PRESENTED AT:
	ISEA (International Symposium on Electronic Art), Helsinki, 1994.

--------------------------------------------------------
Lev Manovich

THE LABOR OF PERCEPTION:  ELECTRONIC ART IN POST-
INDUSTRIAL SOCIETY

I.
--------------------------------------------------------
Just as it would be futile to consider video art in 
isolation from television, it would be equally 
unproductive to theorize new emerging forms of computer 
art without considering their uneasy connections to 
contemporary image industries, such as the computer 
graphics industry. Computer artists need this industry 
to provide them with the latest technological toys which 
will set them apart from their colleagues still working 
in the traditional, pre-industrial mediums. The industry 
uses the artists as beta-testers for new software and 
hardware. More importantly, the industry uses the 
mythology of art -- our Romantic-modernist belief that 
the artist is a unique person, a visionary who 
transcends the everyday reality and pushes the 
boundaries, etc. -- as the most effective sales tool. 
What better way to market a piece of software than to 
have an endorsement from the artist? (Thus, 
paradoxically, computer artist is somebody who 
transcends the here and now in the act of creation, but 
can do so only with the help of the very latest tools, 
the tools of here and now). 
	If computer art does not exist in isolation from 
computer graphics industry, let us examine the history 
and the direction of the industry. Why did computer 
graphics -- the industry concerned with finding more 
effective ways to produce, store, distribute and present 
images -- achieve such importance? Why is it that today 
new disciplines which study images and vision continue 
to expand:  image processing, computer vision, research 
on human-computer interfaces, vision science, and so on? 
What are the reasons these currently prominent image 
industries and image sciences have acquired such 
prominence?
	Let us begin with three images (figures 1, 2, 3).  
	The first image: a portrait of Tatlin by a fellow 
Soviet designer El Lissitsky (figure 1). Time: early 
1920s. A compass, extending straight from Tatlin's eye, 
a metaphor of vision for work. 
	The second image:  SAGE (the "Semi-Automatic Ground 
Environment") -- the first human-machine interactive 
display system (figure 2). Time: mid 1950s.
	The third image: virtual reality interface designed 
at NASA/Ames Human Factors Research Center (figure 3). 
Time: now. Instead of the metaphor of the eye-compass, a 
reality: video monitors strapped to the eyes. The notion 
of vision as work is now fully realized: the operator 
wearing the gear works by mentally processing visually 
presented information. The gear is designed using all 
the available knowledge accumulated by experimental 
psychology about human vision. In the photograph we see 
the last leftover from the age of manual labor -- an arm 
in a DataGlove. It will soon disappear since through 
gaze tracking the operator can control the system by 
merely looking at different points in virtual space.

II.
--------------------------------------------------------
Modernization brought with it a special discipline 
concerned with efficiency -- engineering. The job of an 
engineer was to ensure maximum performance with a 
minimum investment of energy, materials, and time, be it 
the performance of machines (mechanical engineering), 
communication systems (communication engineering) or 
human bodies (scientific management, time and motion 
studies). Inspired by modern engineering, the avant-
garde of the 1920s tried to systematically apply its 
principles to vision. 
	To engineer vision meant to eliminate waste, to use 
minimal material resources. Thus, constructivist graphic 
design streamlined typography, eliminating complicated 
typefaces in favor of block letters consisting of 
straight lines; it also eliminated illustrations and 
"wasteful" decorations by making type itself the main 
element of design. The goal: maximum impact with minimum 
use of ink (figure 4). 
	To engineer vision also meant to minimize the 
psycho-physical resources required of the viewer. Dziga 
Vertov writes in his famous 1923 manifesto: "The least 
advantageous, the least economical communication of a 
scene is theatrical communication." [1] In contrast, 
montage forces the eye to see the right thing at the 
right time, thus eliminating the visual waste of 
theater, ballet, painting, and other traditional forms. 
In montage, "camera drags the eyes of a film viewer from 
hands to legs, from legs to eyes and the rest in the 
most advantageous order..." [2]
	To engineer vision also meant to ensure perception 
in the shortest possible time. Here as well, the avant-
garde promoted montage as an example of possible 
economy, in this case economy of time. Maud Lavin 
describes the 1930 manifesto of the group of leading 
German designers headed by Kurt Schwitters: "Walter 
Dexel writes that modern man has the right to expect 
communications in the shortest possible time. Willi 
Baumeister points out that photomontage is efficient, 
allowing for the quick grasp of several images at once." 
[3]
	Finally, to engineer vision also meant to be able 
to measure its efficiency, or, to use the language of a 
communication engineer, to measure "system performance." 
Eisenstein, fresh from engineering school, invented his 
first theory of artistic communication, the famous 
"montage of attractions":  "Let us search for the unit 
which will measure the influence exerted by art! Science 
has its 'ions,' its 'electrons,' its 'neutrons.' Art 
will have -- attractions!" [4]
	To summarize: The job of the avant-garde artist was 
to engineer vision, and to engineer vision meant to 
affect the viewer with engineering precision, 
predictability, and effectiveness.  

