Digital Autonomy

This essay was originally written as a response to a symposium held at Goldsmiths, Uni of London, in June 2011, entitled: New Notes on Sculpture

“Is an ephemeral image, a moment in a streaming video, a thing? Or if the image is frozen as a still, is it now a thing? Is a dream, a city, a sensation, a derivative, an ideology, a decay, a kiss? I haven’t the least idea.”

Extract from David Miller, Materiality: An Introduction

In A Thing Like You and Me, Hito Steyerl plays out her ongoing obsession with the copy, skirting briefly over her wider, yet more implicit concern: the digital. Echoing the work of Bruno Latour, Steyerl acknowledges the materiality by which images are created, scarred and destroyed in order to get to a much deeper, ontological question about autonomy. Avoiding the kind of subject/object purification Latour warns about, Steyerl asks us to consider images as something we can participate in, even model our autonomy on. Is it possible to become a thing? And where does Hito Steyerl get off calling us ‘things’ in the first place?

For Latour things cannot be understood in a pure form, material or otherwise. To understand a JPEG means also understanding personal computers, digital economies, carefully programmed image compression software and the relations which connect them:

“…as soon as we are on the trail of some quasi-object, it appears to us sometimes as a thing, sometimes as a narrative, sometimes as a social bond, without ever being reduced to a mere being.” [2]

To map and follow these narratives and social bonds it becomes necessary to understand the history of actants and their relations without manipulating those histories into subject/object hierarchies themselves. Hito Steyerl begins with David Bowie’s pop-video Heroes: a highly produced moving-image made to re-present “three simultaneous angles” of the idealised proto-idol. In order to grant a condition of autonomy to the identity that results, Steyerl becomes more explicit:

“…if identification is to go anywhere, it has to be with this material aspect of the image, with the image as thing, not as representation.” [3]

If we are to identify “with the image as thing”, whilst acknowledging the debt Steyerl’s account pays to Bruno Latour’s ‘object-oriented philosophy’, we must do two things. Firstly, we need to trace the materiality implicit in the technology of image reproduction which Hito Steyerl is interested in. Secondly, we need to isolate the kind(s) of autonomy those technologies are traditionally thought to (dis)allow and, in turn, determine whether Steyerl’s ‘thing-oriented’ participation is compatible with the definition of ‘image’ that results.

Throughout, it is my aim to highlight Steyerl’s major concentration: the digital, maintaining that Steyerl is, above all, arguing for a new, singular conception of autonomy that does away with subject/object dichotomies rooted in the ‘material’ products of a market economy.

Seeming to sever their dependency on the physical processes that underlie them, digital technologies:

“incorporate hyper-redundant error-checking routines that serve to sustain an illusion of immateriality by detecting error and correcting it…” [4]

The alleviation of error and noise, is then, an implicit feature of digital materiality. Expressed at the status of the image it is the visual glitch, the coding artifact, [5] that signifies the potential of the image to loosen its digital shackles. Enabled by over-used compression software, images are forced to atrophy as they navigate multiple bandwidth streams. Hiding in the random access memory of a user’s personal computer, the coded image waits patiently for browser software to determine which type of de-compression will best suit its use, before, in microseconds, the image re-emerges as a 60 frames per second flicker on a dirty, regularly fingered, poorly maintained computer monitor. Iman Moradi on ‘the glitch’:

“The visual glitch is an artefact resulting from an error. It is neither the cause, nor the error itself, it is simply the product of an error and more specifically its visual manifestation. It is a significant slip that marks a departure from our expected result.” [6]

Here Steyerl is calling for us to “tap into” the power of the bruise and the glitch, in order to participate in the forces that compel contemporary digital-capitalism. Blurring the boundary between things (their networks and relations) becomes a way of exposing the market forces that bind us. In the 80s and 90s digital networks, including most prominently the World Wide Web, were talked about in utopian whispers (listen closesly and you can still hear them). With the idea of the ‘original’ banished, corporate oligarchies could concentrate on distributing perfect digital simulacra to customer nodes. These processes, according to W. Daniel Hillis, define:

“the essence of digital technology, which restores signal to near perfection at every stage.” [7]

Maintenance of the network and the routines of error management thus became of primary economic concern: control the networks and the immaterial products would manage themselves.

