I’m going to acknowledge up front that my theme this time may be a little bit tenuous. I’m talking broadly about awareness as much as literal sight, but a number of the things I’ve been thinking about do have to do directly with the visual realm.
“On the Nose”, by Chris Ip, describes the work of Sissel Tolaas, an expert on smells. She’s an installation artist whose work has appeared in numerous museums, as well as a consultant to corporations specializing in scents and flavors. In the piece, Ip and others are following her around Detroit as she smells various things, including several that would seem to be viscerally unpleasant if not downright dangerous to inhale (insulation from abandoned buildings, for example). The visit is part of a project in which she creates “smellscapes” of individual cities.
Tolaas believes that “most people are increasingly disembodied, out of touch with their natural senses, with the real world. Through her work, she wishes to put us back in our bodies.” Ip argues further that smell, in particular, is ignored or denigrated in favor of hearing and, especially, sight, and that this is both culturally and temporally specific:
In the West, the sensuality of smell has also relegated it below the supposedly higher, rational senses of vision and hearing. Kant, Freud and Darwin all characterized scents as animalistic and primitive. Then, with the Industrial Revolution, workers flocked to cities, away from the odors of the natural world, while the mass hygiene campaigns of the 19th century made cleanliness not only a public health issue but also a moral one. The middle and upper classes took pride in the fact that they didn’t have to do manual labor and had access to modern plumbing to wash regularly.
But the last century of consumer technology has embedded this hierarchy of the senses into our everyday devices. We have exponentially augmented our sight (through TVs, virtual reality), sound (phones, voice assistants) and touch (keyboards, controllers) but hardly smell nor its cousin, taste.
The result, said [anthropologist David] Howes, is an “unbalanced sensorium” — today’s humans have overpowered eyes and ears while our other interfaces atrophy. Modern computing — which is smooth, sterile, anti-odor — has allowed us to connect with one another and the world at great distances and speeds but mostly through aural and visual means. So we are, increasingly, an aural and visual culture.
I’m not sure how far I buy the notion that one sense “atrophies” because of a cultural emphasis on others, but perhaps I am reading that word in too literal or medical a sense. The idea, though, that we value certain kinds of sensory data more than others, and that this has something to do with the affordances provided by contemporary technology, seems plausible. Tolaas herself seems to see her work in these terms, as a way of drawing attention to aspects of our environment that we typically ignore:
“My work is about life and being alive,” she said. “Information that surrounds all of us, how to focus on that properly. Which from the very beginning, that was a concern. What is the potential in the air that you breathe? How does it determine which life we have? How does it determine who we are as humans and our role toward each other?”
Our sense of the world in which we live (and I am struck here by the way the meaning of the word “sense” is fluctuating as I write) is a result of a process of selection or filtering, by which some elements are noticed and remembered and others drop out of the picture. Some of that process is directed by evolution, focusing our attention on potential threats— or simply preventing us from becoming overwhelmed by the sheer amount of information that we are capable of perceiving about the world. But some of it is also a matter of culture and habit, and Tolaas’s work is interesting for the ways in which it tries to subvert those habits.
The priority of the visual is also important in “Afterimage” by Joshua Rothman. The piece discusses the rapid development of image synthesis, by which whole images, or parts of images, are fabricated with increasing sophisticaltion by computers, and artificial intelligence systems in particular. Rothman’s focus is mainly on the ways in which these technologies undermine the reliability of images (and videos) as representations of the facts, though we continue in general to attribute great credibility to them. One of the subjects of the piece is Hany Farid, an expert in images synthesis and photo forensics, and he asks:
“Why did Stalin airbrush those people out of those photographs?” he asked. “Why go to the trouble? It’s because there is something very, very powerful about the visual image. If you change the image, you change history. We’re incredibly visual beings. We rely on vision—and, historically, it’s been very reliable. And so photos and videos still have this incredible resonance.” He paused, tilting back into the sun and raising his hands. “How much longer will that be true?”
The answer seems to be “not much longer,” and that we are not remotely prepared for that reality.
I’m also interested in what Rothman says about why image synthesis has improved so much recently. Part of the answer is predictably, increased computing power, particular the development of very fast, relatively inexpensive GPUs; part of it is also the vast quantity of visual data that we are all putting online, which serves as a library for the training of neural networks learning to generate more realistic images. One reason that those systems are becoming so successful, though, is because of human predictability. There are some fascinating ideas about “texture” here— that almost any predictive system (which is what image synthesis involves) can be thought of in terms of generating a texture from a very small number of elements, repeating in a number of different possible arrangements.
“But, you know, we’re boring” [says Berkeley computer scientist Alexei A. Efros]. We always build the same kinds of buildings on the same kinds of riverbanks. And then, as we walk over bridges, we all say, along with a thousand other people, ‘Hey, this will look great, let me take a picture,’ and we all put the horizon in the same place.”…One of the lessons of image synthesis is that, with enough data, everything becomes texture. Each river and vista has its double, ready to be sampled; there are only so many faces, and your doppelgängers have already uploaded yours. Products are manufactured over and over, and new buildings echo old ones. The idea of texture even extends—“Zzzzt! ”—into the social dimension. Your Facebook news feed highlights what “people like you” want to see. In addition to unearthing similarities, social media creates them. Having seen photos that look a certain way, we start taking them that way ourselves, and the regularity of these photos makes it easier for networks to synthesize pictures that look “right” to us.
