A conversation between Shane Denson, author of Discorrelated Images (Duke University Press, 2020), and Caetlin Benson-Allott
Caetlin Benson-Allott (CBA): It’s my pleasure to be sitting here this afternoon with Shane Denson to talk about his new book Discorrelated Images, just out from Duke University Press. Shane, I want to begin by praising the intervention that the book makes to our understanding of post-cinematic media regimes and the way in which you’re extending the concept of post-cinema beyond the issue of format to a pre-personal, sensory epoch as well as a perceptual epoch. The archive of your book is especially stunning. It ranges from international art films to Hollywood blockbusters to NASA photographs and drone surveillance to art installations and video games. So the first question I wanted to ask you was how do you understand—or how did you, as you were working on this book, understand—your archive? Were there particular principles or structural imperatives that guided you towards the case studies you’ve collected?
Shane Denson (SD): Yes, thanks a lot, Caetlin. First of all, thanks for taking the time for this conversation. I’m really happy to have this opportunity to talk to you about the book. We’ve been in conversation about some of these ideas over the years as they’ve developed; for this reason and because of the relevance of your own work on electronic image and media cultures, I think you are in many ways the perfect interlocutor for this. We should also recognize the exceptional circumstances in which we meet today, over Zoom, just days after a coup attempt and in the midst of a global pandemic, as well as widespread social unrest and everything else. So, thank you for taking the time.
Regarding the question of the archive, I think this is a really important question, and it goes straight to the heart of what I’m trying to do in Discorrelated Images. Like you pointed out, with the wide range of media texts I look at in the book, it is a very eclectic form of archive to have constructed or to try to work through. The short answer, as to why I would have such an eclectic archive, is because I am trying to get a grip on what I see as an incredibly broad and foundational shift in the parameters of the visual—or the sensible or the perceptible more generally—and this is a shift that really has no regard for the distinctions that we make between various forms, genres, or other cultural distinctions of value or even medium. This is a shift that is taking place across all areas of culture, and it has to do with the fact that all of our images now are transformed when they come into contact with computational processing. Of course, there are still exceptional images out there: there are images on canvas; there are images on celluloid still. But most of our contact with them is through digital screens and other interfaces. What I’m arguing is that the computational processing that underlies and enables our access to images today impinges upon us at a bodily level that bypasses our consciousness and really goes straight to the pre-subjective parameters of our processing of sense or sensory information.
So what does that have to do with the archive? I’ve already said it cuts across all these different categories, but I think it’s also important to think about this question of the archive in terms of methodological questions. One of the ways that I’m trying to approach these very different images and the way that they affect us today is by bringing together a kind of media-archaeological focus alongside careful phenomenological attention to sensation, and visual forms in particular. That is because I’m trying to think about the relation of these perceptible forms to the imperceptible infrastructures of our images. The question of the archive is something that informs a media-archaeological approach, originally taking off from Foucault’s notion of the archive as defining the parameters of the seeable and the sayable. I’m really trying to address this question, on the broadest terrain possible, of what is at stake in this massive shift that takes place across all of these genres and media, in terms of what is seeable and what is sayable under conditions of computational image processing.
CBA: Yeah, that makes sense, and I really like your turn to Foucault there—the seeable and the sayable—in order to articulate how you understand “archive,” because as I was asking my question, I sort of pivoted to “case study” as a synonym for a form of engagement with the archive. Then as you were speaking, I started wondering: what are the theoretical genealogies behind these methodological terms? The way in which you’re engaging your archive isn’t psychoanalytic in the sense of that kind of deep object-work: reading the object for its symptoms. So, I think, as you’re saying, Foucault’s notion of the archive offers a much better understanding for how you move betwixt and between such a wide range of objects, showing connections that manifest shifts in perception that we, or perhaps I, otherwise would never have noticed.
SD: Right. That’s the thing. They’re not necessarily shifts that we can see with our eyes. Because often we’re talking about things that are designed and engineered precisely to be imperceptible. Think of things like video compression, or how we’re talking on Zoom right now and the ways that the buffer functions—so that I don’t notice when things have frozen for a few microseconds; Zoom employs sophisticated techniques for masking that. And so really the question becomes: given the fact that we’re not supposed to see these things, how do those imperceptible differences still have an impact on our perception? That’s why it’s really important for me not just, or not solely, to do a media-archaeological study (in the sense that somebody like Wolfgang Ernst and has taken that in a very technicist direction), but also to interface that media-archaeological focus on infrastructure with the visible form.
