New Spaces for Stories: Technical and Conceptual Challenges to Using Spatial Imagery in Environmental History
EDITOR’S NOTE: This Field Notes Essay is a companion piece to the author’s Gallery essay “Frame DS1050-1006DF129: March 20, 1969” in Environmental History 19.2 (April 2014): 271-280. Available here.
By David Biggs
Surreal and Subterranean
On a cold January morning in 2011, I met a French colleague at a train platform in Saint-Denis, a northern suburb of Paris. We were headed to a 19th century French fort, Fort de l’Est, to re-photograph prints of air photos taken by French forces over central Vietnam in the 1940s and 50s. A few days earlier, she had introduced me to the lone archivist responsible for maintaining all of France’s historic military air photography, a collection covering much of Europe as well as North Africa and Indochina. At his main office at the Service Historique de la Défense (SHD) inside the walls surrounding the historic Chateau de Vincennes, the archivist navigated a wholly analog system of hand-denoted, paper index maps and plastic overlays to identify specific rolls of film and photo frames for my area of interest. The film and photos are stored separately at Fort de l’Est because the “caves” built into the mounds surrounding the fort provide a natural form of climate control.
Compared to most archive sites, the SHD facility at Vincennes and the French Air Force’s storage at Fort de l’Est are surreal. Working inside these buildings, a belle epoch ballroom and an unheated cave of a mid-19th century French fort, its easy to put oneself in other times. The archives at Vincennes occupies a series of elaborate 19th century buildings adjacent to a 14th century castle that housed French kings, a 17th century dungeon that imprisoned the Marquis de Sade, and a plaza that served as the execution spot for Mata Hari in 1917. Approaching both places, one travels through gates that separate the outside world of urban Paris from an internal, military and historical realm. Even outside Fort de l’Est in the industrial suburb of Saint-Denis, I was surprised to see a gypsy camp situated in the moat outside the fort’s walls. Inside the fort was an active military facility. The historic air photography is stored beside the contemporary air surveillance activities of the French military, a program called “Plan Vigipirate.” Whereas photointerpreters pored over air photos in the 1960s, today’s surveillance technicians stare at computer monitors connected to satellites and thousands of cameras. We walked past the contemporary air surveillance activities into concrete-lined “caves” to meet our archivist friend and pour over air photos taken in the early 1950s over Huế, Vietnam. My reward for this long-planned, complicated and expensive archive visit was wholly worth it. We found hundreds of air photos detailing built and natural features in Vietnam in an otherwise obscure period, the two decades after World War II and before American troops arrived.
While American reading rooms in such places as the National Archives at College Park may lack the ambience of a Vincennes, the system for storing air photos and the cans of film is no less surreal. The cartographic records of the Defense Mapping Agency include the air photos used for military photo intelligence (World War 2 era) and photos that served as base materials for the production of topographic maps. The sterile veneer of an American reading room serving up historic air photography belies the national network of subterranean vaults storing millions of cans of film. The American solution is to subcontract the work of maintenance to the world’s largest underground storage company, Iron Mountain, named after the first facility, an iron mine in upstate New York.1 Today the company maintains hundreds of storage sites above and below ground around the world. My cans, I was told, were stored at a facility near Saint Louis MO along with millions of military personnel records. From these modern catacombs, the cans I had selected (Figure 1) were pulled from deep underground and then sent up to the surface. An Archives plane takes off daily from Saint Louis, delivering these materials to the Washington area.
Given the expense involved in storing, accessing, and transporting these cans, its understandable that the National Archives limits public users to approximately 10 cans within one week. The cans in Figure 1 were from 1943 and represent the first American air photographs taken over Vietnam.
Besides air photography, the United States in 1995 declassified its entire collection of spy satellite photography. Discussed in the companion Gallery essay, these images are particularly valuable for their coverage of vast areas at high resolution in the late 1960s. Since declassification, the National Archives preserves the original film while the U.S. Geological Survey allows public users to obtain frames using its online data finder, Earth Explorer. For scholars working outside of countries such as France and the United States that provide free, accurate imagery as a public service, such images are invaluable.
