Signal. image. architecture front cover

This book is not for the faint of heart. This is especially true for readers born sometime during the twentieth century, who learned to read and write, to take photographs, and who enjoy film or love turning the colored pages of hand-made atlases. In the twenty-first century, according to John May, every one of those tasks has become, if not exactly fake, then at least a deep source of inauthenticity. Why? Because today it is by invoking and plugging into a shower of data coming from the Cloud (I did not say “coming from Heaven,” but for the ignorant user it has about the same mystifying effect) that you get, by sheer condescension, some traces, on screen or on paper, that resemble—but only resemble—what used to be texts, photographs, films, drawings, and maps. Poor fools, you believe you take “digital photographs,” but this is not the case at all. What the Cloud has registered is a data file that is dumbed down and simplified into some shape on the screen for minds stuck in the older “orthographic” era.

In a prodigious inversion of Plato’s Allegory of the Cave, May brings us all back into the darkness, where we are occupied looking at screens, except now there is no way to go from the appearances to their hidden reality. The reality is that of data sets invisible to the eye, inaccessible to the human brain, sent telematically everywhere, data sets that project fleeting versions of their contents, sometimes as spreadsheets, sometime as texts, sometimes as drawings. The message from the Cloud is clear: “Thou shall get only images.” Images are no longer a record of anything; they are the provisional translation and a possible rendering of data that could take any other shape.

The feeling of inauthenticity comes from a disconnect. When engaged in the older practice you capture some event through a medium—writing, drawing, taking pictures, filming, and so on; this creates an archive of some sort; then later you interpret it as best as you can. Such a sequence gave you a concrete grasp on the passage of time: the past was revealed by the archive and the future by the interpretations of its traces. Historicity, objectivity, and hermeneutic work were produced together in this era of technical literacy called orthography. People of the twentieth century were still “in history.” But now nothing of such an order is maintained if the Cloud delivers only infinitely revisable versions of its data. One version of an image cohabits with billions of others just as equally possible—and each of them could be easily adulterated, thereby adding yet another version. Historical time, the possibility of objectivity, and the hard work of interpretations are gone. The Cloud is timeless.

When, in my own writing, I’ve pointed out that a lot of scientific referential work could be summarized as a search for the oxymoron of “immutable and combinable mobiles,” I was try-ing to describe activities that encountered a lot of trouble in succeeding at this contradictory task. I was thinking of the immense work necessary to bring birds from the Galapagos to the Science Museum in London, as if brought across long distances in time and space while remaining “intact.” I was alluding to the hard work of La Condamine to bring back to Paris the data sets on the measure of longitude he accumulated over many years in Quito.

But what May’s book makes me realize is that the contradiction between immutability, combinability, and mobility has been solved to such an extraordinary degree that time and space have shifted location, moved above our head, and are now falling back on us—hence, the incredible metaphor of “the Cloud.” Scientists have prayed for centuries, “Give us immediacy!” Well, now they have it for sure (provided we forget the electricity necessary to sustain the Cloud up in the air). “Real time” has displaced historicity. Our present era unfolds as if Leibniz’s calculating God had finally taken over and dissolved time as well as the difference between good and bad, now made simply statistical. It is rather unfortunate that just at the time when we seem to have lost our ground because of the climatic mutation, we are also collectively unsettled by the complete disconnect between older technics of inscriptions and the digital infrastructure that is now activating them from behind. Just at the time when we need to land on an earth that would give us some solidity, we also have to reconcile ourselves with a technical infrastructure for which we don’t have the right bodily apparatus.

No wonder that people born in the former century feel unsettled, and that a general atmosphere of “fakery” has spread everywhere: we are no longer physically adjusted to our imaging technology. As André Leroi-Gourhan would say: to confuse humans born in the orthographic and in the digital eras would be like confusing animals as different as lions and tigers. 


A man and a woman sat before an electric eye in a London laboratory tonight, and a group of persons in a darkened cellar in this village outside New York watched them turn their heads and move from side to side. 

The images were crude, imperfect, broken, but they were images nonetheless. Man’s vision had panned the ocean; transatlantic television was a demonstrated reality, and one more great dream of science was on the way to realization. 

The demonstration was made by the Baird Television Development Company of London, using short-wave radio sets for transmission of the “vision sound,” and the “televisor” invented by John L. Baird… for turning this sound back into vision after its ocean hop. 

The transformed vision of the man and woman in the London laboratory came through the ether in the form of a bumblebee’s hum, a musical buzz, or irregular cadence representing in sound the lights and shadows of their faces—all that was transmitted in the test. 

When the televisor, a black box compact enough to be carried around in a taxi, had done its work with this rhythmic rumble from across the sea the visions gradually built themselves up of tiny oblongs of light suspended in a whirling rectangle of brilliance in the machine’s gaping mouth. 

The vision of the woman appeared broken and scattered, but it was still plain that she was a woman and that she was showing first the full face and then the profile. The vision sound was sent across the ocean by short wave radio station 2KZ, of only two kilowatts power. 