CBA: Actually, your answer there leads me naturally to my next question, which is about pre-perceptual changes. Most, but importantly not all, of the examples of protentional and discorrelated images in your book emerge from prerecorded rather than simultaneously transmitted media (and I’m obviously bracketing video games as a special case here). So, well, I guess we’ve acknowledged where we are in history right now. Having just watched an attempted coup at the US Capitol a couple of days ago, I couldn’t help but think about how discorrelated images and attendant processes of dividuation might be influencing my perception of that terrorist attack. I’m not going to ask you to comment on that particular live media event, per se, but I was wondering if you could say more about how these post-cinematic processes of discorrelation may be affecting spectators’ perception of live or simultaneously transmitted images?
SD: That’s a great question, and it touches on something that I’m arguing throughout the book: there is a temporal shift that has happened in audiovisual media today, and it has to do with their discorrelation from the subjective temporality that glues together the present for us as phenomenological subjects. This is a shift from what I call the “memorial” function of cinema, which records the past, to the protentional function that you’ve just mentioned, where post-cinematic images in a sense anticipate the future: they’re always drawing something out of this moment that is a couple of steps ahead of the temporal window of the “now,” as it’s lived by the phenomenological subject. There’s a reorganization of our relation to the present, and thus to the past and the future as well. So, in a way, one of the things that I’m arguing is that although we don’t necessarily think of it this way, all of the images that we interface with through digital and computational media are now in fact live—if not even maybe, in a sense, alive. They’re live in the sense that they are being generated right now, before our very eyes—always now and anew with each screening, always again at the level of pixels and microtemporal transformations that therefore render all of the images that we see (at least on our computer screens and on our digital TV screens and other devices) as generative images: actively generated in the now, or in the actual future (from the perspective of how we live as subjects).
I’ve also been thinking about this in terms of recent events. First of all, we can think about Zoom as one of the places where we consume most live images right now. A lot of the innovation of Zoom in comparison to Skype, for example, is in its superior masking of the moment of breakage, of buffering and things like that. This has to do with better video encoding and compression, but it also has to do with different ways of dealing with delays, and catching back up after things have frozen. I’m sure you’ve noticed, sometimes you’ll be talking to someone and then their speech becomes accelerated for a few seconds. That’s the app catching back up, whereas Skype usually would have just had a break, a cut, and then they would be back. So Zoom is playing with this processing of the subjective experience of liveness in this way, and this also obviously has consequences for political images. You mentioned the attempted coup the day before yesterday. I was also thinking about when Trump had Covid and was rushed to Walter Reed, and then—I’m not sure about the exact timeline right now—but I think it was when he returned to the White House; he released a video where everybody rolled their eyes and said, “Oh, come on. This is, you know, cheap greenscreen technology. He’s obviously not outside the White House where it looked like he was, but he’s probably inside—and, oh, here’s this moment where it seems like maybe there was even a subtle morph edit to remove a cough or something like that.” So all this quasi-forensic attention was given to the images, and people were pointing to minute things like the way the leaves in the background seemed to be playing on a prerecorded loop. And it was almost like the early cinematic attention to, you know, the lapping waves and the rustling leaves in audiences’ first encounter with film.
But in our very post-cinematic environment there’s this major difference. It’s not just the difference of the lack of indexicality—of course it’s not not related—but it’s also related to the way that images are processed now on your screen, each time anew. One of the mechanisms that allows for an efficient processing on your screen is to reduce the amount of variation among unchanging or minimally changing pixels. And so, if there were in fact a looping video that was being played on a greenscreen background, it might look exactly the same as if it were real and your computer were just processing it this way. So we’ve actually come to this point where it’s not just about fake news and things like that, but where visually and perceptually there is a kind of indistinguishability between the faked and the real. And this is where I think it is extremely important to think about these infrastructures in a very critical way if we’re ever going to get out of this mess that we’re in. Because all of these processes are affecting our political engagement with the image right now.
CBA: Absolutely. One of the things that really stood out to me is when you were describing, in chapter three, the speculative nature of these protentional images, the way in which Zoom and other digital video programs are constantly trying to guess what pixel changes may be up ahead, and as I was reading that and thinking about the ideological stress of “liveness” and the speculative experience of “the live” during the coup—I mean, I don’t want to toss this back to you because that’s a really shaggy wet yucky tennis ball to throw at somebody—but I was thinking that there is a way in which the speculative nature, the pre-perceptual speculative nature of our images, becomes reflected in how we understand live events: we are constantly in a speculative “now” that is actually a protentional future.
SD: I think, at the very least, we have to say that our speculation about the future—whether we are regarding this in longer circuits of the future or in terms of the immediate moment of the future that we are open to at every moment—at the least we have to say that those processes are now deeply imbricated in these technological processes that are actively co-speculating about the immediate future. It’s interesting, and I don’t mean to pick on the word that you use, but it makes me think about how this imbrication works. You were talking about how these technological processes get “reflected” in our anticipation or imagination, our speculation about the future—and I think that’s absolutely right, but it’s also important to say that there is a moment that is prior to reflection or an interval that is more minuscule than that which allows for reflection, at which this imbrication is already taking place. And that’s a really scary thing.
