrism over other senses. How does our interpretation of sculptures change when considering the Forum filled with blaring sounds, potent smells, rough surfaces, bright colors, and hundreds of people?
Second is a need for instruction in how to interpret space, speed, processing, and temporal factors. Scholars trained in the humanities have not developed disciplinary standards for kinetic and sensory experimentation and assessment. There are no agreed upon metrics to facilitate comparative analysis.55 While we all occupy environments, our skills at reading simulated environments while in motion vary widely. The computer adds another layer of mediation. How should users conceptualize their role when moving through the Forum model? Without training, many mimic disengaged observers like 19th-century flâneurs or their 21st-century equivalents, meanderthals.56 More appropriate exemplars can be developed drawing upon the extensive research in cognition and embodiment centered on situation awareness, defined as the perception of environmental elements within time and space and the comprehension of their meaning.57 Further experiments within the 3D environments of the Visualizing Statues project can help to clarify the skills and means needed to evaluate and interrogate how human motion shapes (and shaped) thinking.
Third, extensive cultural contextualization is necessary for all historical simulations. In recent decades scholars have explored visuality, how vision is culturally constituted; initial forays investigating other senses are also underway for the Roman world, though the study of movement in relation to time and space is lagging behind considerations of smell and tactility.58 Numerous queries come to mind. Did the Romans in the Forum privilege movement or sight over smelling, or were all aspects experienced in a fog of urban distraction? Did they consider crowding as positive or negative? How did representations of processions in Roman art mime or determine actions? Was there a class distinction between speeds of movement? How was the Roman time of day or year a factor in kinetic analysis? On an operational level, should scholars develop extensive historical backstories about the spaces and occupants under scrutiny?59 In the current project the exploitation of statues and inscriptions to reinforce the political agenda of Honorius could be further enriched by a broader exploration of sensemaking.60 How did the Romans conceptualize the habitus defined by architectural historian Dell Upton as, “the nexus of spaces and times that repeatedly trigger bodily habits and cultural memories”?61
The fourth challenge is perhaps the most important for the promotion of digitally-based humanistic research in general. If a picture is worth a thousand words, an interactive 3D simulation is worth tens of thousands. Yet there is no agreement about scholarly assessment. What are the criteria for judging the content of intricate digital projects that are constrained not only by the cited research agenda, but also by the capabilities of the technology and delivery system, the costs, the programmatic requirements of funding source(s), and the lack of comparative standards? Despite the proliferation of research and production using 3D modeling in the arts and digital humanities, critical appraisal is unstable. Only a handful of scholarly journals review websites or other digital presentation venues; blog reviews often focus on the use of the technology and the graphics rather than the research substance.62 A distinct advantage of digital projects is their inherent mutability, yet there almost never post-launch reassessments evaluating subsequent updates and continued performance. Equally problematic is the mothballing of many undertakings due to the lack of sustainability plans or archiving standards. In addition, the rapidity of advances in interfaces and rendering techniques make the digital kinetic simulations of 2000 seem graphically and experientially inadequate in 2010, a technically-imposed denigration that too often taints valuable research content. What steps can be taken to avoid the new knowledge produced by the Visualizing Statues project and others from being sidelined by enticing new developments such as controller-free, natural interfaces where users engage with simulations through bodily gestures and spoken commands?63
In Book 8 of Virgil’s famous epic poem, the Aeneid, King Evander leads Aeneas through the future site of Rome, exploiting movement as a way to generate a narrative, to explain contemporary monuments and, ultimately, to animate history and convey values.64 The Visualizing Statues project takes the reader on a tour through a simulation of the city in the late antique period, likewise demonstrating how inscriptions, sculptures, buildings, and monuments in Rome forged enduring memories and transported the reader beyond the here-and-now. Both these simulated kinetic urban narratives were meant to be visited and revisited. Similarly, digital technologies are open to adaptation and redirection, shifting the description from research as a product, to researching as an on-going process, itself giving meaning through motion.
