Preface

[ On using this book … ]

This book is about seeing, so if you are reading this text before looking at the pictures, stop right now. Construct a narrative, an argument, from the images alone.

Do read the text—just do it later. See if reading changes your perception of the images. The opposite experiment might be equally valuable: read first and look only later, discovering the degree to which your understanding is changed by looking.

There is another experiment, which I must fail to describe in full. It is the one in which you use this book as you will.

•••

[ On the writing, if you can call it that … ]

Everything written here is provisional.[i] Even though I wrote it, I don’t stand behind any of it. I don’t defend it as truth. I can’t promise it will be helpful. It was merely the best I could do under the circumstances—possibly not even the best. We are always, after all, subject to our circumstances and hostage to our situation. As soon as I composed something, there was something other to write, something better, more to the point, more accurate, more amusing, more accessible, more scholarly, more artful, more concise, more delightfully baroque, more poetic. In many, if not most cases, I have changed the words or deleted them already. It didn’t help. Well, perhaps it did a little. Still, each letter, each word, each sentence, each paragraph, every page: they only hold the place for something else that might replace it, that ought to replace it, and that would have replaced it if only there were infinite time and inspiration.

This is what is called writing now, but really it is some other kind of word process that is referred to by this name. We call it writing though there is no pen. We call it typing though there is no type.  This is the writing of change. Words are ephemera. Some wise old characters suggested that thinking happens in them, but really it seems that thoughts just visit words—passing through them like a tourist on the way to a more desirable location.

That is the situation. So you, dear reader, please do me the kind favor of substituting for my words, especially the particularly obtuse or opaque ones, something you actually understand. I won’t be offended in the least. I do the same when I read. As a matter of fact, I have come to recognize that that is pretty much what is meant by reading anyway.

Thinking

 [ On Technophiles and Luddites and everything in-between … ]

There are Technophiles and there are Luddites and there is everything in-between. Our unfortunate tendency seems to be to relate to technology in the affective dimension. Love it. Hate it. Beyond that, there are those—individuals or classes, governments or social bodies, organizations commercial and scientific—with particular interests at stake in the affective valuation of technologies, who translate these feelings into desires which are ultimately actualized, if not exactly satisfied, in the global marketplace. The ultimate ramifications of our technological sentiments register in life through the polarization of wealth, the conditions of labor, massive changes to both local and global ecosystems, and in further and differential effects on the living conditions of everyone and everything that shares our planet.

So while we spend much of our time immersed in a technologically mediated relation to life—a totality which includes communication with others, movement of ourselves and our things, production of the things we need and those we want and probably don’t need—we remain conscious predominantly of desire (or lack of desire) for tech. Secondarily we consider what it can get for, or do for us. The realization that it can be useful is an ascent, if barely, into a new realm: a consideration of utility, instrumentality.

A perspective on the world as seen primarily through instrumentality, rather than desire, is characteristic of the subjectivity which identifies itself with capital: industry, applied science, government, and military. Affect shapes perspectives of a laboring general public, which, nonetheless, tends to share an ideological justification of techno-desire in terms taken from utilitarian apologists for techno-instrumentality. In the current situation, where symptoms of the techno-ravaging of the earth (the seemingly inevitable collateral consequences of utility) are no longer easily deniable, there is a complicity between capital and the populous in suppressing an awareness of looming disaster. Anxiety is allayed, ironically, through the distractions afforded by the tantalization of techno-desires.

It is a challenge to think about technology in a way that diverges from the thinking most broadly circulated, which is aligned habitually with emotion, commerce, and utility in a self-reinforcing feedback loop. It is not adequate to simply reject technology, though this is the tactic that arises in Luddism or Romanticism’s enmity to industrialization. Neither machine-breaking nor a counter-romance with a nostalgically idealized past prevented the progress of the machine’s colonization of our world—though perhaps, they altered its course in some way.

Today, we must learn to change, to diverge from current trajectories towards an unknown and menacing future—to differ from ourselves, to reconfigure and re-conceptualize the horizons of possibility. Because aesthetic contemplation tends to abandon instrumental imperatives as the grounds of thinking, it can be a starting place for an inquiry into technology that affords a space for a different sort of understanding. Since in our present situation, we exist almost always in relation to the machine, we must seek to understand the relation between change and the machine.

