The self-referential closure of the form includes the question of the
observer
as the excluded third.
Niklas Luhmann - Art of the Social System
We are always dealing with a historical process that consists of opera- tions, that is, with an evendike process which can realize itself only in time.
The observation accompanying the creation of a work differs from the ob- servation of the beholder in that the former can occur only once, whereas the latter can be repeated (and is thus subject to variation).
With respect to the operativity of its observation, the artwork must therefore be a tempo-
rally abstractedstructure. It is a program for repeated usage which--like to- day's complex computer programs--blocks access to what is actually going on during the execution of the operation. To put it differently, the artwork itself offers no clue as to what happens in the process of understanding the
If,
44 Perception and Communication: The Reproduction ofForms
work. It leaves open how artist and beholder are coupled by the work, while at the same time it guarantees that this coupling is not entirely ar- bitrary--this is what makes art a medium of communication. Physicists would perhaps speak of nonlinear structures of coupling; at any rate, they would exclude tight coupling and noncoupling.
Finally, we should emphasize that our difference-theoretical approach --the idea of tracing art to die unfolding of a paradox of form--implies a radically historical mode of observation: "If words and concepts receive meaning only in sequences of differences, one can justify one's language, and one's choice of terms, only within a topic [an orientation in space] and
104
an historical strategy. "
like everything else, that it comes into being and vanishes. Rather, it ex- plains why works of art must be constructed with an orientation toward time, why they must be new, following the requirement, in place since early modernity, that diey distinguish themselves from everything that has been done before. The demand for novelty means not only that no two things are the same but also that the difference between them motivates both the work's production and die interest of die beholder. Only novel works can please. This is how art exposes itself to die effects of a self-consummation of forms. It positions itself historically. And this is why die future of art be- comes increasingly problematic--to die point where die claim is made and refuted daily that art has no future in the posthistoire.
IX
To determine how art defines itself, one needs to rely on what can be known about works of art. What is it that characterizes art in relation to everything else (to its unmarked space)? The tradition offers a number of specific distinctions. Works of art are made, in contrast to natural objects. Once artificiality as such no longer defines what counts as art, a second distinction is added: works of art serve no externalpurpose; having such a purpose is precisely what disqualifies a work as art. This raises the ques- tion what else could qualify works of art as art. With this question, the theory of art receives its admission ticket to the art system--until die the- ory of die avant-garde will proclaim that art is whatever we call art, thereby accomplishing the dual purpose of making works of art exemplify theory while at the same time relieving itself from the burden of further reflection.
This does not simply mean that art has its time
Perception and Communication: The Reproduction of Forms 45
For lack of a convincing alternative, we shall draw repeatedly on this historically exhausted theoretical framework. However, our considerations concerning perception and communication may yield a somewhat more complex treatment of the topic, particularly via the concept of form.
To say that artistic forms call attention to themselves as being made and without purpose circumscribes the reentry of the form into the form. Any artwork displays talent, no matter whether its figures are beautiful or ugly, distinguished or ordinary, benevolent or malicious, profound or nonsen- sical. The work displays, one might say, itself and its own self-description. It executes the paradox of "reentry" and shows that this is possible, what- ever mathematics or logic may think.
The work of art draws on sensuously perceptible media for its own self- explication, no matter what is subsequently presented as an internal play of forms. It exploits evidence based on such media. Even if one knows that this happens only for the sake of communication, the fact that art draws on perception is not irrelevant to understanding how it happens. The question, then, is the following: How does an individual work of art - present itself to perception in such a way as to be recognizable as art and to provide, by virtue of precisely this recognition, an opportunity and a motive for participating in communication?
The concept of form suggests that two requirements must be fulfilled and inscribed into perception: the form must have a boundary, and there must be an "unmarked space" excluded by this boundary. How both of these requirements coincide and how they are fulfilled in one stroke may vary considerably in different artistic genres. Whenever we think of "marked space"/boundary/"unmarked space" in combination, the constitution of an imaginary space is at stake. However, since every artwork constitutes its own imaginary space, the question is how this space is constituted differ- ently in each case.
The typical case is a work of art enclosed within a beginning and end, within a frame or a stage, a work that ignores and does not interfere with its surroundings. In this case, the imaginary space is construed from the inside out, as if breaking through the frame or creating its own world be-
105
hind the frame. The imagination is driven beyond the work.
at once see and think away the frame in order to gain access to the work's imaginary space. The guaranteed repeatability of observations might be of help in performing this operation.
Sculpture or architecture presents an entirely different case. Here, the
One must
46 Perception and Communication: The Reproduction ofForms
boundary does not draw the viewer's attention inward but instead directs it outward. The work permits no view into its depths, no penetration of its surface (whatever the surface may betray of the work's mass, volume, or material). The imaginary space is projected outward in the form of dis- tinctions suggested by the work itself. Here, too, space is work-specific space, visible so long as the focus is on the work and disappearing from view when the focus shifts to surrounding objects--to the weeds in the casde garden.
106
The boundary cannot be perceived,
rects our attention inward or outward. The boundary itself can be shaped as a form--as a portal, an ornament, or a movement on the sculpture's surface, as the gorgeous frame of a painting or merely as a well-chosen one. Once we recognize this, however, we no longer see the boundary as boundary but observe differences between forms, which--one thing lead- ing to another--we attribute to the work itself.
Classical aesthetics, which presupposed a creative or receptive subject in all of this, had no difficulties with these problems. Everything could be ex- plained with reference to the enigmatic nature of the subject. The strict distinction between perception and communication, which dissolves the subject, changes this situation. Now it becomes important that, and in what ways, the boundaries of the individual work of art mark the struc- tural coupling between perception and communication. Boundaries, however, cannot be observed in their capacity as structural couplings, since neither the perceiving consciousness nor communication can ex- plode their operative closure and reach beyond their own systems into the environment.
Giving up the notion of the subject requires reconstructing the object,
107
which loses its opposite.
"unmarked space," objects appear as repeated indications, which, rather than having a specific opposite, are demarcated against "everything else. " Objects are forms whose other side remains undetermined. The unattain- ability of its other side accounts for the object's concreteness in the sense that determining its unity "as something" becomes impossible. Every analysis remains partial, depending on further specification of the other side--for example, with regard to color, magnitude, purpose, consistency.