III.
--------------------------------------------------------
In its desire to engineer vision, the avant-garde was 
ahead of its time. The systematic engineering of vision 
took place only after World War II with the shift to 
post-industrial society.
	For post-industrial society, mental labor of 
information processing is more important than manual 
labor. In contrast to a manual worker of the industrial 
age (figure 5), an operator in a human-machine system 
(figure 6) is primarily engaged in the observation of 
displays which present information in real time about 
the changing status of a system or an environment, real 
or virtual: a radar screen tracking a surrounding space; 
a computer screen updating the prices of stocks; a video 
screen of a computer game presenting an imaginary 
battlefield; a control panel of an automobile showing 
its speed, etc. In short, vision becomes the major 
instrument of labor, the most productive organ of a 
worker in a human-machine system. And this is why 
following World War II we witness unprecedented amount 
of research into imaging and vision.
	The figure which stands at the gates to this post-
industrial society of perceptual labor is a radar 
operator of World War II. 
	1. First of all, in order to ensure the maximum 
performance of such human-machine system as radar, it 
became necessary to engineer it around the capacities 
and the limitations of human vision. At the end of the 
World War II, a new field emerges -- human engineering. 
Let me quote from the description of its history found 
in an 1965 overview of the field:

"The primary emphasis in time-and-motion 
engineering has been on man as a worker; that is, 
as a source of mechanical power. It was not until 
World War II that a new category of machines 
appeared -- machines that made demands not upon the 
operator's muscular power, but upon his sensory, 
perceptual, judgmental, and decision-making 
abilities. The job of a radar operator, for 
example, requires virtually no muscular effort, but 
makes severe demands on sensory capacity, 
vigilance, and decision-making ability. This new 
class of machines raised some intricate and unusual 
questions about human abilities: How much 
information can a man absorb from a radar screen?" 
[5]

Already before the war, experimental psychologists 
assisted in selecting military personnel for such jobs 
as pilot or airplane observer by administering special 
aptitude tests. During the war, a much greater number of 
pilots, radar operators and other similar personnel 
became needed. The emphasis was shifted, therefore, from 
selecting personnel with particularly good perceptual 
and motor skills to designing the equipment (controls, 
radar screens, dials, warning lights) to match the 
sensory capacities of an average person.[6] And it was 
the field of experimental psychology that possessed the 
knowledge about the sensory capacities of an average, 
statistical person: how visibility and acuity vary 
between day and night; how the ability to distinguish 
colors and brightness vary with illumination or 
distance; what the smallest amount of light is which can 
be reliably noticed; and so on.[7] All this data was now 
utilized for designing better displays and controls of 
the first modern human-machine systems such as radar 
installations or high-speed aircrafts.
	The term "human engineering" was eventually 
replaced by another term standard today -- "human 
factors." The radar operator who in the 1940s and 1950s 
was the prototypical example of a human-machine system, 
was replaced by the 1980s by a new prototypical figure, 
the computer user. Thus, references to "human-machine 
systems" became references to "human-computer systems." 
The same amount of intellectual energy and research 
which in the middle of the century went into theorizing 
the performance of a radar operator and adapting him and 
radar display to each other, today goes into the work on 
new computer interfaces, such as NASA/Ames VR system 
(figure 3).
	
2. The work on radar also directly leads to the 
development of interactive computer graphics. Next to 
photography, radar provided a superior way to gather 
information about enemy locations. In fact, it provided 
too much information, more information than one person 
could deal with. Was there a way to process and display 
information gathered by radars more effectively? The key 
principles and technologies of computer graphics -- CRT 
(cathode-ray tube) display, bit-mapped graphics, 
interactive control, were developed as a way of solving 
this problem. The research took place at MIT. After the 
end of the War, Air Force created a secret  Lincoln 
Laboratory. The job of Lincoln Laboratory was to work on 
human factors and new display technologies for SAGE -- 
the "Semi-Automatic Ground Environment," a command 
center to control the U.S. air defenses established in 
the mid-1950s.[8] The earlier version of the center, 
called Cape Cod network, was operating right out of the 
Barta Building at MIT.
	Each of 82 Air Force officers was monitoring his 
own computer display which showed the outlines of New 
England Coast and locations of key radars (figures 6, 
7). Whenever an officer would notice a dot indicating a 
moving plane, he would use a light gun to tell the 
computer to track this dot.[9] 
	This was the first human-machine interactive 
computer graphic display system, developed to alleviate 
the mental labor of information processing. Vision, 
enchanced by computer graphics technology,  became the 
only means to deal with information overflow.  

