Q&A: Robert Pietrusko on Proxy Landscapes

From April 10-11, Weitzman’s Ian McHarg Center for Urbanism and Ecology will present Proxy Landscapes, a symposium exploring how landscapes and landscape features have been used to represent and stand in for remote places and conditions. The symposium, organized by Associate Professor of Landscape Architecture Robert Gerard Pietrusko, will bring together artists, landscape architects, architects, planners, historians and anthropologists. Susan Schuppli, director of the Centre for Research Architecture in the Department of Visual Cultures at Goldsmiths University, London, will give a keynote lecture on landscape indicators and nine other scholars will present their research in a series of panels.

Pietrusko, who joined the Weitzman faculty in 2022, has long been interested in military test sites and other uses of landscapes as information systems. For him, the symposium is a chance to strengthen connections between scholars working in disparate fields on questions related to landscape representation, remote sensing, and the prehistory of data science. “The question I have is, what’s the bouquet of case studies that one could use in order to argue for the primacy of the landscape in this history?” he says.

Pietrusko recently spoke with Weitzman News about his research and the upcoming symposium. The conversation has been edited for length and clarity.

What is a proxy landscape and how did you get interested in studying that concept?

I noticed in a number of different domains during the early Cold War era, there was this epistemological thing that kept showing up. Something was always standing in and being an indicator of something else that was hidden or latent. The idea wasn’t invented in the immediate Cold War era, but something about this moment made it very resonant. You see it happen with Freudian ideas—that outward activities and personas are always representations of a deeper internal, psychological state that one doesn’t have access to. Things like Freudian slips or even dreams—behaviors are always symbolic of something else. At the same time, there was a renewed interest in Marxist categories and the idea of the commodity fetish: an object that is a physical embodiment of a whole set of social relations that you don’t have access to, and is a stand-in for them, but needs to be deciphered and decoded or analyzed. It went along with a certain kind of Cold War military approach but also a kind of cultural paranoia, that you look at your neighbor and you don’t actually know what political allegiance is lurking within them. There’s something deeper, latent, that’s unavailable, potentially dark, and needs to be deciphered.

I was looking at a set of projects in the early 1950s because they were the precursor to remote sensing, satellite-based remote observation of the earth, which is something I’ve long been interested in. Botanists, geographers and intelligence analysts were all collaborating, and they were doing two things. First, they were taking concepts from the early days of plant ecology, specifically that plants were indicators of underlying hidden ground conditions. And second that the landscapes they were working in could be proxies for areas that were denied to them because of politics. If they were attempting to map the Sino-Soviet bloc, they instead mapped landscapes they had access to that had similar natural categories. That kind of approach, of treating one place as a test site for some other remote spot with common features first really found its ground at this moment. There’s an epistemological category of the proxy that shows up in the Cold War and then it filters through all sorts of other practices. We now take it for granted that we can use one place as a stand-in for other places.

“There’s an epistemological category of the proxy that shows up in the Cold War and then it filters through all sorts of other practices. We now take it for granted that we can use one place as a stand-in for other places.”

Can you give me an example of a place in that era, a landscape that was used as an indicator of something that was happening in a different place?

Because of the elevation, researchers used the Pocono mountains as a climate and vegetal surrogate for parts of Siberia. They needed information about the terrain. Now, we map terrain using radar from orbital altitudes, but back then they didn’t have that. The basic idea was this: They believed that at some point there would likely be a ground war between the US and communist adversaries. And they would need to have very accurate terrain maps of Siberia to know where they could put boots and tanks and whatever else to move across the landscape. So, they flew reconnaissance missions over the Poconos, took reconnaissance photographs, and analyzed them looking for patterns in vegetation. They would say, “OK, this species of coniferous tree corresponds to this particular slope, or this other species likes well-drained soil, whereas this other one likes it wet and muddy.” They couldn’t see the ground in the photographs, but the vegetation told them everything they needed to know about its conditions. This was useful because areas around the globe with similar climates and soils have similar vegetation. So they used the Poconos as a stand-in for Siberia and local vegetation as a stand-in for the hidden ground. Then when they flew reconnaissance flights over Russia, which they did, or found historic aerial imagery with vegetation in it, they could say, “if we can classify these tree species, then we know the ground is going to be.” They enlisted vegetation for military intelligence about territories they couldn’t access.

One of the points I find fascinating is that the basic ideas weren’t invented in this Cold War moment, there’s a sort of bricolage that happens. A number of disciplines already had these indicator concepts in them, but they all come together in this moment because they are useful in a military context, and then they resonate against each other and become one of the main ways of knowing the world. Especially for environmental analysis, we find that the epistemological conceits that emerged out of the Cold War are now just how we look at the earth. We still look at the surface of the earth as a type of information source about other things we’re unable to access, such as the climate.

“We look at the surface of the earth as a type of information source about other things we’re unable to access, such as the climate.”

Who else is doing this work?

A lot of media studies folks are interested in proxies, especially now with climate data, and data science. I think the relationship between data science and environmental proxies is closer than they might appear. I described a particular workflow before. You have a training set like the Pocono mountains. The aerial photographs are like an observable input sequence that have some probabilistic relationship to a latent but known underlying ground condition. They trained the photo-interpreters to learn this correlation. Then they applied the process to a real-world problem like Siberia. They again had an observable input set of aerial images, but they didn’t know the actual state of Siberian ground, they just hoped the process had produced sufficient statistics to accurately make the correlation. This process is very similar to how we now train machine-learning and pattern-recognition algorithms and AIs. One of the things I’m arguing for is the botanical roots of data science. In a former life I studied statistical pattern recognition and was surprised to find such a similar approach in these historical case studies. So with proxies, I like to imagine there’s some Venn diagram in which folks in media studies, folks in landscape architecture who want to have a historical understanding of the datasets they use for understanding the natural world, and historians of the Cold War overlap. The years I work in are tied to a particular historical context where the boundaries between these ideas and disciplines were not clearly marked.

How has thinking about these proxy spaces and or stand-ins changed the way you think about landscapes?

I’ve always been interested in the way that people talk about reading the landscape. The tradition in geography of cultural landscapes in particular, like the early work of [British cultural geographer] Denis Cosgrove. It was really about reading the landscape as a manifestation of our underlying political and cultural assumptions and economic organization. I think framing landscapes as a proxy extends that idea and connects it to all sorts of other practices. Of all the different ways that we think about landscapes, thinking about landscapes as an information source and landscapes as media actually changed the way I thought about environmental data. At first, I was interested in representations of landscapes. Then it flipped, which gave me a different way of thinking: representation as landscape. That was big for me.