Besides obtaining historic images, there are significant technical and conceptual challenges involved in using them. In order to compare different air photographs of a given landscape at different times, one has to find a way to compare them in the same spatial framework. Geographic information systems (GIS) are the most popular means to do this. As a kind of database with important visualization and geo-referencing the digital (raster) images derived by scanning, re-photographing, or other means. Any introductory GIS textbook contains a chapter or two about the mechanics of assigning a coordinate reference system, identifying anchor points between an unknown and a known point in space, and ultimately choosing from one of many mathematical equations to re-project a raw photo into geographic space. Figure 2 below shows the same Corona satellite frame in a linear or affine transformation (top) and a 2nd order polynomial (curved) transformation (below).
Working with spatial imagery requires one to consider authenticity or accuracy of representation in a very different way. With spatial imagery, one has to consider margins of spatial error as well as representational error. When looking at a feature, such as the detail of a helicopter pad at an American base (below), if the resolution of the image is too low, how can we be sure? Further, what does an image permit us to say about the landscape: paved, bare ground, sand, clay, built up? The view on the left is from the satellite photo in Figure 2. The comparable image on the right is from a low-altitude air photo taken in 1972.
The Conceptual Challenge: A Visual Turn in Environmental History?
Assuming we can build up a GIS catalog of historic imagery sources such as maps, air photographs and satellite imagery, what then do we do with this information? Here I think is the primary challenge for environmental historians, historical geographers and others concerned with processes of environmental change. The field of environmental history is largely oriented to the phenomenological flow of stories rather than static, areal depictions of the Earth’s surface. Yet images and maps of the earth’s surface can tell us a lot about both changing land uses over time and spatial relationships within a given moment. Incorporating a temporally static, areal description of a given landscape into a narrative about phenomenological processes — ecological and human — shaping it has been a recognizable goal of geographers well back to the days of Carl O. Sauer at UC Berkeley. Sauer’s classic essay “The Morphology of Landscape” argued for a phenomenological view of landscape as a central preoccupation for geography as opposed to more narrow functions such as chorology and geodesy. Only a more holistic account of the intertwined cultural and physical processes shaping landscapes could accurately describe how and why a given area changed from one form or state to another.2 Sauer’s writing reflects the deep schisms that still divide many geography departments, but his ideas have helped to shape the bases of inquiry in newer sub-disciplines such as environmental history, environmental anthropology and human ecology.
Yet these newer heirs to Sauer’s phenomenological perspective, especially environmental history, have yet to make full use of the spatial or areal elements of the archives. Maps in environmental history, with some notable exceptions, serve the most basic of functions – showing general locations of historic events. In those notable exceptions, mainly works of historic visualization, we may be amazed when gazing upon a visualization of what Manhattan may have looked like in 1609. Digital mapping tools even permit a flythrough of a 3-d visualized landscape, allowing the viewer to fly like a bird through a digital time machine.3
Unfortunately, I have not yet arrived at a comprehensive solution to the problem of space, nature, culture and phenomena as discussed by Sauer and many others. However, in my own research I have developed two primary ideas about the uses of spatial imagery. First, historic air photos carry the potential to uncover new “places for stories,” to borrow from William Cronon’s well-read essay.4 By this I mean that a spatial image as a photo-graph does not discriminate in choosing which features of the landscape to record or not. If it reflects sufficient light and fits within the scope of the lens, then it’s in. Historic air photographs not only show us what we’re looking for; but in many cases they also show us things we didn’t expect to see. Since the advent of aerial photography in the early 1900s, archeologists have used them to delineate crop markings showing the outlines of ancient roads and foundations. In England in 1906, an exhibit of photographs taken by balloon over Stonehenge revealed for the first time a system of roads and other structures stretching across the heath. Over the next few decades, aerial photography played a central role in re-writing the history of Neolithic and Bronze Age Britain.5
How might photographs of modern historical places such as California or Vietnam in the 20th century open up new places for stories? In my own research of historic and contemporary images of the landscapes around Huế, I am repeatedly struck by the incredible persistence of seemingly soft features such as field boundaries, estuarine impoundments (dikes) and bamboo hedgerows. Comparing maps from 1927 with photos in the 1950s and satellite imagery today, these lines are for the most part fixed while urban and industrial spaces including military bases go through many changes. More along the lines of the crop markings at Stonehenge, photos also reveal persistent traces from earlier occupants. These include the ruins of old citadels from the 14th century and family tombs that survived inside the fence lines of American bases and are still visible today. Aerial photographs more than textual records and even topographic maps can help us appreciate the layered nature of landscapes.