George Herbert Mead (following Whitehead) assigned a function to identifiable and recognizable objects, whose primary purpose is to bind time. This function is needed because the reality of experience and actions
unless we know whether it di-
If one starts out from the counterconcept of the
Perception and Communication: The Reproduction of Forms 47
consists in mere event sequences, that is, in an ongoing self-dissolution. Since the experience of others must be assumed to occur simultaneously if communication (Mead: interaction) is to take place, the reality of the other's operations remains in principle inaccessible; it can, however, be
108
symbolized via the identification of objects.
Michel Serres suggests that the stabilization of objects (identification,
recognizability, and so on) is more likely to contribute to stabilizing social
109
relationships than the famous social contract.
Heinz von Foerster arrives at the notion of objects as eigen-behaviors of
110
recursive calculations.
cursive self-application of communication contribute more than any other kinds of norms and sanctions to supplying the social system with necessary redundancies. This may be even more true of objects that have been invented for the sake of this specific function, such as kings or soc-
111
112
Presumably, the objects that emerge from the re-
cer balls.
to this function. They assume a sufficient amount of variance, of recog- nizability in different situations, to keep up with changing social constel- lations. In contrast to concepts that are determined by specified ant- onyms, such quasi objects maintain their concreteness as objects--in the sense of excluding the unmarked space of all other events and conditions --throughout changing situations.
Works of art are quasi objects in this sense. They individualize them- selves by excluding the sum total of everything else; not because they are construed as given but because their significance as objects implies a realm of social regulation. One must scrutinize works of art as intensely and with as close attention to the object as one does when watching kings and soccer balls; in this way--and in the more complex case, where one ob- serves other observers by focusing on the same object--the socially regu- lative reveals itself. The object relation thus helps differentiate recursive connections between observations--the court, the soccer game, the art scene--which subsequently construct their own guiding objects.
This is how the exclusion of the unmarked space is carried along--and forgotten. It might as well be left to religion.
X
Consciousness cannot communicate, communication cannot perceive --this is where we began. Correspondences between modes of operation
Such "quasi objects"
can be comprehended only in relation
Via a different route,
48 Perception and Communication: The Reproduction ofForms
as disparate as these must be formulated in a highly abstract manner. Af- ter all, we are dealing with two entirely different, operatively closed sys- tems that do not interfere with each other. With concepts such as the self- referential event, distinction, form, and paradox, we have gained the necessary level of abstraction. This conceptual background, as we sug- gested earlier, permits conclusions concerning the specificity of art. Art
j makes perception available for communication, and it does so outside the standardized forms of a language (that, for its part, is perceptible). Art cannot overcome the separation between psychic and social systems. Both types of system remain operatively inaccessible to each other. And this ac- countsfor the significance of art. Art integrates perception and communica- tion without merging or confusing their respective operations. Integration means nothing more than that disparate systems operate simultaneously (are synchronized) and constrain one another's freedom. By participating in communication via perception, the psychic system might generate in- tensities of experience that remain incommunicable as such. In order to produce such experiences, it must be able to perceive differences between the forms created in the social system of art for the purpose of communi- cation. Communication through art must present sensuously perceptible objects without being able to reproduce itself within individually encap- sulated psychic systems. The need for structural coupling and the oppor- tunity to establish such coupling impose rigorous demands upon the forms that specify and determine a work of art at the boundary between psychic and social systems.
A kind of quantum mechanical solution to the problem of integration
is conceivable if one thinks of forms in terms of a distinction between two sides. One can always count on the other, operatively inaccessible system to operate in a binary manner, as a system that indicates one side of an ac- tualized form and excludes (for the time being) the other. That much one can expect from communication with regard to perception, and from per- ception with regard to communication, although the rich, reference-filled inner horizons of these systems remain mutually inaccessible. Forms, in other words, ensure identity and difference at once: identity through their fixed outlines, and difference through the recursive system reference of the op- erations that actualize these outlines--in the form of a contrasting per- ception or intuition, or of a place where communication can continue by understanding and reconstructing its connective possibilities.
Since works of art are objects that bind time, such an integration can be
Perception and Communication: The Reproduction of Forms 49
synchronized. It outlives the sheer eventfulness of system operations for as long as a consciousness is occupied with the work of art. Since the focus is on an object, this can happen recursively, that is, by recapitulating and anticipating other perceptions of forms. Recursions of this type generate so-called "aha" experiences, sudden flashes of insight that reveal the or- dered nexus of the artwork. In this case, the coupling is unique in that it does not require a merging of psychic and social systems. Consciousness
remains entirely within itself.
Our analysis has reached a point where it becomes apparent that, and in what ways, communication through art tends toward system formation and eventually differentiates a social system of art. In the following chap- ters, we shall deal both with the historicity of this inner-social process and with its consequences. For the time being, our concern is to show, on the basis of our analysis of artistic communication, that, and in what ways, system formation is possible--much as the spectacular individual exis- tence of artworks and the diffuse heterogeneity of observer perspectives may speak against such an approach.
The problem of system formation is connectivity, the recursive reusabil- ity of events. Operations (conscious perceptions as well as communica- tions) are nothing more than events. They cannot persist, nor can they be altered. They emerge and vanish in the same instant, taking no more time than is needed to fulfill the function of an element that cannot be decom- posed any further. The art system has no reality except at the level of ele- mental events. It rests, one might say, on the ongoing dissolution of its el- ements, on the transitory nature of its communications, on an all-pervasive entropy against which anything that persists must organize itself. Concepts such as connectivity or recursive reusability indicate this process, but they do not explain it. They only show that the stability of a system based on time-sensitive events must be a dynamic stability, a stability that depends on the continual change of the system's resources.