New York Times, reporting from Hartsdale, New York, on February 9, 1928 

John Logie Baird, the first 30-line electrical image, transmitted from London to Hartsdale, NY on February 9, 1928.
John Logie Baird, the first 30-line electrical image, transmitted from London to Hartsdale, NY on February 9, 1928.

We can feel our images changing us. Our relationship to our thoughts, to our sense of time, to the cadence of our attentiveness—all of this is subject to rapid revision now, by processes that seem only to be gathering. We are more akin to John Baird’s electric eye than to the man and woman who, one hundred years ago, sat before it.

A blissful but anxious surrender is taking place, incomplete but very much underway. If it once felt possible, even natural, to retain a kind of mental equilibrium between words and images, or even to subordinate images to text, we now sense a quickening imbalance. With this imbalance comes the somber realization that refusing to surrender to images may only blind us to the hidden contours of a new lived reality.

To patiently describe the world to oneself is to prepare the ground for a politics that does not yet exist. Under the right circumstances, new descriptions can be made to serve as the raw material for political impulses that cannot yet be expressed or lived because their preconditions have not been arranged or articulated.

John Logie Baird with his “televisor,” circa 1926. © Hulton-Deutsch Collection/ Corbis Historical via Getty Images.
John Logie Baird with his “televisor,” circa 1926. © Hulton-Deutsch Collection/Corbis Historical via Getty Images.

For ten years I have been trying to describe the ways in which the design fields misunderstand their own images.1 Wandering in a self-imposed haze of anachronism, we habitually confuse electronic images with other forms of visuality, casually referring to them as drawings, or photographs, or diagrams, or renderings, often exchanging those names without care. Some images are referred to as facts, others as fictions. Some are discussed as though they were both at once. All of this would be of little consequence if the register of everyday language were not welded to the deepest regions of thought and consciousness. There are, it turns out, consequences for our confusion—consequences for our thought and work, and ultimately for our lives.

This is a pathographic manifesto: a techno-political diagnosis of architecture after imaging. Neither a history nor a general theory of architectural images, it builds up a philosophical description of architecture’s contemporary technical consciousness, and its deepening immersion in the culture of electronic images that has swallowed all of life. It is an attempt to outline, as clearly as possible, what happens if the technical basis of architecture is examined very closely, if its technical terms and concepts are taken very seriously and, at times, literally. Unlike the lingering traditions of humanism that still haunt our thoughts, this essay dreams of a time when we have become at once less and more technical— less unknowingly immersed, and more deeply aware of ourselves as permanently and originally technical beings—and, with that awareness, more able to address the implications of our technicity, for ourselves and for the world.

Long before its descent into professionalization and academic irrelevance, one definition of philosophy was simply: to be prepared for whatever may come.2 Reflective thought once counted as a kind of continual acclimation to one’s circumstances—a psycho-immunological activity aimed, in Nietzschean terms, at simply becoming who one is.3 When freed from the demand for false exactitude, thought sheds any obligation to persuasion or utility, and becomes instead a kind of philological exercise regimen, in which one repeatedly lifts, one after another, a series of weighty conditions whose severity is irreducible to mere problem solving.

By regularly exposing oneself to both the joys of existence and the “horrors of the unmanageable,” a sensibility emerges, in which the certain and uncertain, knowable and unknowable, avoidable and unavoidable, are sorted out, worked over, and, ultimately, made bearable, tolerable, even delightful, no matter the shock of the initial encounter.4 Not because any specific questions have been answered, but simply because exercise is somehow, mysteriously and miraculously, therapeutic.

In this mood, we confront the technical collapse of historical consciousness in the design fields, rehearsing a project aimed at clarifying the status of computational images, so that we may eventually fashion a politics commensurate with our lived realities. Our primal scene throughout is Bruno Latour’s “new climatic regime”: a political condition belonging to ideas and ways of life that find their genesis in the tension between the inexorable force of neoliberal globalization and the psychosocial weight of the Anthropocene— twinned abysses that the language of modernity is hopelessly unequipped to navigate.5

In the midst of our serene new world of images, a descriptive revaluation of the conditions of imaging—of its technical basis, and the gestures that divide it off from all previous forms of visuality—is a prerequisite for architecture to pose the question, to itself and to culture, without recourse to irony or scientism: How can we learn to live differently? So differently that we might soon—very soon—become nonmodern?

Today there are no other questions.

