Urban Tomographies. Martin H. Krieger

photographs (1865–68), quite often there are features or buildings, sometimes in the background, that are seen in the 2008 rephotographs.

      There is lots of room for further work in this vein. In the last few years various private firms have systematically photographed the streets of major cities, and made those images available on the Internet. Google. com’s Street View, with its twelve-camera van; pagesjaunes.fr in France, with its squad of photographers going up and down the streets; and bing.com’s oblique aerial views (provided by Pictometry) are perhaps the most well known (as of early 2010). So if you have an archive of old photographs (say, from a newspaper morgue or a utility company’s files), you may do armchair rephotography—although it is likely that the perspectives provided by the private firms will only roughly approximate those of your earlier images. On the other hand, a slightly different point of view reveals important facts, since it is easy for occlusions to occur. We are in a three-dimensional world, and our images are from particular vantage points at particular times.

       Swarming as Tomography

      Imagine sending out a swarm of people or a troop of scouts into a market or crowd, or imagine a crew of celebrity watchers outside a Beverly Hills restaurant, each person equipped with a video cell phone. The resulting number of videos would be large. Rather than edit or try to conform them to each other (as in a montaged panorama), we might display them all together on one screen in a tiling, with locations and times indicated by maps and clocks. If we are fortunate, we have various scales portrayed, from overviews, inside and out, and at various angles, to many different detailed individual interactions.

      We are informed and curious inquirers, and so we bring along notions about what goes on in cities and what we might look out for. By means of these slices in space and in time, we might have the sense that we are looking everywhere at all times. Still, there are likely to be lacunae. Realtime monitoring of the inflow of videos might allow the viewer to send commands to the swarm or troop to focus in on certain places, to get more detail about a particular activity, or to spread themselves out.²⁰

      We might infer which point in space and time is being examined at each moment in a video. We link (in our database) multiple views of the same space and time region. The places we are looking at are quite varied in their structure, with occlusions, multiple layers, and multiple foci of interest. Ahead of time we know little in detail about a place’s organization, which itself changes in real time, but we do have general expectations from past experience. However, that organization is not at all fiducial: for everyday life, you cannot just say something such as, let me look at the left ventricle of the heart from below, or part of the Earth’s surface at latitude-longitude Y–Z, at time X, under a particular range of the electromagnetic spectrum. The everyday world is rarely so fixed and specified.

      Even so, you already have some idea of the organization of the place or process, perhaps even a wrong idea, and in the process of examining the various vignettes you are filling in that schematic organization or perhaps questioning its capacity to accommodate this or that vignette. You do not figure out the structure from the vignettes; you fill in a presumption, if that presumption is sufficiently accommodating. (If “figuring” is given the literal interpretation, in terms of drawing a figure, then perhaps you are figuring out—that act of imaginative draftsmanship.)

Figure 5. Multipanel display of videos produced by three cell phones at Santee Alley bazaar in downtown Los Angeles (Figure 1A and B, #20). tomography.usc.edu.

      Consider a simpler situation, taking many video clips of a well-defined site with just one camera: a video album. Sometimes the world as such is available to us as a patterned, organized whole, as when we enter a small foundry and just look around, perhaps sketching a plan of what we see. Still, we might find the scene overwhelming in its detailed activity. The aspectival variations provided by a video album, viewed again and again, help make sense of the place; yet in our having looked around, we already have a sense of the whole place, albeit perhaps not yet a sense of how foundry work is done, the processes and the flows of materials. So we might spatially organize the vignettes rather more precisely than is provided by geographical positioning (GPS) information, following our plan sketch. Still, the various processes and how they interact may remain rather more opaque. We might infer what is going on by looking at the flows of materials from one vignette to another, arranging them in process order rather than spatial order (although the design of a workshop, or of a factory, should make the two reasonably congruent). Moreover we already know lots about materials, that heat makes things flow and cooling stops that flow, and so we bring physical science to our viewing of the vignettes. To make sense, the place must be organized spatially and processually and in accord with the physics and chemistry of materials. And, of course, as did Diderot, we might ask the people who work there what is going on.

Figure 6A. Mold making, casting, and polishing in sand casting of metal: foundry, downtown Los Angeles (Figure 1A and B, #4).

      Or perhaps we visit a vast and sprawling high-end forge, with monstrous presses, ovens, and sophisticated milling machines. While the individual images or videos give us a sense of each of the activities, still we need a layout diagram of the forge to locate them in the space and time of this industrial complex. We cannot see all of it from a single position. It would help to take notes on what we are told by the foremen and to be able to ask them questions on a return visit. In addition, as with the foundry, knowing about the physics of metals would have made sure that our story made technological and scientific sense.²¹ In each case we might want to know about how people work together in dangerous but mostly predictable work environments—presumably in well-choreographed groups.

      Figure 6B. Overview of foundry floor: ingots at far right (the oven is nearby but not in the picture), ejected sand from castings at far left, and men preparing, pouring, and cleaning castings in the middle ground.

      Once the place makes better sense, the video clips are now seen as aspects of that place, that foundry, that industrial process and its machinery and its workers. Yet it is those video clips and our sketch plan and our interviews that enabled us to articulate more adequately this place’s sense and meaning.

      A place might be a neighborhood street rather than a formal site. Having made multiple videos of an ethnic neighborhood’s main street, such as East César E. Chávez Avenue in Boyle Heights, Los Angeles (Figure 2; Figure 1B, #3), we discover who is on the street at 11:30 A.M. on a weekday in Boyle Heights—elderly adults and mothers and children, or people who are not at work or school—complementary to what is seen at our work sites. The aspects or video clips are aspects of work and family life in an ethnic working-class neighborhood. Much of this is obvious once seen, but perhaps less so ahead of time and likely less obvious fifty years from now.

Figure 7. Engraving showing the making of plate glass, from the L’Encyclopédie (~1760).

       The World Is Consistent

      I have been focusing on the visual and spatial aspects of phenomena, but the aural and temporal ones are as significant. If we have multiple video vignettes taken by multiple cameras, the soundscape is likely to be shared even more than is the landscape. Sound is everywhere, sneaking up from behind, diffracting around corners. Images of different nearby places may well