1 Several studies of Roman walking have been undertaken by Timothy M. O’Sullivan, including, “The Mind in Motion: Walking and Metaphorical Travel in the Roman Villa,” Classical Philology 101, no. 2 (April 1, 2006): 133-152; “Walking with Odysseus: The Portico Frame of the Odyssey Landscapes,” American Journal of Philology 128, no. 4 (2007): 497-532; and his forthcoming book Walking in Roman Culture (Cambridge University Press, 2011). Roman kinetic experience of cities is explored in Fikret Yegül, “Street Experience in Ancient Ephesus,” in Streets: Critical Perspectives on Public Space, ed. Zeynep Çelik, Diane Favro, and Richard Ingersoll (University of California Press, 1994), 95-110; and Ray Laurence and David Newsome eds., Rome, Ostia, Pompeii: Movement and Space (Oxford University Press, forthcoming 2011).
2 Cic.Fin.5.5; see also 5.2; James Frakes, “Monuments of Passage: Roman North Africa and the Emperor on the Move,” ARRIS 18 (2007) 53-69; Anthony Corbeill, “Political Movement: Walking and Ideology in Republican Rome,” inThe Roman Gaze: Vision, Power, and the Body, ed. David Fredrick (The Johns Hopkins University Press, 2002), 182-215; Florence Dupont, Daily Life in Ancient Rome (Wiley-Blackwell, 1994), 74; Nicholas Purcell, “Rediscovering the Roman Forum,” Journal of Roman Archaeology 2 (1989), 165.
3 Diane Favro, “Reading the Augustan City,” in Narrative and Event in Ancient Art, ed. Peter J. Holliday (Cambridge University Press, 1993), 230-257.
4 Christian Norberg-Schulz, Genius Loci: Towards a Phenomenology of Architecture (Rizzoli 1979).
6 Spiro Kostof and Greg Castillo, A History of Architecture: Settings and Rituals (Oxford University Press, 1995), 3.
7 William L. MacDonald, The Architecture of the Roman Empire: An Urban Appraisal (Yale University Press, 1988).
8 Diane Favro, The Urban Image of Augustan Rome (Cambridge University Press, 1996); Kevin Lynch, The Image of the City (MIT Press, 1960); Gordon Cullen, Fred Koetter, The Concise Townscape (Van Nostrand Reinhold Co., 1971); Donald Appleyard, Kevin Lynch, and John Randolph Myer, The View from the Road (MIT Press, 1971); Rob Krier, L’espace de la ville: theorie et pratique (Bruxelles Belgique: Archives d’Architecture Moderne, 1975); Colin Rowe, Collage City (MIT Press, 1978).
9 Lise Bek, Questiones convivales: the Idea of the Triclinium and the Staging of Convivial Ceremony from Rome to Byzantium (Odense University Press, 1983); Bettina Bergmann, “The Roman House as Memory Theater: The House of the Tragic Poet in Pompeii,” The Art Bulletin 76, no. 2 (June 1994): 225-256; Jaś Elsner, Roman Eyes: Visuality and Subjectivity in Art and Text (Princeton University Press, 2007).
10 http://getsatisfaction.com/eastmanhouse/topics/kodak_picture_spot_road_sign. In pre-digital days, the ease of capturing a single image versus recording a video must be underscored.
11 The UST was directed by William Jepson; he and Robin Liggett co-authored, “An integrated environment for urban simulation,” Environment and Planning B: Planning and Design 22, no. 3 (1995): 291–302. The software developed by the UST derived from that used by the military for flight simulation training. Today the Urban Simulation Team continues to focus on the Los Angeles project, but also produces historical models under the direction of Lisa M. Snyder; http://www.ust.ucla.edu/ustweb/ust.html (accessed October 25, 2010).
12 Alison Babeu, Rome Wasn’t Digitized in a Day: Building a Cyberinfrastructure for Digital Classicists Draft Version 1.2—9/29/10, http://www.clir.org/activities/details/infrastructure.html (accessed October 25, 2010).
13 Lisa M. Snyder, “Real-time visual simulation models in an exhibition environment,” presented at the IEEE Virtual Reality Conference (Chicago, 2004), 2-3; resumbrae.com/vr04/snyder.pdf (accessed October 25, 2010)
14 Anthony Corbeill, Nature Embodied: Gesture in Ancient Rome (Princeton University Press, 2004), 107-39.
15 For a full list of the aims of the lab and our generous sponsors see http://www.cvrlab.org/ (accessed October 25, 2010). Bernard Frischer provides an overview of the lab’s aims in “Mission and Recent Projects of the UCLA Cultural Virtual Reality Laboratory,” in Proceedings of the Conference Virtual Retrospect 2003, Biarritz, France 6-7 November 2003, ed. R. Vergnieux and C. Delevoie (2003), 65-76; and “The Digital Roman Forum Project of the Cultural Virtual Reality Laboratory: Remediating the Traditions of Roman Topography,” in The Reconstruction of Archaeological Landscapes through Digital Technologies, British Archaeological Reports S1379, ed. Maurizio Forte (2005), 9-21.