  1. See John May, “Such as that Elegant Blend of Philosophy and Hardware: Preface to a Geographical Autonomy,” Thresholds 31 (2006): 8–15; John May, “Sensing: Preliminary Notes on the Emergence of Statistical-Mechanical Vision,” Perspecta 40 (August 2008): 42–53; John May, “The Becoming-Energetic of Landscape,” New Geographies 2 (February 2010): 23–32; John May, “The Logic of the Managerial Surface,” PRAXIS 13 (December 2012); Zeina Koreitem and John May, New Massings for New Masses: Collectivity After Orthography (Cambridge, MA: MIT SA+P Press, 2014); John May, “Field Notes from the Instruments Project,” Journal Of Architectural Education 69, no. 1 (March 2015): 58–61; John Harwood and John May, “If We Wake Up To Find We’ve Been Too Well-Trained,” in Architecture Is All Over, ed. Esther Choi and Marrikka Trotter (New York: Columbia Books on Architecture and the City, 2017), 178–190; John May, “Everything is Already an Image,” Log 40 (Spring/Summer 2017): 9–26; and Zeynep Çelik Alexander and John May, eds., Architecture and Technics: Archaeologies of Design Practice (Minneapolis, MN: University of Minnesota Press, forthcoming 2019).
  2. “Although our word ‘asceticism’ derives from the Greek word ‘askesis’ (since the meaning of the word changes as it becomes associated with various Christian practices), for the Greeks the word does not mean ‘ascetic,’ but has a very broad sense denoting any kind of practical training or exercise. For example, it was a commonplace to say that any kind of art or technique had to be learned by mathesis and askesis—by theoretical knowledge and practical training. And, for instance, when Musonius Rufus says that the art of living, techne tou biou, is like the other arts, i.e., an art which one could not learn only through theoretical teachings, he is repeating a traditional doctrine. This techne tou biou, this art of living, demands practice and training: askesis. But the Greek conception of askesis differs from Christian ascetic practices in at least two ways: (1) Christian asceticism has as its ultimate aim or target the renunciation of the self, whereas the moral askesis of the Greco- Roman philosophies has as its goal the establishment of a specific relationship to oneself—a relationship of self-possession and self-sovereignty; (2) Christian asceticism takes as its principal theme detachment from the world, whereas the ascetic practices of the Greco-Roman philosophies are generally concerned with endowing the individual with the preparation and the moral equipment that will permit him to fully confront the world in an ethical and rational manner.” Michel Foucault, “Techniques of the Parrhesiastic Games,” Discourse & Truth: Problematization of Parrhesia (one of six lectures, University of California at Berkeley, October–November, 1983), See also Jean-Pierre Vernant, Myth and Thought Among the Greeks (New York: Zone Books, 2006); Peter Sloterdijk, The Art of Philosophy: Wisdom as Practice (New York: Columbia University Press, 2012); Pierre Hadot, What is Ancient Philosophy? (Cambridge, MA: Harvard University Belknap Press, 2002).
  3. See Friedrich Nietzsche, Ecce Homo (New York: Penguin Classics, 1979).
  4. Zygmunt Bauman, Liquid Fear (Cambridge, UK: Polity, 2016), 72–95.
  5. See Bruno Latour, Down to Earth: Politics in the New Climatic Regime (Cambridge, UK: Polity, 2018).

Staring into the mundane depths of any modern technical system requires periodically making peace with an analytical absurdity: namely, that although one of the oldest fallacies concerning technical life is an imagined division between the mental and the material—as if thoughts (subjects) and actions (objects) are somehow separable realms of lived experience—we must nevertheless accept this division, if only temporarily, in order to understand the goals and motivations of the modern technical mind. This concession is required especially when contemplating the consciousness of system designers—meaning, today, all those electrical engineers, computer scientists, and ergonomists whose singular focus is programming the mediation between user and interface. “In the computer field, the moment of truth is a running program; all else is prophecy.”1

Imaging places the fact of automation at the center of our lives, but not in ways relatable to our historical traditions. Never simply the passing of labor from “humans” to “machines,” automation is a dream endemic to technics, always involving the enmeshment of consciousness and gestural routines within habituation processes that belong neither to the organic nor to the machinic, but which reside within both categories simultaneously. We know from Georges Canguilhem that it has always relied on practical-symbolic “theories of organic extension,” and is best understood through a “biological philosophy of technique” that is neither a history of industrialism, nor of Taylorism, nor of cybernetics, but rather a paleotechnical approach that refuses any stable categorization. No humans, no animals, no machines—even if these are the names with which industrialists and system designers (and Humanists) comfort themselves.2

Automation is the concretization of technical objects at points of genesis where the mental and the material codetermine each other in repetitive gestures, giving rise to new physiological processes and new lived experiences.3 “This technical phenomenon is the relation of the human to its milieu”—one in which the question of so-called “technological determinism” can never arise because “none of the terms of the relation hold the secret of the other.” Life cannot be determined by technics, nor vice versa, because the two categories are one and the same, moving, like combustion or catalysis, in processes where distinctions dissolve.4 It is “less a question of opposition than of a composition.” It involves what Gilbert Simondon called transduction: “a relationship whose elements are constituted such that one cannot exist without the other—where the elements are co-constituents.”5

The methods of automation that emerged during the age of orthography belonged to the techno-logic of mechanization—the essence of which, we know from Sigfried Giedion, consists in an “endless rotation” that motivates a geometric translation of energy and force.6 During the last phase of the technical age of mechanization, machines employed electricity only as a continuous source of power (let us call this phase electromechanical, in order to distinguish it from the cybernetic project that replaced it).7

As early as 1934, however, Lewis Mumford was able to write that “we have now reached a point in the development of technology itself where the organic has begun to dominate the machine,” where “instead of simplifying the organic, to make it intelligibly mechanical, as was necessary for the great eotechnic and paleotechnic inventions, we have begun to complicate the mechanical, in order to make it more organic: therefore more effective.”8

Workers on the first moving assembly line, at the Ford Highland Park Plant, put V-shaped magnets on Model T flywheels to make one-half of the flywheel magneto, 1913.
Workers on the first moving assembly line, at the Ford Highland Park Plant, put V-shaped magnets on Model T flywheels to make one-half of the flywheel magneto, 1913.