•••

•••

 [ On what is called a machine … ]

Under the umbrella term of technology, there is a continuity among the things referred to by the words tool, machine, and computer. This series suggests a progress in time, and a quantum staging of technological knowledge. A tool is an improved extension of the human body controlled and driven in connection with that body. It is a singular abstraction from a particular human faculty. A machine is a tool that is powered separately from the human body controlling it. It is a compound apparatus of multiple parts that may combine the abstraction of several human faculties. Finally, a computer is a machine that is controlled in some way separately from the human mind. It is both multiple and indeterminate: an abstraction not of a particular human capacity but of human possibility—any human faculty. Tool, machine, and computer all share a relation to utility: they are means to an end. They also mediate the human relation to the world by interposing their own objectified, hypertrophic version of human facility between people and their objects in the world.

Each subsequent apparatus partakes of the suchness of its predecessor in the series: toolness (utility) is characteristic of them all. Subsequent developments re-characterize their antecedents and even rework them in their image so that finally all machines and tools become or will become computerized—they are already computerized in potentia. Now it is possible to refer to any of these things with any of these terms; the choice is one of emphasis—each term stresses what is particularly characteristic of the shift for which it is the emblem. But the last item in the series is the one that is privileged. It is now the era of computation, so every mention of a machine likely refers to a computer. The persistence of the appellation “machine” for the computer emphasizes its machinic aspect, which is power and movement. It may be that we require a reminder that the automation of control in the computing machine is not a sign of its effeteness. The word machine stands for the muscularity of the system.

The series tool-machine-computer is not intended to suggest that the computer necessarily stands at the pinnacle of a technological progress. By rejecting a teleological notion of progress, the advent of the computer can be seen to represent a technological change addressing the mathematical demands of modern warfare or business practices. Technological change, however, is not the only sort of change with which the computer has a relationship. The computer is a machine that is an agent of change—not that it changes history by its presence, but that it contains change in its mechanisms, in its material substrates, and in its ontological being. Tools are generically agents of change since their use, their utility, is in effecting things in the world; and use changes the tool through wear. But unlike other tools, the computer works by an internal and constitutive changing. Unlike other tools, it works by being anything (in turn) instead of being one thing (always).

•••

[ On aesthetic contemplation … ]

Aesthetic contemplation is the practice of looking and seeing with all its attendant cognitive processes. This is not the rarified practice of experts. While there may be something that could be called “the trained eye,” it doesn’t do anything substantively different from an untrained one. Insofar as all people have been visually programmed through their relations with culture, we all have trained eyes. Patterns of attention and assignments of particular meanings can vary depending on previous visual experience, but what people do in general when seeing is to scan a visual field, isolate objects, compare them with each other and construct meaningful interpretations from these observations.[ii] To the extent that seeing invokes emotive responses and judgments, it might be said to be aesthetic.

Aesthetic contemplation could be thought of as occurring prior to the harnessing of the visual field to mechanisms of mundane communication, or the demands of material circulation in the marketplace, or even the demands of survival in the wild or the workplace. Except in aesthetic contemplation, vision, as with other senses, is subordinated to the seeking out of the affordances of the environment—the senses are instrumentalized.

Before instrumentalization, seeing is available for a range of affective responses: from anxiety to awe, fear to fascination. The pleasure of seeing is possible because of a visceral responsiveness to vision which is separable from what seeing affords as a means to gratification. Before instrumentalization, seeing is available to the mind for thinking—for responding to what is seen by circulating the concepts which the seeing provokes internally, and sharing them with others in language.[iii] In that circulation of ideas that respond to things seen, new concepts emerge, develop, combine, and alter. Thinking is receptive, responsive, and protean. Thinking thinks about its object and it thinks about itself. Thinking thinks about what is sensed and it thinks about sensing: it thinks about seeing; it thinks about thinking.

•••

[ On the relation between thinking, seeing and computing … ]

If thinking responds to seeing, why then state that “computers change the way we see by altering the way we think?” The titular assertion could well be read in reverse, i.e.: seeing changes thinking—which would be the expected direction of the visually centered system of causality underpinning scientific epistemology. Even in science a loopiness prevails; things run round in circles, or they run in figures even more elaborate. The thinking of hypothesis-making creates the context for observations that confirm or refute thought and lead to refinements—digressions too. Thinking moves. Seeing moves. Each, as object, in response to the other; each, as subject, transitively impelling the other towards difference.