We shall call this state of affairs an "autopoietic system. " This means that the elements of the system are produced within the network of the System's elements, that is, through recursions. A communication cannot occur as an isolated phenomenon, as a singular event brought about by a combination of physical, chemical, living, and psychic causes. Nor can it proceed through simple replication, merely by substituting disappearing elements for one another. It is not enough--and it wouldn't work, any- way--to repeat what has been said (shown, perceived, or thought) once it
50 Perception and Communication: The Reproduction ofForms
dies away. Something else, something new must follow, for the informa- tional component of communication presupposes surprise and gets lost in
113
repetition.
art--once it is differentiated as an autopoietic system--must always pre- sent something new, something artistically new; otherwise its communi- cation breaks down or turns into general social communication about artistic quality, prices, the private life of artists, their successes and failures. Operative closure, in other words, requires information in order to move from operation to operation. This is why artworks must convey informa- tion not only in themselves but also in relation to other works--by their novelty or by virtue of the fact that the viewer's observations are not de-
114
Consequently, as we should note in view of what follows,
termined unambiguously and can vary with each observation.
ultimate reward for the complexity of formal arrangements that they pro- vide an opportunity to discover something new, something that strikes us as more astonishing every time we look at them. Conversely, a work that lacks complexity is compelled to offer more conspicuous or, to put it bluntly, more scandalous forms of novelty. Another feature of autopoietic systems is that they are limited to one type of operation, which they must employ for the dual purpose of producing further operations and of creat- ing structures that serve as programs for this production and allow the sys- tem to distinguish between system-immanent/system-external events. An autopoietic system reproduces both its reproduction and the conditions for its reproduction. The environment cannot participate in the reproduc- tion of systems; the manner in which it affects the systems reproduction is never instructive, only destructive. Of course, structural couplings be-
115
tween system and environment are presupposed.
plings, the system could not exist. Works of arts need a material existence. Artists must breathe if communication through art is to be possible. But the effects of dissolving these structural couplings can only be obstructive or destructive. The persistence of such couplings does nothing more than prevent these effects from obstructing the continuation of autopoietic re- production. The evolution of complex systems of this sort displays the complicated structure of a surplus production, of a structure that inhibits
116
and de-inhibits possibilities. We shall take up this point subsequently,
speaking of medium and form.
Taken by itself, the concept of autopoiesis explains very little. It implies that any specification of structures (here, any determination of artistic form) is produced by the system itself; it cannot be imported from the
Without such cou-
It is the
Perception and Communication: The Reproduction of Forms 51
outside. Moreover, the explanation of certain structural developments de- mands further analyses that must take into account structural couplings between autopoietic systems. One cannot derive the existence of worms, birds, and human beings from the singular invention of die autopoiesis of life. The autopoiesis of communication does not tell what kinds of social formations will emerge in the course of evolution. Nor can one infer from die autopoiesis of art which works will be created. The concept's lack of explanatory power stands in disproportional relation to its revolutionizing effect. (This insight could have avoided much controversy. ) Instead of pursuing an ontology or a theory of essences, we follow the injunction: "Indicate the system from which you want to observe the world, draw a distinction, and distinguish yourself from what you observe while ac- knowledging the autological implication diat the same holds when you observe yourself (rather than the external world). "
Needless to say, the autopoiesis of life and the autopoiesis of conscious- ness come about without art, although they may be influenced by art (the brain, for example, or the fingers of a piano player). Neither life nor con- sciousness depends on art in the sense of being unable to reproduce itself without art. The same is true for the communication system of society. We can certainly consider the structural consequences of a society without art. The regeneration of art is autopoietically necessary only for art itself. This is a matter of general agreement, even in aesthetic theories of quite
117
different orientation.
come about without society, without consciousness, without life or mate- rial. But in order to determine how die autopoiesis of art is possible, one must observe the art system and treat everything else as environment.
In the following, we adopt the communication system of art as our sys- tem reference. When using compact concepts such as "observer," "be- holder," "artist," "artwork," and so forth, we are always referring to con- densations of the communication system of art--to the sediments, so to speak, of an ongoing communication that moves from one condensation to the next by means of circumscribed recursions. Artists, artworks, and so on serve a structuring function within the autopoiesis of art. They bun- dle expectations. This is why they are less ephemeral than the basal events
118
of artistic communication.
which operate as events--can reach back and forth while remaining fo- cused on the same thing--on the same work, the same artist, or on the educational qualities of an informed observer. This is not to say that com-
Of course, artistic communication could never
They make sure that communications--
52. Perception and Communication: The Reproduction ofForms
munication reaches down to the physical substratum, to life or conscious- ness, or to the sum total of structural couplings that make such a con- struction of order possible. One can certainly speak of artists as human beings or of artworks as material artifacts; indeed, one would need to the- matize such matters if one's ambition were a complete description of the object. Such a description, however, would have to deploy another system reference or switch back and forth between different system references.
Against this theoretical background, the question of how art communi-
119
cates is no longer trivial. The answer is obvious: by means of artworks. Artistic communication distinguishes itself both from communication that relies exclusively on language and from indirect communications that are either analogous to language or unable to secure the autopoiesis of com- munication, because the communicative intent of conveying information can always be denied. Artistic communication, by contrast, employs per- ceptions that it prepares exclusively for its own use. In so doing, artistic communication realizes specific forms of structural coupling between con- sciousness and society. It communicates by means of distinctions located within the work or by means of forms, for the concept of form, in the sense we use it here, implies a two-sided form, a distinction that can be dis- tinguished. The work of art, then, is anything but an "end in itself. " Nor does it serve external purposes--for example, as ornament. It determines
forms which yield the dual insight that (i) distinctions make possible indi- cations that enter into a play of nonarbitrary combination with other dis- tinctions and indications, and that (a) along with this recognition, it be- comes apparent that this order contains information that is meant to be communicated and understood. Without the fixation of forms in the work and without the opportunity of further actualizations by different ob- servers, a communication of this kind would never occur. It is important that communication can be preserved, just as language is preserved in writ- ing. This does not mean that artistic communication aims at an identical reproduction of the work (consensus and the like! ). The mere fact that the observational sequences that accompany the work's production necessarily differ from those that occur in the perception of the finished work ensures that there can never be a genuine agreement between the two--and yet communication does take place! The artwork makes sure that the observa- tion of observations--a second-order observation--continues, on the side of the producer as well as on the side of the observer of art.