In the case of digital images, error-managed at every stage of their transmission, it is only at the level of the visual that Steyerl’s “bruises” become manifest. Code-savvy artists (and theorists, such as Moradi) do not see a visual glitch as an aesthetic abhorrence, but as a signal of the kinds of errors that image may have been subjected to. Once an error – perhaps resulting in a glitch or a bruise – is allowed to creep into an image file, that error will itself be re-produced accurately by the systems it navigates. In effect, error management will maintain the glitch just as readily as it maintains the intended image. There is no distinguishing between the two at the status of computer code. For Steyerl, autonomy emerges from the participation, not in the error, but in the “bruises of images… its glitches and artifacts, the traces of its rips and transfers”, outer marks signalling “a constellation of forces” petrified within:

“In the commodity fetish, material drives intersect with affect and desire, and [Walter] Benjamin fantasizes about igniting these compressed forces, to awaken “the slumbering collective from the dream-filled sleep of capitalist production” to tap into these forces… Things are never just inert objects, passive items, or lifeless shucks, but consist of tensions, forces, hidden powers, all being constantly exchanged.” [8]

Participating in these glitches then gives us back a certain control over the market networks that images navigate. Immaterial, perfect simulacra become – in what some may see as a modernist reversal – unique material things, scarred by their passage and better for it:

“In a sense we are cherishing the little idiosyncrasies that are absent from the soulless machines churned from the production lines.” [9]

Glitch artists identify with the aesthetic of malfunction, compression artifacts and creative transcoding, tending towards techniques that manifest these qualities in their work. Although the glitch is universally acknowledged as a way to expose digital and technical materiality, much debate surrounds whether it is accurate to say one can intend a glitch. Glitches expose errors, and errors are in opposition to the norm. [10] To manipulate the code of an image, and manifest a particular aesthetic result, is not to produce a glitch, but to produce the representation of a glitch. This problem stabs at the very heart of Hito Steyerl’s appraisal of ‘things’ as, very early on in her text, she turns her back on representation:

“It doesn’t represent reality. It is a fragment of the real world. It is a thing just like any other—a thing like you and me.” [11]

Materially the glitch in the image does not represent error, neither then can error be intentionally injected into images in order to produce a bruise or artifact. But here, perhaps, we may have a way to prise open Steyerl’s rhetoric, in order to uncover the slip (glitch?) she makes between ‘participation’ and ‘autonomy’.

In another of Steyerl’s e-flux essays, In Defense of the Poor Image, we find ourselves allied further with, what she calls, these “Wretched [debris] of the Screen”:

“The poor image has been uploaded, downloaded, shared, reformatted, and reedited. It transforms quality into accessibility, exhibition value into cult value, films into clips, contemplation into distraction… The poor image is an illicit fifth-generation bastard of an original image. Its genealogy is dubious. Its filenames are deliberately misspelled… It mocks the promises of digital technology. Not only is it often degraded to the point of being just a hurried blur, one even doubts whether it could be called an image at all. Only digital technology could produce such a dilapidated image in the first place.” [12] [my emphasis]

So poor images have become, for Steyerl at least, not only a way to expose and betray network economies, but the improbable re-definition of the digital itself. Here perhaps, it is worth mentioning Paul Virilio’s work on ‘failure’. Virilio opposes the idea of a ‘norm’ to which an accident would be an unintended divergence. Instead, Virilio sees accidents and failures as productive – human – acts, they are an integral, crucial, part of the system of relations:

“Failure is not a condemnation! … Failure is an accident: art has tripped on the rug… In my view, the accident is positive. Why? Because it reveals something important that we would not otherwise be able to perceive. In this respect, it is a profane miracle.” [13]

The glitch then, exposes its own causes (i.e. the digital ‘failure’) revealing features of the network “we would not otherwise be able to perceive”.