IV.
--------------------------------------------------------
Computer graphics helped to process radar information 
more efficiently, but was there a way to take the human, 
who was too slow to keep up with the computers, 
completely out of the loop ? This is the third crucial 
development in engineering of vision -- the work on 
computer vision.
	In 1961, the National Photographic Interpretation 
Center (NPIC) was created to produce photoanalysis for 
the rest of the U.S. intelligence community and, as 
Manual De Landa points out, by the end of the next 
decade computers "were routinely used to correct for 
distortions made by satellite's imaging sensors and by 
atmospheric effects, sharpen out-of-focus images, 
extract particular features..." Computer analysis of 
photographic imagery also became the only way to deal 
with the pure volume of intelligence being gathered.  
	The techniques of image processing, which can 
automatically increase an image's contrast, remove the 
effects of blur, extract edges, record differences 
between two images, and so on, greatly eased the job of 
human photoanalysts. But was it possible to completely 
replace them by computers? 
	 Roberts' 1965 paper "Machine Recognition of Three-
dimensional Solids" is considered to be the first 
attempt at solving the general task of automatically 
recognizing three-dimensional objects.[10] His program 
was designed to understand the artificial world composed 
solely of polyhedral blocks (figures 8, 9). Using image 
processing techniques, a photograph of a scene was first 
converted into a line drawing. Next, the techniques of 
3-D computer graphics were used, also developed by 
Roberts. Thus, the two fields were born simultaneously: 
3-D computer graphics and computer vision, automation of 
imaging and of sight.

--------------------------------------------------------
In summary, the rise of modern image industries and 
image sciences, such as computer graphics, human-factors 
research or computer vision, can be seen as a part of 
the shift to the post-industrial society of perceptual 
labor. This shift involves two processes -- two stages 
of automation. 
	First stage of automation: human and machine are 
integrated in new human-machine systems which 
increasingly came to dominate both the battlefield and 
the workplace after World War II (radar screen, aircraft 
controls, computer terminals of the automated factory). 
Human vision became the key instrument of post-
industrial labor as the channel of communication between 
human and machine. This leads into research into more 
efficient human-machine interfaces -- from Ivan 
Sutherland's Sketchpad to today's VR.  
	Second stage of automation: the complete 
replacement of human cognitive functions by a computer, 
such as the substitution of human vision by computer 
vision.  What does it mean to teach a computer how to 
see? In the field of computer vision, "vision" refers to 
two goals. First, it means the identification of various 
objects represented in an image. Second, it means 
reconstruction of three-dimensional space from the 
image. For instance, a missile not only has to identify 
a target but also to determine the position of this 
target in three-dimensional space. Here, vision is not 
meant for the contemplation of a sunset or appreciation 
of art; instead, it is reduced to the common denominator 
shared by humans and low level organisms: to detect an 
obstacle, a predator, a prey.   
	I believe that most of the new research into vision 
and imaging after World War II can be understood as 
following these two directions: on the one hand, making 
human vision in its new role of human-machine interface 
as efficient, as productive as possible; on the other 
hand, transferring vision and other human cognitive 
capacities from human to a computer. 

V.
--------------------------------------------------------
What does this analysis entails for forming aesthetic 
criteria by which we can judge computer art? Let us look 
at the two paradigms in turn.
	First, as I pointed out, in a post-industrial 
society vision acquires a new role of  human-machine 
interface -- from radar screens of World War II to such 
contemporary developments as VR. The industry aims to 
make human vision as productive, as efficient as 
possible.  If we still believe that art is something 
which is anti-productive, anti-utilitarian, the computer 
artist can be defined as designer of bad interfaces: 
interfaces which are inefficient, wasteful, confusing. 
One example of such "bad" interfaces is a display where, 
instead of usual modernist clarity, or "good form," the 
viewer encounters formlessness, chaos, "the madness of 
vision" (figure 10).[11] 
	Another example can be a pseudo-interactive work: a 
screen with a menu where every choice gets you to the 
same place. 
	Second, since we are also witnessing a movement 
towards the complete automation, including the 
replacement of human vision by computer vision, we need 
to completely reevaluate the very term "computer art." 
The term presently refers to the making of art with the 
help of a computer, the art to be enjoyed by human 
observers. The artist is the one who makes the creative 
choices. This Romantic paradigm reaches its extreme in 
the recent trend of artificial life art, where the 
computer is programmed to simulate the laws of 
evolution, mutating images to create endless new 
combinations; while the artist assumes the role of God, 
selecting which of these images will survive.
	I suggest to redefine "computer art" to mean "art 
for computers," art to be enjoyed not by humans but by 
computers. Moreover, using the tools of expert systems, 
artificial life and neural networks, we can evolve not 
only computer artists -- the programs to create images 
-- but also computer critics, the programs to evaluate 
them. What kind of images will be pleasurable for a 
computer? It is hard to make predictions, but I can 
guess that following its human master, the computer will 
adopt efficiency as the main aesthetic criteria. Thus, 
the computer may prefer images which are efficient in 
terms of storage -- images which compress well. 
Rewriting art history from this perspective, the 
computer critic will prefer minimalist abstraction to 
Jackson Pollock, and will champion Malevich, as the most 
important artist of the twentieth century -- the artist 
who anticipated the aesthetics of compression, and thus 
was already ahead of today's computer artists who still 
try to resist the poetics of the productive, functional, 
industrial (figures 11, 12).
	As  Dziga Vertov wrote in 1923, "I am a mechanical 
eye." [12]