This relatively sudden transition from a “mechanical” to an “organic” ideology—which constitutes the essence of cybernetics—effected a transformation in what Leroi-Gourhan called the “technical tendency” that has seemingly always animated our lived reality.9 In the former (the mechanical), the foundational question behind all system design thinking was: to what extent can the organic be treated as mechanical? And because all mechanistic activity originated, paleontologically, in a desire to imitate and extend organic gestures (turning, pushing, lifting, etc.), the answer was: almost completely. Thus the whole history of mechanization, culminating in Taylorism’s nearly unthinkable results.10

“But the realization that technologically superfluous movements were biologically necessary was the first stumbling block” for this industrial mechanicism. “The systematic examination of certain physiological, psychotechnological, and even some psychological conditions… finally culminated in a reversal,” in which the organic was suddenly made to dictate the categorical terms of relation.11

In that moment, automation’s system design question was rotated on its axis and reformulated, becoming instead: to what extent can the mechanism be made to approximate, simulate, or imitate the organic? The answer: so perfectly and extensively that it instantly rendered the entire history of mechanization subordinate to a new family of apparatuses that can only be described as statistical-electrical (eventually, electronic), rather than electromechanical.12

Signalization absorbed mechanization. And in this great absorption, whole regions of biology and physiology—regulation, reflex, sensation, perception, sight and sound, the visible and the invisible—were exposed to the logic of statistical-electrical automation. Norbert Wiener made clear that the feedback logic of these new sensitive apparatuses already existed “as thermostats, automatic gyrocompass ship-steering systems, self-propelled missiles—especially such as seek their target—anti-aircraft fire-control systems.” “What is perhaps not so clear,” he continued, “is that the theory of the sensitive automata is a statistical one.”13

“Subject wearing experimental suit for physiological research,” circa 1895. Published in Otto Fischer, Der Gang des Menschen (Leipzig: B.G. Teubner, 1900).
“Subject wearing experimental suit for physiological research,” circa 1895. Published in Otto Fischer, Der Gang des Menschen (Leipzig: B.G. Teubner, 1900).

In signalization—understood broadly as the ongoing project of converting all experience into discrete, measurable, calculable electrical signals—automation is released from the prison of endless rotation and moves into thinner realms, into the topological and electrical, re-organized into processes concealed from our perception by their size and speed. Today we call this “machine learning,” or “artificial intelligence,” or the “digital revolution,” but the effort remains the same: the substitution of computational processes for lived mental states.

Signalized automation is in no way “thoughtless.” It reformats thinking, displacing it to different arenas. Unlike machines—whose “soul,” according to Bruno Latour, was “a pile of paper”—signalized apparatuses know only the logic of discretization, whose translation of force relies on an electrical communication among their parts. Mechanicism becomes servomechanicism. The two are conjunctive: mechanization aimed to automate manual labor, and signalization now aims to automate the mental processes that can be made to control automated manual-mechanical processes. (Isn’t this mental automation what we mean today when we say “parametric”?) In reality, both mechanization and signalization do, and have done, far more. They have brought entirely new forms of consciousness and ways of life into existence during their respective technical ages. Under the technical conditions of real time, signalization takes command, initiating an exhaustive reformatting of all previous thought and language—resulting in an entirely new orientation toward the world.

Against this background, we can consider a now-ubiquitous family of algorithmic operations that together go by the name “autocomplete,” and which now extend throughout our signalized lives (autofill, autotype, autocorrect, autoreplace). This group, which originated in efforts to assist persons with disabilities in the use of computational interfaces, modestly promises to speed up human-computer interactions, often by leaning heavily on the psychological principles of “primacy” and “recency.” That is, the entire category relies on our tendency to most easily remember the first and last instances in any nonroutine mental task: lists, directions, music choices, books purchased, financial information— and, ultimately, words themselves.

Design, too, has its own forms of autocompletion. Is it possible that the original “copy” command, in the first commercial release of AutoCAD in 1982, constituted a fundamental and decisive rupture in architectural reasoning? A rupture in which a whole series of incredibly laborious (that is to say, time-intensive) orthographic gestures were subsumed within an algorithmic logic whose aim was to automate that labor in the name of efficiency? To contemplate this chasm is not to suggest that either one, orthography or computation, is somehow better or worse, but merely to acknowledge that thinking takes place under radically different conditions in each.