Poor images, crammed full of artifacts, glitches and bruises maintain their autonomy in two ways. Firstly, they “mock the promise of digital technology” by exposing the errors and limitations of the networks they have been parsed through. In turn, these networks – designed to maintain market equilibriums – actually act in favour of glitches, managing the ‘errors’ that produced them as if they were part of the ‘intended’ digital quarry. Secondly, and most crucially in our rhetorical distinction, images avoid having this autonomy controlled by web users and artists because an ‘intended’ error is not an error at all and does not produce a ‘real’ glitch. Images with glitches then suddenly become an interesting thing to aspire to because (and here I will loosely echo Steyerl’s concerns) unlike subjects, they cannot be subjected, and unlike objects, they are impossible to objectify. This brings us back to one of our questions: how to become a thing like an image? With our awareness of digital images refreshed, it becomes possible to read into Steyerl’s text an ontological proposition:

“To participate in the image as thing means to participate in its potential agency—an agency that is not necessarily beneficial, as it can be used for every imaginable purpose. It is vigorous and sometimes even viral. And it will never be full and glorious, as images are bruised and damaged, just as everything else within history. History, as Benjamin told us, is a pile of rubble. Only we are not staring at it any longer from the point of view of Benjamin’s shell-shocked angel. We are not the angel. We are the rubble. We are this pile of scrap.” [14]

With true participation in the agency of the image now exposed as improbable, the question of what this “potential agency” is shouts back at us. To finish I want to re-pose Steyerl’s proposition, asking more explicitly for us to identify with the glitch; with the “pile of scrap”. The image as thing maintains its autonomy through the glitches it harbours. These glitches, in turn, are non-physical, non-subjective elucidations of the digital. Michel Foucault’s interest in the work of Georges Canguillhem, the grand expositor of biological disease and disorder, bears a striking similarity to Hito Steyerl’s interest in digital images. Foucault sees in Canguillhem’s work on disease a self-sustaining paradigm of scientific principles, where error “is eliminated not by the blunt force of a truth that would gradually emerge from the shadows but [at all stages] by the formation of a new way of ‘truth-telling’”.[15] This epistemology of science as productive failure is further aligned with the status of the biological systems Canguillhem studied:

“It has not been possible to constitute a science of the living without taking into account, as something essential to its object, the possibility of disease, death, monstrosity, anomaly, and error. Although one may come to know, with increasing exactness, the physiochemical mechanisms that cause them, they have their place nonetheless in a specificity that the life sciences must take into account, lest they obliterate the very thing that forms their object and their particular domain.” [16]

In other words, disease exposes the errors that are the constituting factors of life. It is the only thing about life we can be sure of – in effect, Foucault says “life is [productive] error” – and in this regard also stands as a meta-paradigm for the scientific method. The turn towards error[17] and glitch as the defining principle of the network effectively removes ‘intention’ from the system of communication (the digital market?). What Virilio calls “the accident inherent”[18]in the network. The computer virus, the technical glitch, router noise et al. all then become actants exposing the topology of networking. “Through a malfunction, a glitch, we get a fleeting glimpse of an alien intelligence at work.”[19] What’s more – and this is where my alignment with Steyerl’s proposition again rears its head – because we are living systems, error is the fundamental principle exposed by human ‘being’:

“At the center of these problems one finds that of error. For, at the most basic level of life, the processes of coding and decoding give way to a chance occurence that, before becoming a disease, a deficiency, or a monstrosity, is something like a disturbance in the informative system, something like a ‘mistake’. In this sense, life – and this is its radical feature – is that which is capable of error… Further, it must be questioned in regard to that singular but hereditary error which explains that fact that, with man, life has led to a living being that is never completely in the right place, that is destined to ‘err’ and to be ‘wrong’.” [20]