Seeing, in that circuit, is either a method through which the phenomenal world is perceived, or, alternatively, a thing in and for itself: seeing as such. Seeing, in this latter sense, is other to thinking—it is alien enough to become thought’s object. The transparency of seeing reveals itself in opaqueness, and seeing comes to have a particularity, which might be contained by description.

It is not necessary to argue that one or the other, seeing or thinking, has any kind of priority—and neither is it necessary to conceive of them within the complementarity of binary opposition. We might, for instance, want instead to imagine seeing contrasted with language as the possible media for thought itself. I think in words; I think in pictures. Either thing can serve as the constituent elements of thinking: a circulation of concepts no matter what way they are formed, remembered, or transcribed.

So what is meant in the title parallels the linguistic relativism (the weak variant, not linguistic determinism) of the Sapir-Whorf hypothesis: that culturally and historically bound differences in language can be correlated with cognitive and perceptual differences.[iv] To consider something like perceptual relativism means rearranging the terms and understanding perception in its denaturalized aspect as a cultural technique. Seeing is an acculturated practice, learned, transmitted, and disciplined through the artifacts of visual culture and the attendant practices of reception. These artifacts and practices are distinct, diverse, and dynamic, as culture is generally.

To aver that computers change the way we see by changing the way we think is first to assert that seeing is changeable and that seeing is part of a cultural practice that exists within history. Second, it is to assert that seeing is in relation to thinking; and lastly, it is to assert that thinking is in relation to computation.[v] The word “by” should be understood as suggesting proximity, or possibly method, as much as causality.

Since “computers change …” we must, finally, take up the question of change once again: this time change in the sense of temporal difference. Within the last century we have seen a shift from a world dominated by industrial culture and its mass-produced materiality to one dominated by what we call information technology and its culture of the screen. It matters less that this is new or recent than the fact that it is different, that whole ways of being are transformed as everything becomes submitted to the logic of a data ontology. No singular aspect of this regime is unprecedented, but in its totality, it represents a significant intensification and condensation of life around specific concepts that form the basis of computational culture.

•••

[ On automation and difference … ]

What follows are three linked meditations on visual difference, but they are presented under the rubric of automation. It would not be unexpected to consider automation as inherent in computation, but the significance of difference is not necessarily obvious.

Computers are the tools by which machines take control. Automation, from the Greek automatos, means acting of itself—the attribution of agency to the inanimate. Specifically, the 17^th-century coinage relates to the mechanical automaton which acts in imitation of a person (or animal). Our prerogative to will and control action is taken over by the machine and things proceed “automatically,” without the need of our intervention, though still usually requiring our initiative.

It is crucial to understand how it came to be that machines can automate what had been formerly accomplished only by people. As a fantasy, the idea of animating the inanimate is almost ubiquitous, becoming something of an obsession in the 18^th century. While sophisticated movement was achieved by analogy with clockworks, early claims of machinic thought were a fraud. Wolfgang von Kempelen’s Mechanical Turk of 1770, for example, played chess, but depended on the complicity of a hidden diminutive person to produce the appearance of mechanical autonomy. The calculating engines of Charles Babbage (1791-1871) did make some headway in the area of automated computation, though none of his machines where completed in his lifetime.

The specific version of automaticity that lies at the core of the computer’s conception comes from Alan Turing’s 1936 paper, “On Computable Numbers, With An Application to the Entscheidungsproblem,” in which he describes how calculations performed by people could be accomplished by a machine:

We may compare a man in the process of computing a real number to a machine which is only capable of a finite number of conditions […] The machine is supplied with a ‘tape’ […] divided into sections (called ‘squares’) each capable of bearing a ‘symbol.’ At any moment there is just one square […] ‘in the machine’ […] only one of which the machine is, so to speak, ‘directly aware’ […]

The work of the machine is accomplished through carrying out one of three possible operations. It can erase or print a symbol on the scanned square, move one square to the left or right, and change its state, which is likened by Turing to the state of mind in a human. The rules that govern what operation the machine carries out are given in a table of behaviors, indexed by the state and the symbol read. In each moment of discrete time, the machine acts according to the operation described for the encounter of the state it is in with the symbol that it reads in the current square (e.g, if being in state 3 it reads a 1, it should write a 2, move left, and change to state 4).