As I argue in the book, obviously there are these mechanisms which are designed and engineered by corporations located right outside my apartment here in Silicon Valley that are designed to capture our future selves; they’re designed to predict, and ultimately to control and to shape, who we will be because we’re much easier to market to and, you know, sell insurance to, for example, if our movements and ideas can be predicted, shaped, or influenced. There are layers of tighter and tighter circles of control which go down through streaming processes and buffering and video compression and down to what computer scientists call “speculative execution,” which takes place at the level of the data cache at the CPU and which is about speculating in a special sense, when the computer—and I don’t mean to anthropomorphize it too much—but when the computer doesn’t know what’s going to happen in the next microseconds. It will then execute multiple lines of code, as in a conditional of the form “if this, else this, else that.” It can execute many of those different lines at once and then discard whichever ones turn out to have been wrong. What I’ve tried to emphasize is that, although that is part of the technical control circuit of the future, it is also a point of contingency where control can never be total. And so I’m trying to think about how we can open up these spaces of contingency and maybe reclaim some kind of control of our own. I recognize, though, that that’s a very slim, microtemporal margin of hope, but it could still be crucial for thinking about the ways that these various temporal levels interact with one another.
CBA: I think you do a really great job of that, in the last chapter of your book in particular, of layering temporal scales on top of each other for the reader in order to start thinking about how an ethics might emerge from the awareness that you’re offering us—and I kind of want to leave that idea there as a little Easter egg for readers of your book.
There’s another question, though, that kept occurring to me as I was reading. It probably goes without saying that Discorrelated Images is primarily interested in visual phenomena, and yet, because of your investment in and connections to media archaeology as a methodology and a tool set, I kept thinking that there were a lot of generative connections to work being done in sound studies right now. So I was wondering if there were observations about post-cinematic sound that had been floating in your head, that perhaps hadn’t come out in the book, or connections that you would recommend to readers to pursue.
SD: Well, definitely, Jonathan Sterne’s work on MP3 and everything that he’s done since then, including the more recent work on disability and sound, is operating in a similar space; his work is foundational for a lot of the work that I do here, in that he’s also interested in a kind of archaeology of formats and protocols as sites of politics of technologies—which is something I’m also gesturing towards. But it’s true that I don’t deal a lot with sound directly in the book. There’s a couple of different reasons for this. First of all, computers don’t care, obviously, if they’re processing images or sounds, and so to that extent sounds are going to be subject to all of the same processes that I’m talking about with images. That’s a very basic observation, I think, but it’s ultimately a very important one when read through the kinds of lenses that Jonathan Sterne opens up.
Why, then, do I focus on images? Beyond the boring, pragmatic reasons—which have to do with my background, with the field of film and media studies and its foundational ocularcentrism, I guess you could say, as well as the fact that I work in an Art & Art History department that deals mostly with visual art—I do think that there’s perhaps a more substantive reason for the necessity of a visual (though not exclusively visual) approach, and that has to do with the phenomenological centering of the subject. What I’m trying to say is that ocularcentrism is not a simple mistake. It might be a mistake, but it’s not a mistake that we can easily or simply repeal, because our subjectivity—conceived as individual subjectivity—is so heavily centered around the visual field, in contrast to the way that the aural is much more dispersed and therefore also possibly a much richer site for a kind of oppositional politics. But I think that if that is true, then it needs to be part of a double-pronged approach: we also need to attack whatever it is that dominates our image or our sense of individuality—or individualism, which I take to be a problem, especially in the ecologically entangled world that we live in. We really need to think about the way that the phenomenological correlation itself is wrapped up in ocularcentrism.
That’s a long preface, but what I’m interested in, and what I think your question opens up, is a kind of comparative approach to the way that the aural is not only “naturally” less individualistic than the visual (because different frequencies affect us differently, like the way low-frequency bass tones surround us and don’t have a kind of focal “center” in the same way that a visual object does), but also how there’s something new happening, which is technologically conditioned, and that I think we can start to sense in digital media—for example on Zoom with its microtemporal rewinding and the slight pitch shift that we notice when things have gotten out of joint. Here, we might start to think about how the construction of the individual is open to these contingencies of time and space through sound and image alike.
CBA: Yes, and I think it also really speaks to the importance of this book for sound studies, which is something I wanted to flag. It’s going to be a fantastic tool to think with, and I didn’t want anyone from sound studies to not pick it up simply because it has “images” in the title. So I’m glad we got to talk about this.