Until now we have focused on how a work of art mediates communica-
Perception and Communication: The Reproduction of Forms 53
tion. An isolated work of art, however, is not yet a communication system of art. We therefore need to ask how and what the individual artwork con-
120
tributes to the social system of art.
The question we raised in conjunc-
tion with the basal elements of artistic communication thus poses itself
again: How do we get beyond the compact communication condensed in
die individual artwork? And how are individual works of art possible
within the reproductive network of art? To be sure, the artwork, in con-
trast to a communicative event negatable in the moment of its occurrence,
is not a basal element of the system; nevertheless, it only comes into being
by virtue of recursive networking with other works of art, with widely dis-
tributed verbal communications about art, with technically reproducible
copies, exhibitions, museums, theaters, buildings, and so forth. This ap-
pears to be beyond dispute. A work of art without other works is as im-
possible as an isolated communication without further communications.
The same holds for different types and genres of art--for sonatas and son-
nets, for statues and still lifes, for novellas as much as for comedies and
121
tragedies.
Works of art, broadly speaking, lead the autopoiesis of art in two direc-
tions and, in so doing, expand and secure its continuation. On. the one hand, one can learn from works of art how to observe and subsequently reintegrate what one has learned into the work's form. One can realize new variants of certain ideas--perhaps in better, more convincing ways and with less of an investment--or one can derive suggestions for new be- ginnings from works that are exhausted. An observer might then confront die task of understanding all of this as being communicated along with die artwork--Manet's black, for example, as a color. On the other hand, these and other issues can become a topic for art-related conversation and writing. One turns to the medium of language while retaining art and its works as a topic. It is well known that the romantics celebrated art criti- cism as the perfection of art, as the production of its history, even as its "medium of reflection" (Benjamin). Whatever one might think of this to- day, the fact that people write and speak about art contributes signifi- candy to the stabilization and destabilization of its autopoiesis up to the remarkable point where the search for a concept of art and the probing of its limits begins to influence the avant-garde and their exploration of forms at the level of the artworks themselves.
? 2 Observation of the First and of the Second Order
I
The activity of observing establishes a distinction in a space that re-
mains unmarked, the space from which the observer executes the distinc-
tion. The observer must employ a distinction in order to generate the dif-
ference between unmarked and marked space, and between himself and
what he indicates. The whole point of this distinction (its intention) is to
mark something as distinct from something else. At the same time, the
observer--in drawing a distinction--makes himself visible to others. He
betrays his presence--even if a further distinction is required to distin-
1
guish him. Once a distinction is established as a form, it points back to
the observer, thus generating both the form's self-reference and its hetero- reference.
The self-referential closure of the form includes the question of the observer as the excluded third.
The forms of possible distinctions are innumerable. But when several observers select a certain distinction, their operations are attuned to one another. What they have in common is generated outside of the form in a manner that remains unspecified. (To call this shared space "consensus" as opposed to "dissent" would require another observer who employs just this distinction. ) This common ground entails the prospect of a formal calculus that leads all participating observers to the same result. We can therefore say that the form is the observer. This notion presupposes the reduplication of the unobservable world in the imaginary space of math- ematical forms. The same procedure--which makes communication pos- sible by creating a shared space--is used in art. Here, too, there is no pres-
54
Observation of the First and of the Second Order 5 5
sure to select a certain distinction and to transform the unobservable world into an imaginary space--now a space of art--by cutting up die world. But when this selection occurs by means of certain artistically de- termined forms, then the observers employing these forms all observe in the same vein. This is how the artist manages to bind the free-floating
23
awareness of other observers. As we indicated earlier, a distinct "object"
is thus secured as an "eigenvalue" of communication. Judgments might still diverge on the basis of different standards of artistic quality, but to deal with such differences one must focus on observing how different ob- servers observe--which is certainly possible.
Each observation immediately observes something that can be distin-
4
guished--objects or events, movements or signs. We cannot get rid of the
immediately given world, although the philosopher may doubt whether it exists at all or whether it exists as it appears and may express his doubt by withholding judgment (Husserl's epoche). Nor can we completely divorce ourselves from the intuitively apprehended world, not even in our imagi- nation; we can only simulate what we might observe under different cir- cumstances. When reading novels, we need to have the text in front of our eyes, even though our "inner eye" can furnish the text with lively details and recall its fictional world once the text is no longer at hand. We know perfectly well tliat no real world corresponds to our imagination, just as we "know away," so to speak, an optical illusion while seeing it nonetheless. Yet, we still follow an experience that accepts the world just the way it could be. Nothing can change that.
We remind the reader of this basic situation in order to introduce the-- not quite simple--distinction between first- and second-order observa- tion. Every observation--this holds for second-order observations as well --uses a distinction to mark one side (but not the other). No procedure can get around that. Even negations presuppose the prior distinction and
5
indication of what one wants to negate. One cannot start from an im-
mediately given nondetermination--an unmarked space, a primordial en- tropy or chaos, an empty canvas or a white sheet of paper--without dis- tinguishing this state from what is being done to it. Even when moving toward fictionality, away from the real world in which we exist, we need this distinction in order to indicate the "whence? " and "where? " This is how we construct reality as reality.
We shall call observations of observations second-order observations. Con- sidered as an operation, the second-order observation is also a first-order
$6 Observation of the First and of the Second Order
observation. It observes something that can be distinguished as observa- tion. Accordingly, there must be structuralcouplingsbetween first- and sec- ond-order observations, which guarantee that something is observed at all in the mode of second-order observation. As usual, there are two sides to the concept of structural coupling: the second-order observer is more irri- table as a result of his first-order observations (of textual features, for ex- ample, or of the characteristics of another observer's observations), while at the same time, he is equipped with a higher degree of indifference against all other conceivable influences.
As first-order observer, the second-order observer remains anchored in the world (and, accordingly, observable). And he sees only what he can distinguish. If he wants to observe from a second-order perspective, then he must be able to distinguish observations from other things (objects, for example).