The Turing Machine is proposed at a level of abstraction beyond that which attains in regular procedures of human calculus; it is an extreme simplification that allows for its actual electro-mechanical implementation. The logic, though, must have proceeded from a more basic understanding of calculation as requiring paper, pen, and procedure. For a machine to compute, it only needed to be able read, write, and follow instructions. The computing machine is a thing that can write according to what it reads. What it is able to do follows from this facility.

Turing goes on to demonstrate that the machine he describes is capable of making any calculation that can be made. If this is not the computer, it is the equivalent of any computer, and it is essentially a substitutional system: its workings depend on a series of changes to a single variable, the square of which the machine is aware. The machine changes the contents of this square based on a system of rules. And this functional difference—the sequential change of the contents of a space—at the heart of computation is also at the heart of computational culture.

The changeability characteristic of digital media stems from the system of differencing that enables computation. Substitution, variation, translation, versioning, seriality, temporal plasticity, customization and personalization, architectural liquidity, fractal and chaotic effloresces all emanate from the ramification of parametric difference in compound systems of repetition and difference within the machine.

Seeing

The Automation of Variation

as formal difference within a computational system

Examples of variation surround us. The efflorescence of life forms and the proliferation of cultural forms are the background of daily living. Our sensory access to this variety inspires our sense of autonomy in choice, our sense of creative possibility in imitation, and our sense of understanding in science through theories like natural selection and evolution.

The play of theme and variation has a rich history in art practice. Repetition itself seems to invite changes in form. The difference in the circumstances of execution, the difference between the hand and the habit of one artist and another, mistakes and accidents, observed difference, and the apprehension of possibility in improvisation, all conspire to move cultural producers to vary their output. The computer might have brought an end to variation by enabling the exact perfect repetition of even the most complicated procedures, but it didn’t. It brought instead an intensification of variation.

In the computer, aesthetic systems are abstracted into protocols, form-making techniques into procedures, skills into rules, and styles into parameters, variable factors which configure the system. At every point, the constituent elements make space for parametric substitutions. Every constant becomes a variable and the permutations of difference explode in number.[vi] The spaces of variability become more readily apparent when they are transcribed into rules and the arbitrariness of any particular variant is obvious.

The virtuality of the computational space facilitates change. The inhibition to explore diversions and digressions, which comes from the perception of permanence and expense associated with physical media, disappears. In the computer, all creation amounts to writing in air. Time becomes the most significant obstacle to variability. Visual practices can assume the fluidity of performance, of music, and of writing.

When a program runs it produces a variation from the universe of possibilities encoded by the program. The machine can be set running to instantiate any number of those possibilities, either systematically or at random. In a sufficiently complex system, the set of possible instantiations will be difficult to exhaust. The user of the program explores the style and reach of the system to apprehend and characterize its particular gesture. That use, and that understanding, will suggest new possibilities, new parameters and variations, new spaces to explore. The aspect of variability in software comes because in it an aesthetic system is transcribed as a model of substitutional possibility: a program for the machine. The potential vastness of software’s parameter space is countered by the mechanisms of execution that the computer affords. In complex systems though, the instantiations will remain incomplete, a small subset of the possible.

The Automation of the Sublime

as the promise of infinite instantiations of difference within a computational system

The aesthetic implications of seriality as practiced by modernists—from the impressionists through the minimalists and beyond—are multiplied in contemporary works by the scale and rapidity with which computational methods can produce variation. The intensification and multiplication of variation changes the essential nature of image production.[vii] This shift of degree and magnitude invokes the sublime.

The sublime is a quality that inspires a peculiar mixture of fear and pleasure: awe in the face of the vast, the powerful, the excellent, the beautiful, and the overwhelming. It impresses. It makes one feel small and insignificant.