16 The evolution of Digital Humanities as a term and field is examined in Matthew G. Kirschenbaum, “What Is Digital Humanities and What’s It Doing in English Departments?” ADE Bulletin, 150 (2010), 1-7.
17 Randall Davis, Howard Shrobe, and Peter Szolovits, “What is a Knowledge Representation?” AI Magazine 14, no. 1 (1993): 17-33.
18 The team’s decision to view the 3D models using vrNav, an open source 3D scene navigation program developed at UCLA, was also a factor in the early use of open source software ; http://www.ats.ucla.edu/portal/research_activities/vrNav/ (accessed October 2010).
19Ruth Tringham and Rosemary Joyce, “Feminist Adventures in Hypertext,” Journal of Archaeological Method and Theory 14 (2007), 340.
20 http://www.ats.ucla.edu/portal/about_the_portal/default.htm (accessed October 25, 2010).
22 The US Military was researching issues of navigator orientation in 3D simulations during the same period; S.R. Goerger, R.P. Darken, M.A. Boyd, T. A., Gagnon, S.W. Liles, J.A. Sullivan, and J.P. Lawson, “Spatial knowledge acquisition from maps and virtual environments in complex architectural spaces,” Proceedings of the 16th Applied Behavioral Sciences Symposium, U.S. Air Force Academy, Colorado Springs, CO. (1998): 6-10.
23 The speech inserted in the model was performed by Bryan Levitt; Sander Goldberg, “Performing Cicero,” http://cicero.humnet.ucla.edu/index.htm (2003-2004; accessed October 25, 2010). The Roman Forum model was also enriched by localized Latin conversations.
24 At the time, the use of 3D glasses in a theater such as the Visualization Portal could not compare to the more dynamic experience of observers inside Cave Automatic Virtual Environments (CAVEs) such as that installed at the University of Michigan in 1997; http://um3d.dc.umich.edu/resources/hardware/res_hardware_cave.html (accessed October 2, 2010).
26 In the period before the widespread proliferation of GPS data, the Digital Roman Forum carefully included geospatial data; http://dlib.etc.ucla.edu/projects/Forum/geospatial (accessed October 25, 2010).
27 Day lighting studies were made for the “then” and “now” views on the Digital Roman Forum website, though in other instances the CVRLab team relied on the generalized lighting effects included with the modeling software.
28 Diane Favro, “In the eyes of the beholder: Virtual Reality re-creations and academia,” Journal of Roman Archaeology Supplement 61 (2006): 321-334.
29 Diane Favro, “The City Is a Living Thing: Performative Roles of Urban Sites in Ancient Rome,” in The Art of the Ancient Spectacle, edited by Christine Kondoleon and Bettina Bergman (National Gallery of Art, Yale University Press 2000), 205-219.
30 Plutarch, Moralia, 559. Diane Favro, “Fluctuating Edge Conditions in Ancient Rome,” in Rome des quartiers : des vici aux rioni. Cadres institutionnels, pratiques sociales et requalifications entre Antiquité et époque moderne. Actes du colloque international de la Sorbonne (20-21 mai 2005), ed. Étienne Hubert, Manuel Royo, and Agnès Béranger (Paris, 2009), 279-298.
31 Diane Favro, “The iconiCITY of ancient Rome,” Urban History 33, no. 1 (2006): 20-38.
32 Regarding Roman visuality, the cultural constitution of vision, see Diane Favro, “Ancient Rome through the veil of sight,” in Sites Unseen: Landscape and Vision, ed. Dianne Harris and Dede Ruggles (Pittsburgh, 2007), 111-130.
33 www.etc.ucla.edu (accessed October 25, 2010). The ETC team plays an active role in promoting digital humanities initiatives at UCLA; http://idre.ucla.edu/hasis/; www.keckdcmp.ucla.edu; www.digitalhumanities.ucla.edu (accessed October 25, 2010).