According to a certain tradition, which, however, we shall not be bound by here, one might say that he must be able to distinguish subjects from objects. But this linguistic usage itself needs explanation and unduly re- stricts the topics we seek to approach. Hence our attempt to use a more formal terminology and to speak of second-order observations only in terms of observations that observe observations. In so doing, we remain at the level of operations. Whether we are dealing with an observation of other observers is a different question. It certainly facilitates the observa- tion of observations when we can hold on to an observer to whom we can attribute these observations. In the case of art, however, this is true only with certain reservations. We may decide to observe a work of art solely in view of its intrinsic observations without observing the artist; it is enough to know or to recognize that we are dealing with an artificial, rather than a natural, object.
The proposition that a second-order observer is always also a first-order
observer rephrases the familiar insight that the world cannot be observed
from the outside. There is no "extramundane subject. " Whoever employs
this figure of thought or wonders how a transcendental subject becomes an
6
empirical subject is thinking in the long shadow of theology or is led
astray by a philosophical theory. As we know from operative epistemol- ogy--which is widely accepted today--the activity of observing occurs in the world and can be observed in turn. It presupposes the drawing of a boundary across which the observer can observe something (or himself as an other), and it accounts for the incompleteness of observations by virtue
Observation of the First and of the Second Order 57
of the fact that the act of observing, along with the difference of the ob- servation that constitutes it, escapes observation. Observation therefore re- lies on a blind spot that enables it to see something (but not everything).
7
A world equipped to observe itself withdraws into unobservability. Or, to
use a more traditional formulation, the unobservability of the observing operation is the transcendental condition of its possibility. The condition of possibility for observation is not a subject (let alone a subject endowed with reason), but rather a paradox that condemns to failure those who want the world to be transparent. Many an artist may have dreamed of making it in another world, while being bound to reproduce the unob- servability of this one.
In order to observe the world as an object, one would have to indicate the world as distinguished from something else; one would have to pre- suppose a metaworld containing the world and its other. What functions as world in each case resists observation--as does the observing operation. The retreat into unobservability leaves nothing behind in the world; it erases, to speak with Jacques Derrida, its own traces. At best metaphysics
8 (or theology? or the rhetorical theory of how to use rhetorical forms? or
a second-order observer? ) may just barely catch a glimpse of "the trace of
9
the erasure of the trace. "
These considerations are meant to irritate philosophers. What matters
to us is deriving foundations (which are no foundations) for an operative concept of observation, so that we can describe more accurately what is going on and what we have to expect when society encourages observers 10 observe observations or even demands that the conditions of social ra- tionality be met at the level of second-order observation.
II
Once the conditions for second-order observation are established, so- ciocultural evolution embarks on a detour that--like the detour of capital according to Bohm-Bawerk--proves to be extraordinarily productive. One restricts one's observation to observing other observers, which opens up possibilities (social psychologists would speak of "vicarious learning") diat are unavailable when the world is observed directly and on the basis of the belief that it is as it appears. Second-order observing maintains a distance from the world until it dismisses the world as unity (wholeness, totality) altogether, henceforth relying entirely on the "eigenvalues" that
58 Observation of the First and of the Second Order
emerge from the dynamic-recursive process of ongoing observations of observations.
This is both true in general terms and a typical trend of all modern functional systems (and of their self-reflection as well). If one looks for further specifications within the larger framework of an operative episte- mology, one discovers a variety of starting points. This is due pardy to the multitude of participating disciplines or research fields, partly to the con- cept of observation itself, which refers to a number of quite disparate em- pirical facts. The operative execution of observations can be described in physical, biological, or sociological terms, whereby in each case potentially disruptive realities come into view as well. As a physicist and mathemati- cian, Heinz von Foerster employs the notion of computing a reality. Humberto Maturana proceeds from a very general, biologically grounded concept of cognition. George Spencer Brown develops a formal calculus that builds on the concept of "indication. " Indication presupposes a dis- tinction, but it can use only one side of this distinction as the starting point of further operations. In semiotics, one would describe the basal op- eration in terms of a sign use that makes the difference between signifier
{signifiant) and signified (signifie) available for operations (primarily, but not only, linguistic ones). Gotthard Giinther investigates logical structures that can describe in adequately complex terms what happens when a sub- ject observes another subject, not as an object but as another subject, that is, as an observer. For others, the problem is how to attribute observations to observers, and diey tend to think of psychological processes that are in- vestigated in attribution research. In the social sciences, too, one typically thinks of psychological manifestations when methodological problems arise. This is because an observer who is doing research is observed in turn and can absorb only what is presented to him as a consequence of the ob- servation of his being observed. Cybernetics, which must be mentioned here, concerns itself with operations of regulation and control, whatever
10
All of these starting points remain capable of dialogue and mutually ac- cessible--although mainly because of an extreme formalization of the con- cept of observation, which is beginning to announce itself in the literature without yet assuming the form of an integrated interdisciplinary theory. By formalization, we mean the coinage of an operative concept with em- pirical reference, which leaves open at what level of reality a given opera- tion occurs and what kinds of realities must be presumed in order to guar-
the apparatus might be that carries out these operations.
Observation of the First and of the Second Order 59
antee that the operation proceeds smoothly and without destructive inter- ference from the outside.
For our purposes, it suffices to define observation with Spencer Brown as the use of a distinction for the purpose of indicating one side and not the other. We disregard any reference, in the concept of observation, to the
11
material conditions that make observation possible.
conditions would explode the concept's unity and lead us into quite dis- parate fields of reality. Moreover, the concept embraces--this distinguishes
12
it from standard usage--experience and action,
both of which depend
on distinguishing and indication (as opposed to mere behavior). As noted
earlier, this conceptual disposition helps describe how artists and observers
participate in communication. We further assume that observation does
not merely happen (in the manner in which an avalanche tears down one
portion of a snow field and leaves the other half intact); if it did, any oper-
ation that yields an effect would be an observation. Instead, the concept
entails that the other side of the distinction is co-presented along with it,
so that the indication of one side is decoded by the system as information
according to the general pattern of "this-and-not-something-else," "this-
and-not-that. " That is to say, one can also observe operations that are not
13
observations.
ating), distinction and indication are executed simultaneously (rather than consecutively, that is, in the sense that a distinction is selected first and then followed by an indication). The observing operation--this is its dis- tinctive feature--realizes the unity of the distinction between distinction and indication. The motivation for this unity resides in observation itself; it does not depend on the prior existence of corresponding objects in a world that exists separately from this unity. Finally, we shall speak of ob- servations only when the indication of one side of a distinction is moti- vated by recursive interconnections--partly by prior observations, hence memory, and pardy through connectivity, that is, by anticipating what one can do with the distinction, where one can go with it, what kinds of pos- sibilities the observation discloses or forecloses. In this sense, observation is always the operation of an observing system (even if this fact remains un-
I observed). Observation cannot happen as a singular event; more precisely, ! when singular events occur, they cannot be observed as observations.
