A founder of Rhizome, Mark Tribe is known for his early contributions to the field of new media art and his socially-engaged performances and installations. His current practice engages the power of aesthetic experience to illuminate the challenges we and future generations will face in the climate crisis. Since 2012, he has made landscape pictures that unpack American ideas about nature and land, from Manifest Destiny to contemporary environmentalism. In this interview, Mark talks about his views on the climate, his landscapes, and his integration of machine learning tools (AI) into his latest project Learning to Love the Future.
Could you just give a brief definition of solastalgia and tell me how your thinking about it has evolved since we last discussed your work?
I don’t remember who coined the term solastalgia, but it’s been used to describe feelings of loss, longing and grief that people experience when their homelands are threatened by climate change and other kinds of environmental destruction.
Where are you now in your thinking about the nature of climate change and loss?
Like a lot of people, I had been feeling pretty pessimistic about the future of nature, particularly the future of wild places like the woods I grew up with in the Northeast. As recently as five years ago there were still a lot of people denying climate change, and there was very little hope that we would be able to pivot and slow it. I was convinced we were headed for global ecological catastrophe: loss of habitat, ecosystem collapse, mass extinction and possibly our own extinction. I do a lot of hiking, and I started to notice that my experience of the outdoors was tinged by a sense of dread and a heightened awareness of how precious and impermanent wild places are. I was haunted by the idea that our descendents would never know what the great forests of North America were like at the dawn of the Anthropocene, when things were just starting to go haywire climatewise. That sense of despair mixed with appreciation made me want to capture that experience in paintings and in audiovisual recordings.
But in the long summer of 2020, when COVID lockdowns had stalled the global economy and those who could were working remotely and fleeing cities, people somehow seemed to wake up, collectively, to the real urgency of the climate crisis and to realize that things could change much faster than they’d imagined. I mean, wild animals were roaming suburban streets, carbon emissions were plummeting, the Black Lives Matter Movement was going mainstream… suddenly, another world seemed possible, and I started to feel, yeah, more upbeat about the future, our ability to adapt, and our resilience as a species. On the other hand, I still think we are facing a couple of centuries of chaos: terrible storms, massive wildfires, epidemics, mass migrations. It’s going to be tough, but in the middle of the catastrophe there could also be really amazing creative responses, experiments in new ways of living… pocket utopias, to borrow a term from the artist Austin Thomas. I’m imagining things like community food forests, floating cities and urban rewilding.
So tell me, how has this optimism about the future, with a caveat that it’s going to be very difficult, entered into your practice?
Well, lately I’ve been using AI image generators to imagine future landscapes that I might actually like to see.
How does that work?
I tried out Dall-E and Stable Diffusion, two of the AI-based image generators that became publicly available last year, and settled on Stable Diffusion. I wasn’t liking the images it was making when I gave it one of my photographs and suggestive text prompts, so I fed it phrases generated using an AI prompt generator, so the prompts basically described the image I’d given it. Then I played around with other settings to see what the model does when it’s sort of left up to its own devices, and the results got interesting. I mean, it started to make images that I wanted to paint. Could I show you an example?
Wow! So this image was produced by your AI process?
Yes, this is one of several pictures I’ve made with an open-source web interface to Stable Diffusion called Automatic1111, using landscape photos and self-referential prompts. After tweaking the resulting images and converting them to black and white in Photoshop, I print them on canvas, then paint over the print with thin layers of acrylic using the glazing techniques I’ve been refining for the past several years.
It’s difficult for me to articulate why I find these images so compelling. I look at a lot of landscape paintings, and was very influenced by the Hudson River school, but more recently I’ve been looking at French landscape paintings, particularly Corot and other Barbizon painters, and also contemporary landscape painting. So this image is filtered not only through AI but also through the aesthetics of contemporary painting. You look at this for a while, and it doesn’t actually look like an image from nature, but it looks similar. It’s naturalistic, but it has this quality of being filtered through some kind of unnatural perceptual or mimetic apparatus.
My reaction to this is I know I’m looking at a landscape, but to what degree is it something that’s produced by hand or by other means is absolutely not clear to me. It plays on my familiarity with something photographically produced and something painted.
Right. It lies in this uncanny valley, and flips between these different kinds of images, photography, virtual reality and painting.
Could you say a little more about your painting process?
In college I took a class on Renaissance painting techniques and learned how to make oil paint from scratch, how to prepare a panel, and how to paint with glazes. You begin with a reddish ground, then make a grayscale underpainting, which ends up looking kind of like a charcoal drawing of your subject. Once that dries, you start layering over thin, translucent layers of color to build up the flesh tones, or the rich hues of draped fabric. A face, for example, might include alternating layers of red, yellow and blue. Light passes through these layers, bounces off the opaque underpainting, then travels back through the colored glazes. Instead of mixing colors on the palette and then applying them to the surface, the colors are applied separately, one on top of the next, so there’s this kind of optical mixing that gives those paintings chromatic complexity and luminosity.
Years later I was scouting locations for a landscape recording, and was shooting still photographs as part of that process. It occurred to me that I could make paintings based on those photographs using the Renaissance glazing techniques I’d learned at RISD, but instead of doing a grayscale underpainting, I could print them on canvas in black and white. I wanted to make the images into objects that draw the viewer in and reward a longer, more contemplative look.
How do these paintings relate to your video practice, what you call landscape recordings? What’s the relationship between them?
They’re both about preserving a kind of experience that, like nature itself, is endangered. With climate change, forests and other ecosystems are changing, no matter how carefully we try to protect them. But through art, we can preserve something of what they’re like, what it’s like to just sit in a woodland meadow and experience the play of light as it falls through the trees. So there’s real value, I think, in capturing these landscapes for the present and the future.