35 Declan Butler, “Virtual globes: The web-wide world,” Nature 439, no. 7078 (February 16, 2006): 776-778.
36www.romereborn.virginia.edu. In 2006, the UCLA models were licensed to Past Perfect Productions which utilized them on various commercial projects including an edutainment 3D movie currently showing in a virtual theater in Rome; www.pastperfectproductions.com; www.3drewind.com.
37 The project utilized procedural modeling, a rule-based generative computer technique based on shape grammars, to generate infill buildings; Simon Haegler, Pascal Müller, and Luc Van Gool, “Procedural Modeling for Digital Cultural Heritage,” EURASIP Journal on Image and Video Processing Volume 2009, Article ID 852392 (2009 1155): 11 pages.
38 Teaching assignments developed for a Google-sponsored 3D curricular competition engaged students through role playing in the recreated city, but the interrogation largely took the form of 2D mapping; http://www.google.com/educators/romecontest.html. To address the challenge of navigating in geo-browsers, the ETC team created lengthy QTVR movies around and through buildings, as well as animations of ritual processional paths for the Digital Karnak project launched 2009; http://dlib.etc.ucla.edu/projects/Karnak.
39 In contrast to print publications, on-line digital research projects necessitate continuous monitoring by authors as updates and alterations to external geo- and web-browsers can negatively affect functionality.
40 Christopher Johanson, “Visualizing History: Modeling in the Eternal City,” Visual Resources: An International Journal of Documentation 25.4 (2009): 403–418.
41 Regarding sound, the article included graphic renderings of soundscapes and aural clips of recreated Roman funerary music.
42 Johanson, “Visualizing History,” 403-418.
43 Christopher Johanson, Spectacle in the Forum: Visualizing the Roman Aristocratic Funeral (Ph.D. dissertation, UCLA 2010); “A Walk with the Dead: A Funerary Cityscape of Ancient Rome,” in A Companion to Families in the Greek and Roman Worlds, edited by Beryl Rawson (Blackwell Publishers 2011), 408-430.
44 Diane Favro and Christopher Johanson, “Death in Motion: Funeral Processions in the Roman Forum,” Journal of the Society of Architectural Historians 69, no. 1 (March 2010): 12-37. The careful analysis of mobility in the center of Rome, in turn, generated research on how building materials were moved through the city; Diane Favro, “‘If the axle breaks, what is left of their bodies?’ Construction traffic in Ancient Rome: Building the Arch of Septimius Severus,” in Rome, Ostia, Pompeii: Movement and Space, ed. Ray Laurence and David Newsome (Oxford University Press, forthcoming 2011).
45 Not all readers are trained to comprehend arguments presented visually. Many complain that pop-up windows not only block underlying representation, but that they disrupt the narrative and visual flow through models, an effect exacerbated by the disjunction between 2D content in windows and 3D content in underlying models.
46For example, Professor James E. Packer and architect Gilbert Gorski aim to convey the attraction and impact of completed Roman buildings by modeling structures of a specific period with heightened realism; renderings of the models will appear as still images in a forthcoming book on the Roman Forum (Cambridge University Press). In contrast, the ETC team aims to interrogate geo-temporal and experiential issues, creating knowledge representations of urban spaces over time and utilizing these as real-time research environments where space and mobility, among other topics, can be investigated.
47 This work is an outgrowth of Snyder’s dissertation research; The design and use of experiential instructional technology for the teaching of architectural history in American undergraduate architecture programs (Ph.D. dissertation, UCLA 2003).
49 http://hypercities.com/; Todd Presner, “HyperCities: A Case Study for the Future of Scholarly Publishing,” in The Shape of Things to Come, ed. Jerome McGann (Houston: Rice University Press, 2010), 251-71, http://cnx.org/content/m34318/latest/ (accessed, October 25, 2010).
50 Attempting to create an improved information experience, Qwiki on-line inquiries aggregate images, videos, and sound clips with structured data imported from web sources, and present the results on the fly as a story read by a female voice; http://www.qwiki.com/.
51 The “Mapping Statues” section situates sculptures from the fourth and fifth centuries CE on an interactive, geo-temporally constructed map; the “Inscription Database” catalogs the preserved inscriptions along with images, references, and text descriptions with full epigraphic markup using EpiDoc, an XML encoding project applying Text Encoding Initiative (TEI) standards to texts from inscriptions.