We shall apply this concept consistently whenever its features are given (to an observer), especially when dealing with second-order observation. Perhaps (we leave the question open) a capacity for observation can be at-
In the act of observing (as distinguished from simply oper-
Considering such
60 Observation of the First and of the Second Order
tributed even to chemical processes in living systems, if the necessary "si- multaneity" of distinction and indication can be represented in chemical terms. At the level of nervous systems or immune systems, one might con- sider the capacity for discrimination--which such systems undoubtedly possess--to be a form of observation. Animals certainly observe, and the same is true for systems of human consciousness that process meaning psychologically. One may equally well attribute a capacity for observation to communication systems on the grounds that such systems, in using language, simultaneously handle distinctions and indications. This raises the question as to whether the shaping of material by an artist's hand rep- resents an instance of observation, in that it creates a difference not just for its own sake but in view of a two-sided form, in view of a difference that produces meaning. We shall return to this question below.
What interests us here are the consequences for the notion of second- order observation. We shall speak of second-order observation when two observations are coupled in such a way that both fully realize the features of first-order observation while the second-order observer, in indicating his object, refers to an observer of the first order and, in so doing, distin- guishes and indicates an observation as observation. This raises the fol- lowing question: What needs to be observed from a first-order perspective so that second-order observation becomes possible and can, so to speak, unfold what first-order observation observes directly? Or, what tells us that the activity of observing and indication is going on somewhere? Does it suffice to say that an "observer" must be observed? Or should we prefer formulations that, rather than aiming at compact, self-organizing realities, speak of the materialities involved in the process of observation? Wouldn't this lure us onto dangerous ground in the vain attempt to distinguish mat- ter from "Spirit"?
In the face of such thorny questions, we retreat to a constructivist start- ing point: an observation of the second order is present whenever the fo- cus is on distinctions or, to use a more pointed formulation, when one's own activity of distinguishing and indicating refers to further distinctions and indications. To observe in the mode of second-order observation is to distinguish distinctions--however, not simply by placing distinctions side by side in the manner of "there are large and small objects, pleasant and unpleasant things, theologians and other academics" and so forth ad in- finitum. Rather, what is observed as a distinction--a distinction that both distinguishes and observes--must be observed in operation so that the
Observation of the First and of the Second Order 61
features we identified above as constitutive of die concept of observation come into view: the simultaneity of distinction and indication (keeping an eye on the other side) and their recursive networking with prior and subsequent observations, which, for their part, must also be distinguish- ing indications.
First-order observation is an indication of something in opposition to everything that is not indicated. In this kind of observation, the distinc- tion between distinction and indication is not thematized. The gaze re- mains fixed on the object. The observer and his observing activity remain unobserved. With the occurrence of second-order observation, the situa- tion changes--whether or not the observer is the same. Now the observa- tion indicates that the observation occurs as observation, that it must use a distinction, and perhaps even what kind of distinction it must use. The second-order observer encounters the distinction between distinction and indication. He treats the instrument of observation as the form of obser- vation, which implies that there might be other forms (and observers). The form of observation (there is no need to elaborate this point) also im- plies a reentry of the form into the form, since the distinction employed in the observation presupposes a distinction between distinction and in- dication. This distinction is always already copied into itself as a distinc-
14
tion that distinguishes itself from the indication that it makes possible. The second-order observer does not need to be as complex as to be able to observe this reentry, but he presupposes it as implied in the form he ob- serves as the form of an observation.
The unobservability of first-order observation thus becomes observable in an observation of the second order--on the condition that the second- order observer, considered as first-order observer, can now observe neither his own observing nor himself as observer. A third-order observer can point this out and draw the autological conclusion that all this applies to himself as well. Focusing one's observation on the means of observation-- on artistic means (such as the twelve-tone technique)--excludes a total view of the world. No further reflection can get around that. Nor is there a dialectical Aufhebung that would elevate the blindness of distinguishing into a form of "Spirit" for which the world, including Spirit, becomes fully transparent. Rather, second- and third-order observations explicate the world s unobservability as an unmarked space carried along in all ob- servations. Transparency is paid for with opacity, and this is what ensures the (autopoietic) continuation of the operations, the displacement, the
6 2 Observation of the First and of the Second Order
diffirance (Derrida) of the difference between what is observed and what remains unobserved.
Second-order observation observes only how others observe. Once the question "How? " is posed, a characteristic difference between first- and second-order observations comes into view. The first-order observer con- centrates on what he observes, experiences, and acts out within a horizon of relatively sparse information. He may be puzzled by some things and look for explanations when his expectations go unfulfilled; this, however, is the exception rather than the rule and happens in accordance with his capacity to process information. The first-order observer lives in a world that seems both probable and true [wahr-scheinlich]. By contrast, the sec- ond-order observer notices the improbability [Unwahrscheinlichkeit] of first-order observation. A movement of the hand, a sentence spoken-- every such act is extremely improbable when considered as a selection among all other possibilities. But since this improbability characterizes every operation, it is also quite normal and unproblematic. It remains la- tent in the operation itself, even the operation of first-order observing. There is no need to thematize this improbability, nor can it be thematized. One could never begin if one were to consider all possible beginnings. This holds for second-order observation as well, to the extent that it is an oper- ation. It cannot think through all possible ways of observing an observer before setding on observing a certain observer. Second-order observation recognizes (and experiences in observing itself) that the total information contained in the world cannot be concentrated in one point--unless one assumes a God. But as second-order observation, it can at least thematize the improbability of first-order observation (including its own). It can comprehend more extended realms of selectivity and identify contingen- cies where the first-order observer believes he is following a necessary path
15
or is acting entirely naturally.
say that only the second-order observer notices that the first-order observer "reduces complexity"; it makes no sense to ask a first-order observer to re- duce complexity. Or to put it in yet another way, the world of possibility is an invention of the second-order observer which, for the first-order ob- server, remains necessarily latent.