VI.
--------------------------------------------------------
The preceding examples, of course, should be taken only 
half seriously. My main point is to urge computer 
artists to examine their relationship to the computer 
graphics industry, and to address the impact of this and 
other contemporary image industries not just on art 
practice but on society at large. 
	 The notion that the artist functions outside of 
society, history and industry is a modernist myth. 
Modernist artists were not only the pioneers of the 
utilitarian aesthetics of modern industrial design or 
the pioneers of the techniques of modern advertisement 
and political propaganda; as I suggested in this essay, 
they have also pioneered post-modern engineering of 
vision, the integration of human and machine in human-
machine systems and the replacement of human by computer 
vision. Today computer graphics industry is one of sites 
of this engineering. Whether computer artists 
acknowledge or ignore their relationship to this 
industry, it exists. Acknowledging rather than ignoring 
this is the first step toward a critical computer art 
practice.


--------------------------------------------------------
NOTES

1. Dziga Vertov, "Kinoki. Perevorot" (Kinoki. A 
revolution), LEF 3 (1923): 139.
2. Ibid., 139. Emphasis in the original -- L.M.
3. Maud Lavin, "Photomontage, Mass Culture, and 
Modernity. Utopianism in the Circle of New Advertising 
Designers," in MONTAGE AND MODERN LIFE: 1919-1942, ed. 
Matthew Teitelbaum (Cambridge: The MIT Press, 1992), 54.
4. Qtd. in Jacques Aumont, MONTAGE EISENSTEIN (London and 
Bloomington: BFI Publishing and Indiana University 
Press, 1987), 41. Emphasis mine -- L.M.
5. Alphonse Chapanis, MAN-MACHINE ENGINEERING (Bemont, 
CA: Wadsworth Publishing Company, Inc., 1965), 9-10.
6. Ibid., 8.
7. William Estes, "Experimental Psychology: an Overview," 
in THE FIRST CENTURY OF EXPERIMENTAL PSYCHOLOGY, ed. 
Eliot Hearst (Hillsdale, NJ: Lawrence Erlbaum 
Associates, Publishers, 1979), 630.
8. See Paul Edwards,  "The Closed World. Systems 
discourse, military policy and post-World War II US 
historical consciousness," in CYBORG WORLDS: THE 
MILITARY INFORMATION SOCIETY,  ed. Les Levidow and Kevin 
Robins  (London:  Free Association Books, 1989); Howard 
Rheingold,  Virtual Reality (New York: 1991).
9.  Panel proceedings of SIGGRAPH '89 (Boston, Mass., 
July 31-August 4, 1989), in COMPUTER GRAPHICS 23, 5 (ACM 
SIGGRAPH: New York, 1989), 22-24.
10.  L.G. Roberts, "Machine perception of three-
dimensional solids," in OPTICAL AND ELECTO OPTICAL 
INFORMATION PROCESSING, ed. J.T. Tippett (Cambridge: The 
MIT Press, 1965) .
11. The notion of "the madness of vision" is explored by 
the French philosopher Cristine Buci-Glucksmann. 
Describing her work, Martin Jay writes: "Resistant to 
any totalizing vision from above, the baroque explored 
what Buci-Glucksmann calls 'the madness of vision,' the 
overloading of the visual apparatus with a surplus of 
images in a plurality of spatial planes. As a result, it 
dazzles and distorts rather than presents a clear and 
tranquil perspective on the truth of the external 
world." Martin Jay, DOWNCAST EYES: THE DENIGRATION OF 
VISION IN TWENTIETH-CENTURY FRENCH THOUGHT (Berkeley: 
University of California Press, 1933), 47-48.
12. Vertov, "Kinoki,"  141.