On a different note, your chapter on algorithmic animation uses the perpetual refiguring of Frankenstein’s monster as a metaphor or emblem for cinematic media and explores how that’s not precisely replaced but perhaps remoulded or reworked through the fembot in 21st-century post-cinematic features. While reading, I kept thinking about Dirty Computer and Janelle Monáe’s character, Jane 57821. You note that “it should therefore be possible to mobilize such images [of the post-cinematic fembot] for a politics of de- and resubjectivation, for the artificialization and queering of subjectivity.” So I thought of Jane; I thought of Dirty Computer. And I was wondering whether or how we might connect that queer Afrofuturist impulse to the figure of the fembot or other emblems of post-cinematic algorithmic animation.
SD: It’s a really great question. As you know, my focus in that chapter is on more mainstream and in some ways less critical appropriations of the artificial creature—and it’s always a woman—in movies like Her and Ex Machina and Blade Runner 2049. Part of what I’m doing there is trying to get a grip on the dominant ideology of this moment, which extends through technologies like Siri and Alexa, these digital servants or chatbots that are always women or coded female in some way (and in a non-queer way, I think: it would be quite a stretch to claim any kind of queerness at the root of Siri). What Janelle Monáe’s movie really brings out is a queer potential that is nevertheless there, that could be made—but only through an act of subversion—to play a role in these kinds of fantasies; and I emphasize the fantasy part, the gendered fantasy of these other, more normative or non-queer scenarios, where it’s really about the spectacle and the fetishization or objectification of these fembots, chatbots, holographic women, etc. that allows me to think about the dominant normative ideology. Janelle Monáe, on the other hand, is taking a more critical approach that utilizes many of the same techniques in order to expose this ideology.
In comparison, Ex Machina stages what ultimately feels like a rather fake rebellion against patriarchy, but I’m less interested in the story that it tells than in the visual emblems that it uses. What’s interesting about these images is that they refocus visual attention in terms of its dispersal across both an “operational aesthetic,” a kind of fetishization of visual effects itself, as well as the diegetic seamlessness that the movie fantasizes at the same time. This duality, inherent in the visual object of Ex Machina’s fembot, can help us to uncover some of the contradictions that Janelle Monáe is pointing to much more directly, in a much more politically invested way.
The quote that you mentioned—that “it should therefore be possible to mobilize such images for a politics of de- and resubjectivation, for the artificialization and queering of subjectivity”—was actually responding to the idea of xenofeminism (about which I remain somewhat ambivalent, as I am unsure how its reappropriation of Prometheanism, for example, can ultimately be squared with the queer feminism that xenofeminism self-avowedly is). But what I like about the xenofeministic approach is its awareness of discorrelation as a site of denaturalization and reartificialization, as something that could be appropriated even at the microtemporal scale, at the level of computational processes subtending representations—for example, the processes by which memes are transmitted across networks and reconstructed on someone’s computer. And I think that what Janelle Monáe offers is something like a mid-level or meso-scale negotiation between those two—between the dominant ideology’s fetishization of the image of the artificial woman, on the one hand, and the microtemporal and subrepresentational denaturalization of gender that xenofeminism envisions, on the other. Monae’s narrative intervention in Dirty Computer mediates between these levels, providing a very powerful way of making sense of discorrelation while exposing the political dimensions not only of images but also their infrastructures. In that sense, her intervention is very much in line with the archive and the dual phenomenological and media-archaeological interests that inform Discorrelated Images.
Shane Denson is Associate Professor of Film and Media Studies and, by Courtesy, of German Studies at Stanford University. His research interests span a variety of media and historical periods, including phenomenological and media-philosophical approaches to film, digital media, and serialized popular forms. He is the author of Discorrelated Images (Duke University Press, 2020) and Postnaturalism: Frankenstein, Film, and the Anthropotechnical Interface (Transcript-Verlag, 2014) and co-editor of several collections: Transnational Perspectives on Graphic Narratives (Bloomsbury, 2013), Digital Seriality (special issue of Eludamos: Journal for Computer Game Culture, 2014), and Post-Cinema: Theorizing 21st-Century Film (REFRAME Books, 2016). See shanedenson.com for more information.
Caetlin Benson-Allott is Provost’s Distinguished Associate Professor of English and Film and Media Studies at Georgetown University. She is the author of Killer Tapes and Shattered Screens: Video Spectatorship from VHS to File Sharing (University of California Press, 2013), Remote Control (Bloomsbury, 2015), and The Stuff of Spectatorship: Material Cultures of Film and Television (University of California Press, 2021). She is also Editor of JCMS, the scholarly publication of the Society for Cinema and Media Studies. Her work on US film cultures, exhibition history, spectatorship theory, and gender and sexuality studies has appeared in Cinema Journal, South Atlantic Quarterly, Journal of Visual Culture, Jump Cut, Film Quarterly, Film Criticism, Feminist Media Histories, FLOW, and multiple anthologies. She also writes a regular column on politics, platforms, and contemporary media for Film Quarterly.