From the viewpoint of an observer whose reality is guaranteed by his own observations, there are two ways of engaging in second-order obser- vation: self-observation and hetero-observation. Such abstract terminol- ogy has certain advantages. It permits both types of observation to be con-
Simplifying matters somewhat, one might
Observation of the First and of the Second Order 63
sidered as parallel possibilities, as two instances of the same principle, and it calls attention to the necessity of a system reference for the self/hetero distinction. Above all, we begin to sense that there may be a link between the possibility for self-observation, on the one hand, and of second-order observation of others, on the other. If one can see others as observers, then why not oneself, too?
But can observation turn reflexive in this manner? How does such a hyper-formation emerge on top of first-order observation?
It is important to preserve our astonishment in view of this phenomenon, because our goal is to combine second-order observation with a theory of modern society which claims that second-order observation is at once a highly improbable evolutionary fact and an entirely normal occurrence.
Ill
There is nothing extraordinary in the assertion that the functional sys- tems of modern society should establish themselves at the level of second- order observation. Take the system of science [Wissenschaft], for example. Recent investigations of scientific laboratories amply demonstrate that first-order observation plays a role in science and that the behavior of sci-
16
entists can hardly be explained in terms of a "striving for Truth. "
trary to what such research suggests, however, scientific practice*does not preclude second-order observation. The instrument that mediates between first- and second-order observation and ensures their structural coupling is die publication of articles. From a first-order perspective, these are pro- duced and read as texts, but they acquire genuine scientific significance by
17
providing a window onto the observational mode of other scientists. Publishing a text (including summarizing the current state of research and quoting other publications) becomes the basis of scientific produc-
18
tion, the operation of the autopoiesis of science.
meory of science, the code true/untrue along with its own supplementary semantics, the special programs containing theoretical and methodologi- cal directives that rule over the code values true/untrue--all of these be- come meaningful only in relation to texts that are published for the sake of communication. This is how publications secure the continuity of the differentiated system of science at the level of second-order observation.
We can identify similar features in the economic system. Assisted by the
19
market, the economy, too, has shifted toward second-order observation.
The semantics of the
Con-
64 Observation of the First and of the Second Order
This shift requires a focus on a medium of first-order observation. One observes payments in the context of transactions: How much for what? This requires and makes possible variable prices that indicate someone's
20
readiness to sell or buy.
bring about) a temporary stability of prices. This makes second-order ob- servation possible. A player in the market watches others (and himself), asking whether they will buy/sell at a certain price; whether or not it may be worthwhile to produce and invest in view of expectations concerning future market prices. At the same time, the markets for products, raw ma- terials, labor, and money generate situations at the level of second-order observation that are subject to change and need to be monitored con- stantly. Without market-dependent prices, there can be no second-order observation and thus (as socialist state planning learned the hard way) no specific economic rationality. This is why economic theory must distin- guish values and prices, depending on whether it observes an observer of the first or of the second order. And it makes sense to transform values (ecological harmlessness, for example) into prices--not to make sure that such values are implemented, but for the sake of observing how an obser- vation of observations establishes itself under these structural conditions.
Our third example is taken from the political system. Not surprisingly, the same structure is realized in this system as well, albeit in an entirely dif- ferent context. Politics employs power to enable collectively binding deci- sions. At the level where power is exercised, this can be observed directly in institutions specifically designed for this purpose. Classical political the- ory insisted that the ruler ought not to be indifferent to public opinion; speaking with Machiavelli, he must have his fortress in the hearts of his
21
22
The transactions assume (and simultaneously
people.
slaves must observe (whether and) how the master observes them, whereas the master is master only to the extent that he can be content with first- order observation--to the master, the slaves are merely objects who carry
23
out orders or fail to do so.
ment, an asymmetry must be preserved wherever second-order observa- tion occurs--whether it looks down from above or up from below.
In the wake of the so-called democratization of politics and its depen- dence on the media of public opinion, this situation has changed. Today, hierarchy plays a role only at the organizational level. Those participating in politics--politicians and voters alike--observe one another in the mir- ror of public opinion, and behavior is "political" when participants react
In Hegel's dialectic of master and slave, on the other hand,
the
According to the political concept of govern-
Observation of the First and of the Second Order 65
to how they are being observed. The level of first-order observation is guaranteed by the continuous reports of the mass media. Initially, the ef- fects of such media are information and entertainment. Second-order ob- servation occurs via the inferences one can draw about oneself or others, if one assumes that those who wish to participate politically encounter one another in the judgment of public opinion, and that this is sufficient. Public opinion is not an aggregate of psychic system-states, but rather a product of a specific communication that provides the starting point for
24
further communications.
We could supply further examples. In the system of religion God has al-
ways been conceived as an observer, and this is why observing this ob- server became a problem that surfaced in the Devil's fate or the theolo- gian's own, and for some courageous theologians even in the very notion of God. Under the logic of intimacy, modern families (for which there was no concept in the Old World) are hot cells of observation, burdened with a pressure to observe that makes spontaneous behavior difficult and gen-
25
erates either communication routines or pathologies.
tem, the relationship between legislation and jurisdiction is considered in terms of mutual observation, and in the so-called realistic legal doctrine, die law boils down to a mechanism for predicting the decisions of judges (instead of securing the implementation of norms that are recognized as correct). These analyses, evoked for the sake of comparison, cannot be pursued any further here. Our question is, rather, whether art--at least since it differentiated itself as the fine arts against the artes in general--un- folds its own dynamic at the level of second-order observation and distin- guishes itself at this level as a social system from other social systems.
rally abstractedstructure. It is a program for repeated usage which--like to- day's complex computer programs--blocks access to what is actually going on during the execution of the operation. To put it differently, the artwork itself offers no clue as to what happens in the process of understanding the
If,
44 Perception and Communication: The Reproduction ofForms
work. It leaves open how artist and beholder are coupled by the work, while at the same time it guarantees that this coupling is not entirely ar- bitrary--this is what makes art a medium of communication. Physicists would perhaps speak of nonlinear structures of coupling; at any rate, they would exclude tight coupling and noncoupling.