52 For more on spatio-temporal argumentation see Chris Johanson’s analysis in the “Methodological Essays” section of Visualizing Statues.
53 The discussion here centers on research issues rather than technological solutions. Regarding the latter, the Visualizing Statues project underscored the need to refine strategies and technologies for depicting sculptures and humans in 3D. Faced with uncertainty about the scale and orientation of ancient statues, and the amount of work required to create attractive and efficient figures, the ETC team opted to use 2D billboards and simplified 3D forms. Such representations seem inferior to audiences familiar with the high-quality, human avatars found in gaming, a field that exploits proprietary software and funding sources distinctly different from those found in academia.
54 Research centered on living subjects has progressed from surveys, to field experiments, to brain responses studies; Seyhmus Baloglu and David Brinberg, “Affective Images of Tourism Destinations,” Journal of Travel Research 35, no. 4 (March 20, 1997): 11 -15; Nold; Bio Mapping / Emotion Mapping, http://www.biomapping.net/ (accessed October 25, 2010); The Art & Neuroscience Project at the Italian Academy for Advanced Studies in America, Columbia University, http://www.italianacademy.columbia.edu/artandneuro_about.html (accessed August 10, 2010).
55 Attempts to develop notational systems for experiential documentation in archaeology have not coalesced around a single system, though developments in gestural interfaces hold promise for digital driven alternatives; Dan Saffer, Designing Gestural Interfaces: Touchscreens and Interactive Devices (Sebastopol 2008).
56 http://www.urbandictionary.com/define.php?term=meanderthal (accessed October 23, 2010). Charles Baudelaire characterized the flâneur as a person who walks the city in order to experience it; Chris Jenks and Tiago Neves, “A walk on the wild side: Urban ethnography meets the Flâneur,” Cultural Values 4, no. 1 (2000): 11.
57 M. R. Endsley, “Toward a theory of situation awareness in dynamic systems,” Human Factors 37, no.1 (March 1995), 32–64. Though derived from military needs, the field of situation awareness offers a content and evaluative structure that can enhance historical experiential research; Brian Tomaszewski, “Situation awareness and virtual globes: Applications for disaster management,” Computers & Geosciences In Press, Corrected Proof, section 5; http://www.sciencedirect.com/science/article/B6V7D-50CDSN0-4/2/111f71213c142a1f67d47c8353ed1140 (accessed October 26, 2010).
58 This preference was evident at the 2010 UCLA conference entitled “Synesthesia. Classics beyond the visual paradigm,” organized by professors Shane Butler, Alex Purves, and Mario Telo. The engaging papers directed scholarly attention directly toward ancient sounds, colors, smells, tactility, and taste; while mobility was interwoven among these topics, it was not analyzed in depth; http://www.classics.ucla.edu/people/faculty/butler/Synesthesia/.
59 Lessons might be learned from the film industry where digital “pre-viz” (pre-visualization) is used both for storyboarding cinematic narratives and analyzing the character’s experiences through animated back stories.
60 Sensemaking can be defined as the combined sensory and cultural reading of experiences; as such it builds upon the knowledge underlying situation awareness; Gary Klein, Barry Moon, and Robert R. Hoffman, “Making Sense of Sensemaking 1: Alternative Perspectives,” IEEE Intelligent Systems 21, no. 4 (7, 2006): 70-73.
61 Dell Upton, “Architecture in Everyday Life,” New Literary History 33, no. 4 (2002): 720.
62Several journals, including the JSAH, now have digital review editors, but overall the annual number of projects vetted remains small and few consider the experiential aspects of either the content or the user interface; Diane Favro, “Review of Ut natura ars. Virtual reality e archeologia,” Journal of Roman Archaeology 22 (2009): 621-624. The lack of scholarly evaluations thwarts both advances in the field and academic promotions.
63 Technologies for bodily engagement are becoming increasingly available to broad audiences as with Kinect for the Xbox 360 video game platform (pre-launch October 2010) which allows users to manipulate movement on screen with their bodies rather than with a controller; http://www.xbox.com/en-US/kinect/.
64 O’Sullivan, “The Mind in Motion,” 150.