I thought here of a recent Teju Cole column, in which he noted that
People don’t merely go to the same places or take photographs of the same monuments and sites; they take photographs of the same monuments and sites in the same way. This applies to tourist sites, public spaces and ordinary buildings. The same gestures and vantage points and compositions are repeated, and the images come out so uncannily similar that it’s as though everyone were subject to the same set of instructions.
Cole argues that this is in part a straightforward response to the physical arrangement of sites: if you’re taking a picture of the Statue of Liberty, you probably want the front, and if you’re taking a picture of a city or other large site, you probably want to get up high, and there are only so many options for doing that. (Cole also uses the concept of “affordances” to talk about this). But it’s also the influence of seeing so many images of particular places, a repetition or patterning that creates a specific, if unconscious, set of expectations about what we will see in an image of a particular place, and so tends to make certain views or angles look more “right” to us than others. The consistency of both the places we construct and the images we create of those places makes the task of synthesizing images easier, because patterns can be replicated while randomness is, well, random.
I’ve also recently begun listening to the podcast The Anthropocene Reviewed, hosted by author John Green. The premise—Green “reviews facets of the human-centered planet on a five-star scale”— might sound a little hokey, and the range of those facets— pennies, diet Dr. Pepper, cholera— might also contribute to the impression that this is not intended very seriously. But actually, the star ratings are a very minor aspect of the show, which is mostly Green explaining the history of his subjects and explaining why they are in some sense a product or result of the anthropocene world. This might seem obvious for, say, diet Dr. Pepper, but Green makes a pretty compelling case for the drink as the perfect beverage for the anthropocene, because it is an imitation of a product (regular Dr. Pepper) which is itself not an imitation of some natural flavor, as almost all other sodas ostensibly are— its flavor is an expressly human creation. A less obvious example, from the same episode, is the Canada Goose, which is of course a wild creature but whose numbers and geographic reach have grown dramatically because of the proliferation of human-made landscapes, like golf courses and parks, which provide ideal environments for them. Cholera, similarly, is obviously a natural phenomenon, but can only spread in global epidemics because of both the connectedness of human societies and the failure of wealthier parts of the world to extend improvements in hygiene and infrastructure to poorer ones.
The admittedly tenuous connection to my theme here is that, by drawing connections between the individual subjects he chooses, Green is in effect making the anthropocene visible. Since it is quite literally the world we live in, the air we breathe and the water we drink, the only way to see it is to think counterfactually, considering how it might be different— and in fact was different, not very long ago. Historicizing mundane things, or things that seem inevitable, reveals them as contingent, and so the world which they make up as contingent as well.
The last two pieces are likewise more about awareness than seeing as such. In both “America’s Other Family-Separation Crisis”, by Sarah Stillman, and “U.S. Law Enforcement Failed to See the Threat of White Nationalism. Now They Don’t Know How to Stop It”, by Janet Reitman, a problem persists because its scale is unknown, unappreciated, or simply ignored. Reitman’s piece is about exactly what it sounds like: a single-minded focus on Islamist terrorism led American law enforcement to ignore the growth of far-right white supremacist movements, despite the fact that people motivated by these ideologies have killed significantly more Americans than those motivated by ISIS or similar ideologies. Incidents of violence, like Dylann Roof’s attack on an A.M.E. church in Charlotte, are characterized as acts of disturbed individuals, rather than the result of ideological radicalization by organized groups. Attempts to bring more attention to the problem were castigated by “conservative” politicians and pundits as a conspiracy to demonize conservative or right-wing voices. (The way far-right groups have managed to get people to take seriously the claim that criticism of them is simply leftist intolerance of “conservative” ideas is not really addressed directly here, but it is striking— and should be very troubling to people on the right). Here, there’s a sense that white supremacist terrorism is not visible as such, simply because it is not named as terrorism.
Stillman’s piece is about the high numbers of mothers imprisoned in the United States, many of whose children are in care as a result. Many of these women are in jail for petty offenses, or for failure to pay court imposed fines and fees; often, child abandonment or endangerment is added to their charges purely as a consequence of trying to deal with their arrest. Most strikingly, women who are victims of domestic abuse may often be charged with child endangerment for not getting away form their abusers, and then get much longer sentences than the men who are clearly actually the problem. I suppose the connection to the theme here is just that the effects on children are an often-invisible consequence of mass incarceration, but mostly I am pointing to this piece because it really pissed me off and I want people to read it.
A couple of other, unrelated recommendations:
“Deep River”, by Will Bostwick, describes the efforts of the Black Gospel Music Restoration Project to collect, preserve, digitize, and catalogue music from the “golden age” of gospel music (1945-75). Musicologists estimate that 75% of the music from this era is already lost.
Selva Oscura, by William Basinski and Lawrence English
If you have any interest in ambient/drone/experimental music, you already know these two names, and their collaboration is as good as you’d expect it to be.