If the world of the orthographer was simultaneously a text and a drawing, the world of the postorthographer is simultaneously an image and a model—an electrical image and an electrical model, signally mapped onto one another. Models are images that “refresh” at a speed anterior to perception. Because all signalization requires the materialization of a statistically managed signal-to-noise ratio, all postorthographic image-models are probabilistic in their underlying logic. If orthographic thought imagined and realized objects, in postorthographic surface labor, Antoine Picon asserts, “one no longer deals with objects but with theoretically unlimited series of objects.”14 Postorthography knows only objectiles, “conceived and produced as a single instance in a series.”15

Caught up in all this is time. The fact that, for many thousands of years prior to the emergence of orthographic writing, time was conceived of as a circle or cycle is proof that we are not born thinking linearly or historically.16 We trained this way of thinking into ourselves and our cultures by way of orthographic media: texts, drawings, and other forms of linear inscription and notation. Cultures can train themselves out of linear, historical thinking—and we are now doing just that, through our immersion in postorthographic surfaces.

Put in the most basic terms: if orthography was predicated on linear historical time, materialized in texts, drawings, numerical calendars, and mechanical clocks, postorthographic technical systems now enmesh our work in “real time,” materialized in signals and image-models. Unlike historical time, which was predicated on technical regimes and gestures that continually related present and future to the past, real time relates the present to all possible futures at once (or at least as many as can be recorded and computed). Real time is the time of statistical thought, in which futures knowable and unknowable are posed simultaneously, some more calculably probable than others, but all possible. This probabilistic conception of time is fundamentally different from the linear, mechanical conception that structured orthography.17

“Time can only be deferred… so-called real time is not time; it is perhaps even the de-temporalization of time, or at least its occultation.”18 Real time is, by design, fundamentally not the time of orthographic or preorthographic life, and this is its entire raison d’être.

The always-present experience of all calculably possible future states—the techno-logic of real-time modeling—is a very different imaginative framework than the orthographic imagination, which always drew upon (traced, overlaid, re-presented) the historical past to “project” the future. We see this difference in the kinds of evidence that are now used to justify architectural form. If the graphic language of historical precedent was once used to legitimize architectural objects, we now use the imagery and language of real-time data: images of performance, efficiency, fidelity, and control (or conversely, of noise, error, and electro-pastiche).19

It might be imagined that this technical succession simply involved the erasure of one set of routines and gestures by another. But in hindsight we can now recognize, in the transition from orthography to postorthography, the emergence of a sensibility that can only be described as a kind of intermediary consciousness. Can we now finally say what it has meant for architecture to use the word digital (or now, post-digital)? It has meant, and still means “pseudorthography” (or its reactionary correlate, “post-pseudorthography”). Pseudorthography is not at all “fake.” It is the very real, residual psychology of orthography laboring in the absence of its own technical-gestural basis—a pathology in which familiarity, that crucial element of comprehension, is preserved as a coping mechanism in the face of completely unfamiliar conditions.

Screenshot of Galapagos form optimization fitness test in Grasshopper/Rhino.
Screenshot of Galapagos form optimization fitness test in Grasshopper/Rhino.

Seen in this way, any account of digital architecture might equally be understood as a kind of genealogy of the last orthographers—a perfect reversal of Husserl’s speculations “back into the most original sense in which geometry once arose,” in which “it appeared in history for the first time”—and a generalized biography of an orthographic consciousness persisting within a technics that had no use for it, for which it in turn had no language, as it searched to resolve its previous world with technical arrangements that had completely destabilized its reality. To say this is not to underestimate the immense achievements of that period, but simply to establish a line of technical demarcation between two distinct architectural cultures, currently treated as though they were the same: those who were born into orthography but later learned to compute, and those who were simply born into—born as— computational imaging.

Pseudorthography is orthography after simulation: a mobile army of skeuomorphisms in which the world appears just enough as it used to. Immediately beneath those appearances is another world, “produced from miniaturized units, from matrices, memory banks, and command models—and with these it can be reproduced an indefinite number of times.” Untethered from the gestural basis of history, pseudorthography wanders instead through the past: an instantaneous archiving of the present, “nothing more than operational” from the standpoint of communication engineering, “the product of and irradiating synthesis of combinatory models.” The postorthographic past is not the historical past. It is an electrical distension of the present—a “temporary storage cache” of the “joy of instant participation.”20 (Does building information modeling—BIM—involve any less of this instant participatory euphoria than social media itself?)

Everything postorthographic finds its meaning and energy in the sensation of immediacy. Postorthography is “micro-archival by its very process-oriented algorithmic ontology,” but an electronic archive of past recordable events is not the same as historical consciousness, whose pace of contemplation and reflection are obliterated by David Joselit’s “epistemology of search.”21

Wherever architecture today imagines itself laboring over a drawing (“computer aided” or otherwise)—wherever it imagines the act of drawing as even still possible—we are in the presence of pseudorthography.22 Far from extinct, pseudorthography retains dominance over a current generation of practitioners who have recently and emphatically declared their allegiance to an imagined “culture of drawing.” At the same time, this generation blissfully immerses itself in the telematic ubiquity of the present, producing “drawings” in telematic image formats, advertising itself on telematic social platforms whose technical structures bear absolutely no relation to the gestural basis of drawing, and which want nothing more than to forget drawing, writing, and history in favor of real-time imagery.