My second observation on potential uses of aerial photography in environmental history stems from more of an activist interest in environmental sustainability and addressing what I perceive as residual problems from the past documented in historic imagery. As Sanderson and others have acknowledged with the Manahatta project, visualization can serve as a powerful motivational tool. It stokes the imagination, allowing one to experience a god-like distance and visual perspective relative to objects on the ground. There are obvious shortcomings to people assuming this perspective, what Donna Haraway calls the god-trick. Her writing on “The Persistence of Vision” gets to the core of this problem:
The eyes have been used to signify a perverse capacity-honed to perfection in the history of science tied to militarism, capitalism, colonialism, and male supremacy-to distance the knowing subject from everybody and everything in the interests of unfettered power. The instruments of visualization in multinationalist, postmodernist culture have compounded these meanings of disembodiment. The visualizing technologies are without apparent limit… But, of course, that view of infinite vision is an illusion, a god trick. I would like to suggest how our insisting metaphorically on the particularity and embodiment of all vision (although not necessarily organic embodiment and including technological mediation), and not giving in to the tempting myths of vision as a route to disembodiment and second-birthing allows us to construct a usable, but not an innocent, doctrine of objectivity.6
By digitally resurrecting satellite and airplane photography over the war-scarred landscapes of Vietnam, my intent is not to facilitate nostalgic god-trickery, although I do admit to a degree of pure childish wonder at bird’s-eye views. Rather, my activist concern is more focused on the post-war silence surrounding the full extent of war-related environmental impacts in Vietnam.
While the United States and other countries including France possess well-catalogued, mostly-classified collections of imagery that could be used to locate such persistent hazards as military base landfills, unexploded ordnance, bomb craters and herbicide spray sites, these governments have yet to declassify and make public their collections. The same images that once permitted a bird’s-eye view of the battlefield in 1969 now via archives can potentially offer concerned citizens a view into their community’s environmental past. I am not sure that any state, including Vietnam, is excited for people to have such a clear view of a given landscape’s potential hazards. Nevertheless, access to such information played a powerful role in citizen-led campaigns for environmental cleanup in the United States, Europe and Japan; and it may facilitate similar efforts in less affluent, more war-torn places such as Vietnam.
By incorporating spatial imagery into the analysis, environmental historians can find new places for stories. Spatial imagery is inherently story-less in the sense that a single image typically contains no text, and only in comparing images can someone detect patterns of change or persistence. However, by detecting new patterns in studies of historic imagery, we can greatly enhance the diversity of storylines, perhaps even avoiding tendencies to fall into declensionist or techno-positivist trajectories. Along the same lines, air photography lacks the familiar geo-lines of national boundaries and place names. Thus by contemplating the areal patterns and features captured in a single image, we are leaving the boundary lines and place names of the derived map for a less structured space.
2. Carl O. Sauer. “The Morphology of Landscape,” University of California Publications in Geography, Vol. 2, No. 2, pp. 19-54 (1925).
3. One great exception is the Spatial History Project at Stanford. However, a reading of the papers published by the Project shows a general tendency to quantify or categorize areal changes over time with less discussion of underlying logics or phenomena. Most of the work aims at producing historic visualizations using maps, oblique, bird’s-eye views, and reliance on GIS software. Another popular example of such historic and spatially-referenced visualization is the Mannahatta/Welikia Project. The Wildlife Conservation Society collaborated with other institutions and many volunteers to visualize New York City’s pre-European ecology and landscapes. See also Eric W. Sanderson. Mannahatta: A Natural History of New York City. New York: Abrams, 2009. An excellent critique with Sanderson can be found at Reuben Rose-Redwood. Encountering Mannahatta: A Critical Review Forum. Cartographica 45:4, pp. 241-273.
4. William Cronon. A Place for Stories: Nature, History and Narrative. Journal of American History, March 1992, pp. 1347-1376.
5. For a comprehensive history of English aerial photography and archeology see Martyn Barber. A History of Aerial Photography and Archeology: Mata Hari’s glass eye and other stories. London: English heritage, 2011.
6. Donna Haraway. Situated Knowledges: The Science Question in Feminism and the Privilege of Partial Perspective. Feminist Studies, 14:3, pp. 575-599.