I want to drill down on these video recordings, because they fit nicely with your framing of how complicated the idea of natural versus cultural is. I also wanted to throw in your mindfulness practice and ask how that might be related.
The recordings are very straightforward, in a way: I place a digital cinema camera on a tripod in the woods and record for 24 hours: no camera movement, no cuts, just a high-fidelity audiovisual recording of a forest. The recordings themselves are cultural artifacts, of course, but it would be hard to argue that the landscapes they represent are entirely natural. Humans arrived in North America at least 13,000 years ago, probably much earlier. The idea that indigenous peoples had no impact on the land is a fantasy inherited from colonial settlers—one that’s been used to justify an approach to conservation that excludes Native Americans. Agriculture, towns, trade, even hunting and fishing: humans have been intervening in the ecosystems of this continent for a long time. And with industrialization and extraction, this process has accelerated exponentially.
So nature and culture exist on a spectrum, but there is still something precious about relatively wild places. I’d like to think they have inherent value, independent of us—the lives of the individual organisms, the remarkable diversity of species, the intricate webs of life. And they clearly have value for us. We evolved in nature, and the idea that experiencing nature is good for us isn’t just a product of 19th-century Romanticism; it seems to be a fact borne out by empirical research. And that relates to what you were getting at when you asked about mindfulness. Making the recordings involves mindfulness–paying attention to what’s happening as it’s happening. I’ve noticed that after the sun sets, and I’ve been sitting by the camera for 18 hours or so, and it gets dark again, my awareness becomes really heightened, and I start to hear and see things that I wasn’t aware of in the wee hours of that first night.
I show the recordings in real time, so it would take a full 24 hours to see the whole thing, and I synchronize playback to the time of day, so at 11am, say, the viewer sees and hears what was happening in front of the camera at 11am. Things happen—wind blows, leaves flutter, eventually the sun sets—but they happen slowly. People often sit for longer than one might expect and seem to find it quite relaxing, even moving sometimes. I’d like to think that the work encourages a kind of contemplative experience.
Could you talk a little bit about the process of actually making one? In an earlier conversation I was impressed by the complexity, especially since the end product seems so transparent.
It’s true that the recordings are deceptively simple. It’s a lot harder than you’d think to record 24 hours of high-quality video in the backcountry and get something that’s actually a pleasure to watch. I do quite a bit of research to find locations that meet my criteria. For example, they have to be quiet, which is to say far from roads, airports, even campgrounds. Then I explore the area on foot, usually on my own, for a few days to select the exact location, looking for a spot that will be visually interesting at different times of day and night. Next I schedule the shoot for a full moon, rent a camera, lens, field monitor, memory cards, microphones, audio recorder, batteries and a bunch of other gear, then head back to the location for a few days with a small crew.
After the shoot I grade the color and output two versions: a set of uncompressed files as a preservation master and a set of compressed presentation files that get installed on an industrial-grade video player that syncs to the time of day and can run pretty much indefinitely. The project is complete when it has been accessioned into the collection of a museum for long-term preservation.
So where are the locations?
I shot the first one in the Catskills: Balsam Lake Mountain Wild Forest in Upstate New York. That was 2016. Then in 2019 I made a second recording in the Cascade-Siskiyou National Monument in Southern Oregon. I’m planning to make a third recording in a stand of first-growth forest in the Adirondack Mountains.
Let’s circle back to Artificial Intelligence. We are in very interesting territory with AI now. It’s clearly not conscious or sentient, but it is sentient-ish in the sense that it gives us answers that we don’t expect.
Consciousness in particular is difficult to define: more difficult, I think, than intelligence. But what I find most fascinating about this new crop of AIs is how it feels to work with them. I know Stable Diffusion isn’t conscious, but after using it, interacting with it for a while, it almost seems to have a mind of its own, as if it had some kind of unnatural creative agency. It’s unpredictable. In that sense, it’s different from any other tool I’ve used.
These so-called deep learning models are incredibly complex, and they use some spooky math, like neural stochastic differential equations, that I don’t pretend to understand, but the upshot is that they can surprise you, which makes working with them feel almost like collaboration sometimes. We humans have a tendency to anthropomorphize, and I’m sure that plays a role in what I’m talking about, but as AIs get smarter and more capable, they’re going to feel more creative, more autonomous, more… human. And that is going to change how it feels to be human, how we understand ourselves. Something similar is already happening with social media as the algorithms that maximize user engagement affect what we see, which affects our behavior, our emotions and desires. Our brains are malleable, so social media can actually affect things like attention and memory. It changes our brains. I think AI is going to change us in ways we can hardly imagine.
Human nature is affected by the technologies we produce. We’re becoming cyborgs, and that’s scary, but it’s also interesting. I think about the interesting things that the future might bring, what it’s going to be like to go for a hike in a forest that has been geoengineered to sequester more carbon. It’s hard to imagine. But having spent the last several weeks interacting with unnaturally intelligent machines, using them to imagine landscapes of the future, I feel like it’s starting to come into focus.
About the writer: Hovey Brock is a painter, climate artist, and writer who has shown his works in the US and internationally. He is a frequent contributor to the Brooklyn Rail’s Art Seen column. He is currently one of the editors of Hot Air, the climate column for Art Spiel. His current project, Crazy River, which includes painting and writing, looks at the climate crisis as it is unfolding on a river he has known all his life, the West Branch of the Neversink, using the filters of personal memory, historical incident, and geologic time