Screenshot of Autodesk Revit image-model and simulated orthography.
Screenshot of Autodesk Revit image-model and simulated orthography.

Some still imagine that drawings will persist because they are needed to build buildings, and that this indexical connection will preserve orthography as the solid center of our practices. But in a technical sense, drawings have not been used to build anything in decades. Everything is now built from simulated orthography (imaging), with its attendant forms of transmission, duplication, repetition, and instantaneous modification— all of which have coalesced into a form of telematic managerialism unknown to orthography.

Images of drawings are not drawings. Using the “Make2D” command is not at all the same as drawing an orthographic plan, precisely because the time contained in their respective surface laboring is so radically different. “Lines,” “drawn” by computers or by the nostalgic hands of architects whose only oxygen is sociotelematic imagery, will never again amount to a drawing.

Images do not and cannot make drawings. They can only make more images, some of which we “print” by electromechanically depositing material (ink, starch, plastic, concrete) with a speed unimaginable to any orthographer.

The psychogestural residue of orthography is rapidly disappearing from an architectural culture becoming ever-more indistinguishable from telematic life itself. 

  1. Herbert A. Simon, The Shape of Automation for Men and Management (New York: Harper & Row, 1965), xv.
  2. Georges Canguilhem, “Machine and Organism,” trans. Mark Cohen and Randall Cherry, in Incorporations, ed. Jonathan Crary and Sanford Kwinter (New York: Zone Books, 1992), 61.
  3. “The zootechnological relation of the human to matter is a particular case of the relation of the living to its milieu, the former passing through organized inert matter—the technical object. The singularity of the relation lies in the fact that the inert, although organized, matter qua the technical object itself evolves in its organization: it is therefore no longer merely inert matter, but neither is it living matter. It is organized inorganic matter that transforms itself in time as living matter transforms itself in its interactions with the milieu.” Stiegler, Technics and Time 1, 49.
  4. Stiegler, Technics and Time 1, 52.
  5. Quoted in Stiegler, Technics and Time 2, 2.
  6. “The difference between walking and rolling, between the legs and the wheel, is basic to all mechanization.” Sigfried Giedion, Mechanization Takes Command: A Contribution to Anonymous History (New York: Oxford University Press, 1948), 47.
  7. Although both employ electricity, the distance between electromechanics and electronics is as drastic and thorough as that separating a toaster and a computer: the former is continuous, the latter discontinuous.
  8. Lewis Mumford, Technics and Civilization, 1934 (Chicago: University of Chicago Press, 2010), 367.
  9. Throughout his work, Leroi-Gourhan elaborates a paleontological claim regarding hominid technicity that he refers to as a general “technical tendency” seemingly animating all of hominization. See André Leroi-Gourhan, L’Homme et la matière (Paris: Albin Michel, 1943).
  10. “One of Taylor’s fundamental contributions to the methods of industrial management was his extension of the search for lost work beyond the physics of power transmission. He looked at the relationships among machines, the men who operated them, and those who maintained them.” Michael Osman, Modernism’s Visible Hand: Architecture and Regulation in America (Minneapolis, MN: University of Minnesota Press, 2018), 131.
  11. Canguilhem, “Machine and Organism,” 63.
  12. “Among the numerous modifications that scientific thought about nature has undergone in the course of the centuries, it would be difficult to point to one that has had a more profound and far reaching effect than the emergence of the conception of the world usually called mechanical or mechanistic.” E. J. Dijksterhuis, The Mechanization of the World Picture: Pythagoras to Newton, trans. C. Dikshoorn, 1961 (Princeton, NJ: Princeton University Press, 1986), 3. In a footnote to this same section, Dijksterhuis con-tinues: “It is difficult to fix upon an entirely satisfactory terminology. ‘Mechanical’ smacks too much of automatic in the sense of thoughtless. ‘Mechanistic’ in itself has no such objectionable connotation, but calls for ‘mechanism’ as the corresponding noun, a word which is, however, also used for the internal construction of a machine. We therefore prefer the noun ‘mechanicism’ for the designation of the system of thought” (emphasis mine).
  13. “To sum up: the many automata of the present age are coupled to the outside world both for the reception of impressions and for the performance of actions. They contain sense organs, effectors, and the equivalent of a nervous system to integrate the transfer of information from the one to the other. They lend themselves very well to description in physiological terms. It is scarcely a miracle that they can be subsumed under one theory with the mechanisms of physiology.” Norbert Wiener, Cybernetics: or Control and Communication in the Animal and the Machine (Cambridge, MA: MIT Press, 1948), 43.
  14. Antoine Picon, “Architecture and Mathematics; Between Hubris and Restraint,” in Architectural Design 81, no. 4 (2011): 35.
  15. Patrick Beaucé and Bernard Cache, “Towards a Non-Standard Mode of Production,” in Objectile: Patrick Beaucé and Bernard Cache, vol. 6, Consequence Book Series on Fresh Architecture (Vienna: Springer, 2007), 26–39.
  16. “Primitive thought appears to take place within a temporal and spatial setting which is continually open to revision … The fact that verbal language is coordinated freely with graphic figurative representation is undoubtedly one of the reasons for this kind of thinking… The thinking of agricultural peoples is organized in both time and space from an initial point of reference—omphalos—round which the heavens gravitate and from which distances are ordered. The thinking of pre-alphabetic antiquity was radial, like the body of a sea urchin or the star fish.” Leroi-Gourhan, Gesture and Speech, 211.
  17. See Harwood and May, “If We Wake Up to Find We Have Been Too Well-Trained.”
  18. Stiegler, Technics and Time 2, 63.
  19. See John May, “Under Present Conditions Our Dullness Will Intensify,” Project 3 (Spring 2014); and Zeynep Çelik Alexander, “Neo- Naturalism,” in “New Ancients,” ed. Dora Epstein Jones and Bryony Roberts, special issue, Log 31 (Spring/Summer 2014): 23–30.
  20. Jean Baudrillard, “Simulacra and Simulations,” in Jean Baudrillard, Selected Writings, ed. Mark Poster (Stanford, CA: Stanford University Press, 1988), 167; and Ernst, The Delayed Present, 38.
  21. David Joselit, “The Epistemology of Search: An Interview with David Joselit,” with Troy Conrad Therrien, ARPA Journal 2 (2014),
  22. For example, see Jacob, “Drawing in a Post-Digital Age”; and Dora Epstein Jones and Bryony Roberts, eds., “New Ancients,” special issue, Log 31 (Spring/Summer 2014).
“The ‘Cloud person’ with one random variable,” 1983. Published in Delle Rae Maxwell, “Graphical Marionette: A Modern-Day Pinocchio” (master’s thesis, MIT, 1983).
“The ‘Cloud person’ with one random variable,” 1983. Published in Delle Rae Maxwell, “Graphical Marionette: A Modern-Day Pinocchio” (master’s thesis, MIT, 1983).