Kant’s discussion of the mathematical sublime in the Critique of Judgment, is particularly useful in understanding the impact of computational work. The mathematical sublime relates to our experience of the innumerable and the extremely large or small. The dynamic sublime, by contrast, relates to the impossibility of directly experiencing the dangerous forces of nature (without risking bodily harm). The sublime is at once the humbling of human faculties as they confront what exceeds them, and the overcoming of that which overwhelms the human scale by a trick of the mind. We contain the sublime with a conceptual encapsulation, a naming, or a representation. Reason can grasp what cannot be directly experienced. An artist can show it. We experience pleasure in this overcoming: a congratulatory laurel.[viii]

Modern notions of sublimity shift their focus from the natural world to the human one. The power of the machine holds a fascination for the witnesses to industrialization as a kind of shock of the new. At the height of industrialized modernity the machine takes center stage as a metaphor for the body itself. The fascist aesthetic idealizes an armored, hardened, and powerful model of the statuesque human being. The social body is also likened to the machine and choreographed as a mechanized spectacle.[ix]

The soft world of the computational sublime shares aspects of Kant’s mathematical sublime. Computers operationalize the contradiction between the symbolic specificity of a parametric system and the infinitude of its possible variations—they concretely describe within a finite language a system which can be uniquely instantiated infinitely. While the profundity of remarking on this paradox may seem to decline as the novelty of the thinking machines wears off, the machines will continue to reproduce the trick we find such pleasure in—containing the infinite in the finite—and we will likely continue to be amused. The trick is arguably what gives digital art its impact and its effects.

In addition to a quantitative sublimity, computation also yields a sublimity of complexity, which is achieved by way of the iterative power of computational logic. Computers produce a fractal baroque of infinitely ramified involutions. Software recursively enfolds substitutional systems within themselves. The self-similar and simultaneously divergent inventions of form are amazing. They tickle us with the same kind of paradox: infinite systematic variation. The infinite is again contained in the finite. The feelings provoked range from awe and wonder, to dread and terror, to stupefaction.[x]

The Automation of the Image

as the informatic instantiation of difference within the image in a computational system

Major changes in the human relationship to image can be broadly periodized in terms of image-making techniques: painting, photography, (digital) image. Throughout images remained a representation of things, whether in the world or phantasmatic, whose existence was elsewhere from the image.

Under the regime of computation, the very notion of image becomes generalized so that the word “image” stands at once for the specifically digital image and also for the generic image which is inclusive of painting and chemical photography. These latter techniques have anyway been subsumed into image—just as photography subsumed painting—because encounters with all images come to be mediated by the electronic, the computational—the screen translation of whatever the original form of the image was.

Photography, the technical image,[xi] brought with it an intensification of the opticality of image making. Photography depends on the science of light and the technology of chemical sensitivity in order to create an image that indexes the real. Every area of the chemical photograph has an analogical relation to a place in the world from which light was reflected onto the sensitive surface where it left a physical trace. The ideological imprint of that trace is the notion of photographic realism, in which photography is the guarantor of truth.

The reproducibility of the photographic image brings with it the mass distribution of images, which circulate in larger and broader publics. Images are freed from their fixed locations and the cumbersome supports on which they are made: cave, building, frame, canvas. As a consequence, the aura of authenticity that surrounded the original image dissipates and the sense of sacrality in the image disappears.[xii]

The digital image intensifies the possibilities of reproduction and distribution because its digital format allows additional means of electronic reproduction and transmission. The digital also removes the idea of image as truth; instead we confront the dubative image in which the awareness of manipulability destroys the guarantee.[xiii] The digital image is also a move away from opticality, as image capture in the camera becomes only one of a multitude of possible paths to imaging. The real makes way for the virtual: the synthetic image, the simulated image, the abstract image, and the data image. Images can be constructed from numbers, from theories, from systematicity itself. The idea of image as construction replaces that of image as optical capture even when some notion of capture remains—i.e., when there actually is a camera.

In the realm of the digital, the ontology of the image has shifted as well: an image exists as a set of discrete quantifications. The pixel, the constituent atom of the digital image, is a numeric variable—just another empty square subject to the computational mechanisms of substitution. Where chemical photography sampled explicitly within time and place, digital photography samples also within resolution: the number of pixels to assign to a place. The parameters of indexicality become variable.

The image is a third degree truncated fractal of variability. The pixel is degree zero where systems of mutability morph the image one atom at a time. The image in its entirety is also subject to variation—the image itself is seen to be filling a square and subject to paradigmatic substitutions by other images and even variants of itself. The image is encapsulated by metadata, information about itself: names, classifications, dates, titles, technical data, authorship, copyrights, etc.—all of these held in their own variables and subject to manipulation. Finally, the image is stored within a database or a file system, in a particular place, at a particular index, and again subject not just to retrieval, but to substitution.

Preface

The Automation of Variation

The Automation of the Sublime

The Automation of the Image