The autonomy we participate in has, it seems, been uncovered as a principle feature of our constitution all along. Dealing with images, with their glitches, with the networks of failure they expose, is simultaneous to dealing with our own natures as beings composed of matter. These principles are not representations of our autonomy. In a very fundamental sense, we expose the possibility of our autonomy. For it to be any other way our material natures would have to be ontologically different. Participating in the glitch, in the artifact that exposes the error, is to align oneself with material reality. Perhaps this is my own, expanded reading of Steyerl, but I want to suggest that doing away with subject/object distinctions might just be the only way we have to claim a true autonomy in the world. A world, if Latour and Steyerl are to be taken seriously, made up of actants related through networks of failure, accident and error.

To identify with the “image as thing” means doing away with the notion of autonomy bound to the free-floating subject. We material-things express our “own conditions of existence” :

“How about acknowledging that this image is not some ideological misconception, but a thing simultaneously couched in affect and availability, a fetish made of crystals and electricity, animated by our wishes and fears—a perfect embodiment of its own conditions of existence? As such, the image is—to use yet another phrase of Walter Benjamin’s—without expression.” [21]



Footnotes / Bibliography

[1] David Miller, Materiality, 2nd ed. (Durham [u.a.]: Duke Univ. Press, 2006), 7.

[2] Bruno Latour, We have never been modern, 7th ed. (Cambridge Mass. Harvard University Press, 2002), 89.

[3] Hito Steyerl, “Hito Steyerl, A Thing Like You and Me / e-flux”, April 2010,

[4] Kirschenbaum, M. G. Mechanisms : New media and the forensic imagination. (Cambridge, Mass.: MIT Press, 2008), 12.

[5] Although Steyerl (and others quoted here) uses the English spelling of ‘artefact’, I have very consciously chosen to spell mine with an ‘i’, widely known as the American spelling of the term. This spelling of the word, though, aligns it with computer/programming terminology (i.e. “compression artifact”), leaving the ‘e’ spelling free to echo its archaeological heritage. I think the distinction is worth making, and consider that perhaps Hito Steyerl chose the English spelling to purposely implicate the glitch as an “object formed by humans”. In any case, multiple meanings for the word can be read in each instance. See for an overview of the word’s various uses and spellings.

[6] Iman Moradi, Glitch : designing imperfection, 1st ed. (New York: Mark Batty Publisher, 2009), 6-7.

[7] W Hillis, The pattern on the stone : the simple ideas that make computers work, 1st ed. (New York: Basic Books, 1999), 18.

[8] Steyerl, “Hito Steyerl, A Thing Like You and Me / e-flux.”

[9] Iman Moradi, Glitch, 10.

[10] The distinction between error, norm, intention – some may even add the concept of mutation – is ahuge subject, and requires more attention than I can give it here. Perhaps another time.

[11] Iman Moradi, Glitch, 11.

[12] Hito Steyerl, “Hito Steyerl, In Defense of the Poor Image / Journal / e-flux,”

[13] Paul Virilio, “The Art of the Accident,” New York, Semiotext, 2005, 63.

[14] Hito Steyerl, “Hito Steyerl, A Thing Like You and Me / e-flux.”

[15] Michel Foucault, “Life: Experience and Science,” in Aesthetics, method, and epistemology (The New Press, 1999), 471.

[16] Ibid., 474.

[17] Human error/machine error.

[18] Virilio, P., 1993. The Primal Accident. In The Politics of everyday fear. Minneapolis: University of Minnesota.

[19] Vanhanen, J., 2001. Loving the Ghost in the Machine: Aesthetics of Interruption. Available at:

[20] Ibid., 476.

[21] Hito Steyerl, “Hito Steyerl, A Thing Like You and Me / e-flux.”