Insofar as the world wells up within us in secret and mostly unseen ways—through techniques and routines, through institutions and habits—our work now finds itself immersed in an immense cultural experiment called imaging.

If our technics are always pulled taut between life and thought, if they are a fact in the world while also being a vision of that world, then we should see an architectural culture that disregards the philosophical dimensions of its own technical practices as self-delusional. Architecture devotes so much effort to establishing its possible influence over culture (its so-called agency) that it forgets to see culture’s overwhelming influence within it, imagining that somehow, through amnesia or mental isolation, its practices can resist the culture of imaging in which all of life now either swims or drowns.

What happens to the architectural mind when it stops pretending that images are drawings? When it finally admits that imaging is not drawing, but is instead something that has already obliterated drawing? What happens when it stops pretending that databases are the same as geometric objects?1 What happens when it realizes that all politics are first a politics of imaging? These are questions that architecture has scarcely begun to pose.

What is an image in our time? It is at once our field of experimentation and our field of politics. It is the technical format in which experimental lives—lives consciously lived differently than our own—might one day find not only their form but also, we hope, their political expression within a new statistical literacy capable of navigating the conditions of telematic culture.

The statistical time of imaging, and its relation to the disappearing historical time of orthography, contains the most pressing political questions facing contemporary architecture and urbanism. If we continue to confuse these two technical time signatures—if we continue to think of images simply as more-efficient drawings, or as technical enhancements of an otherwise undisturbed orthographic life—we will continue to drift unknowingly in an ocean of simulations for which we have no compass or concepts. The longer we refuse to see our work and our screens as belonging to a larger culture of imaging, the longer we continue to confuse images with drawings, the more apolitical we become. Telematic images, Flusser warned, “form bundles that radiate from centers, the senders… Bundles in Latin is fasces. The structure of a society governed by technical images is therefore fascist, not for any ideological reason but for technical reasons.”2

We should find ways of becoming image— of establishing meaningful expression within imaging itself, all the while acknowledging that our images no longer mean anything at all. This is the central paradox of our contemporary telematic consciousness: as imaging becomes the primary way in which we give meaning to our lives, the specific content of each individual image becomes less meaningful, bending toward meaningless.

The content of an image (its “meaning”) is now an automated ratio between signal and noise. Again, Kittler: “because systems of communication that would transmit a single message (e.g., the number pi, a determinate sine wave, or the Ten Commandments) are now superfluous and can be replaced by two separate signal generators, the messages themselves are as meaningless to information theory as their statistics are meaningful.”3 Any truly new telematic politics will, for this reason, no longer be a politics of the content of images but of the structure, composition, capacities, and limitations of imaging itself.4

Do we not already see in electrical images? Do we not now experience life first as transmissible? Telematic lives are statistical lives, constantly animated from a distance.

Going forward, architecture will effortlessly shed the historical consciousness of orthography. It cannot possibly hold. Orthographic consciousness cannot exist within postorthographic technics for more than one or two lived generations. At best, it makes an on-screen appearance, simulated and mummified in stunning resolution, on pseudorthographic software platforms that are both dream states and horizonless museums.5

Illustrations of “Digital Data Entry Glove Interface Device” by Gary J. Grimes, US Patent 4414537A, issued November 8, 1981.
Illustrations of “Digital Data Entry Glove Interface Device” by Gary J. Grimes, US Patent 4414537A, issued November 8, 1981.

Touching, swiping, scrolling, selecting, filtering, cropping, resizing, zooming, channelizing, compressing, tagging, batching—in short, image processing—are not minor expressions of technical systems external to thought, or instrumentalizable techniques with known or controllable affects. They are the gestural basis of an entire consciousness that, for now, continues to refer to our practices as architecture, but which will soon loosen and forget that name if architecture stubbornly clings to “the pieties of essentialism and persistence” and confuses “longevity with profundity.”6

No more drawings, only images. No more orthography, only telemetry. No more points, only addresses. No more geometry, only associations. No more syntax, only commands. No more tectonics, only management. No more machines, only apparatuses. No more stasis, only animation. No more research, only search. No more subjects, only users. No more contemplation, only transmission. No more representation, only presentation. No more perception, only sensation. No more ethics, only statistics. No more aesthetics, only physiology. No more archives, only caches. No more history, only retrieval. No more future, only probabilities. No more signification, only signalization.7

These are not predictions. They are a minimal accounting of exchanges that have already happened.

Only now, in the electric light of lives lived out after the end of orthography, are we finally becoming digital. 

  1. “Designing on an associative software program comes down to transforming the geometrical design in a programming language interface… to create a point at the intersection of two lines no longer consists of creating a graphic element, but in establishing a relationship of intersection on the basis of two relationships of alignment.” Beaucé and Cache, “Towards a Non-Standard Mode of Production,” 1.
  2. Flusser, Does Writing Have a Future?, 61.
  3. Kittler, “Signal-to-Noise Ratio,” in The Truth of the Technological World, 165.
  4. One need look no further than the Oval Office— which is no longer really a place but, increasingly, simply an electrical feed—for definitive evidence that our present political field is defined by the telematic tendency of the structural time of image-messaging to eliminate the historical meaning of its specific content. If, as Baudrillard once pointed out, Ronald Reagan was our first truly televisual president, Donald Trump is undoubtedly the first telematic president. His is a regime in which the specific contents of incessant messaging are nothing more than simulations of ideas, and power resides in the discursive ambiguities inherent in imaging.
  5. Hito Steyerl, “In Free Fall: A Thought Experiment on Vertical Perspective,” in The Wretched of the Screen (Berlin: Sternberg Press, 2012), 14–20.
  6. Robin Evans, Translations from Drawing to Building and Other Essays (Cambridge, MA: MIT Press, 1997), 15.
  7. “The architect is a worker whose mode of production is conditioned by digital technologies, but the development of these has nothing natural about it… in this field it is a strategic concept that will determine the form standard architecture will take in the years ahead: this is the concept of associativeness. What are we to understand by associativeness? Associativeness is the software method of constituting the architectural project in a long sequence of relationships from the first conceptual hypothesis to the driving of the machines that prefabricate the components that will be assembled on site. Designing on an associative software program comes down to transforming the geometrical in a programming language interface.” Beaucé and Cache, “Towards a Non-Standard Mode of Production,” 68.

Columbia Books on Architecture and the City
An imprint of the Graduate School of Architecture, Planning, and Preservation

Columbia University
1172 Amsterdam Ave
407 Avery Hall
New York, NY 10027

Distributed by Columbia University Press

Signal.Image.Architecture. (Everything is Already an Image)
By John May

Graphic Design:
Laura Coombs

Copyeditor: Stephanie Salomon


© 2019 by the Trustees of Columbia University in the City of New York
Text © the author.

All rights reserved.

No part of this book may be used or reproduced in any manner without the written permission of the publisher, except in the context of reviews. Every reasonable attempt has been made to identify the owners of copyright. Errors or omissions will be corrected in subsequent editions.

This book has been produced through the Office of the Dean, Amale Andraos, and the Office of Publications at Columbia University GSAPP.

Director of Publications: James Graham Assistant Director: Isabelle Kirkham-Lewitt Managing Editor: Jesse Connuck

Library of Congress Cataloging-in-Publication Data

Names: May, John (Architect), author. Title: Signal. image. architecture. : everything is already an image / John May. Description: New York : Columbia Books on Architecture and the City, 2019. | Includes bibliographical references and index. Identifiers: LCCN 2019020856 | ISBN 9781941332467 (pbk. : alk. paper). Subjects: LCSH: Architecture–Philosophy. | Image (Philosophy) | Architecture and technology. | Digital images. Classification: LCC NA2500 .M377 2019 | DDC 720.1–dc23. LC record available at