Finally, we should emphasize that our difference-theoretical approach --the idea of tracing art to die unfolding of a paradox of form--implies a radically historical mode of observation: "If words and concepts receive meaning only in sequences of differences, one can justify one's language, and one's choice of terms, only within a topic [an orientation in space] and
104
an historical strategy. "
like everything else, that it comes into being and vanishes. Rather, it ex- plains why works of art must be constructed with an orientation toward time, why they must be new, following the requirement, in place since early modernity, that diey distinguish themselves from everything that has been done before. The demand for novelty means not only that no two things are the same but also that the difference between them motivates both the work's production and die interest of die beholder. Only novel works can please. This is how art exposes itself to die effects of a self-consummation of forms. It positions itself historically. And this is why die future of art be- comes increasingly problematic--to die point where die claim is made and refuted daily that art has no future in the posthistoire.
IX
To determine how art defines itself, one needs to rely on what can be known about works of art. What is it that characterizes art in relation to everything else (to its unmarked space)? The tradition offers a number of specific distinctions. Works of art are made, in contrast to natural objects. Once artificiality as such no longer defines what counts as art, a second distinction is added: works of art serve no externalpurpose; having such a purpose is precisely what disqualifies a work as art. This raises the ques- tion what else could qualify works of art as art. With this question, the theory of art receives its admission ticket to the art system--until die the- ory of die avant-garde will proclaim that art is whatever we call art, thereby accomplishing the dual purpose of making works of art exemplify theory while at the same time relieving itself from the burden of further reflection.
This does not simply mean that art has its time
Perception and Communication: The Reproduction of Forms 45
For lack of a convincing alternative, we shall draw repeatedly on this historically exhausted theoretical framework. However, our considerations concerning perception and communication may yield a somewhat more complex treatment of the topic, particularly via the concept of form.
To say that artistic forms call attention to themselves as being made and without purpose circumscribes the reentry of the form into the form. Any artwork displays talent, no matter whether its figures are beautiful or ugly, distinguished or ordinary, benevolent or malicious, profound or nonsen- sical. The work displays, one might say, itself and its own self-description. It executes the paradox of "reentry" and shows that this is possible, what- ever mathematics or logic may think.
The work of art draws on sensuously perceptible media for its own self- explication, no matter what is subsequently presented as an internal play of forms. It exploits evidence based on such media. Even if one knows that this happens only for the sake of communication, the fact that art draws on perception is not irrelevant to understanding how it happens. The question, then, is the following: How does an individual work of art - present itself to perception in such a way as to be recognizable as art and to provide, by virtue of precisely this recognition, an opportunity and a motive for participating in communication?
The concept of form suggests that two requirements must be fulfilled and inscribed into perception: the form must have a boundary, and there must be an "unmarked space" excluded by this boundary. How both of these requirements coincide and how they are fulfilled in one stroke may vary considerably in different artistic genres. Whenever we think of "marked space"/boundary/"unmarked space" in combination, the constitution of an imaginary space is at stake. However, since every artwork constitutes its own imaginary space, the question is how this space is constituted differ- ently in each case.
The typical case is a work of art enclosed within a beginning and end, within a frame or a stage, a work that ignores and does not interfere with its surroundings. In this case, the imaginary space is construed from the inside out, as if breaking through the frame or creating its own world be-
105
hind the frame. The imagination is driven beyond the work.
at once see and think away the frame in order to gain access to the work's imaginary space. The guaranteed repeatability of observations might be of help in performing this operation.
Sculpture or architecture presents an entirely different case. Here, the
One must
46 Perception and Communication: The Reproduction ofForms
boundary does not draw the viewer's attention inward but instead directs it outward. The work permits no view into its depths, no penetration of its surface (whatever the surface may betray of the work's mass, volume, or material). The imaginary space is projected outward in the form of dis- tinctions suggested by the work itself. Here, too, space is work-specific space, visible so long as the focus is on the work and disappearing from view when the focus shifts to surrounding objects--to the weeds in the casde garden.
106
The boundary cannot be perceived,
rects our attention inward or outward. The boundary itself can be shaped as a form--as a portal, an ornament, or a movement on the sculpture's surface, as the gorgeous frame of a painting or merely as a well-chosen one. Once we recognize this, however, we no longer see the boundary as boundary but observe differences between forms, which--one thing lead- ing to another--we attribute to the work itself.
Classical aesthetics, which presupposed a creative or receptive subject in all of this, had no difficulties with these problems. Everything could be ex- plained with reference to the enigmatic nature of the subject. The strict distinction between perception and communication, which dissolves the subject, changes this situation. Now it becomes important that, and in what ways, the boundaries of the individual work of art mark the struc- tural coupling between perception and communication. Boundaries, however, cannot be observed in their capacity as structural couplings, since neither the perceiving consciousness nor communication can ex- plode their operative closure and reach beyond their own systems into the environment.
Giving up the notion of the subject requires reconstructing the object,
107
which loses its opposite.
"unmarked space," objects appear as repeated indications, which, rather than having a specific opposite, are demarcated against "everything else. " Objects are forms whose other side remains undetermined. The unattain- ability of its other side accounts for the object's concreteness in the sense that determining its unity "as something" becomes impossible. Every analysis remains partial, depending on further specification of the other side--for example, with regard to color, magnitude, purpose, consistency.
George Herbert Mead (following Whitehead) assigned a function to identifiable and recognizable objects, whose primary purpose is to bind time. This function is needed because the reality of experience and actions
unless we know whether it di-
If one starts out from the counterconcept of the
Perception and Communication: The Reproduction of Forms 47
consists in mere event sequences, that is, in an ongoing self-dissolution. Since the experience of others must be assumed to occur simultaneously if communication (Mead: interaction) is to take place, the reality of the other's operations remains in principle inaccessible; it can, however, be
108
symbolized via the identification of objects.
Michel Serres suggests that the stabilization of objects (identification,
recognizability, and so on) is more likely to contribute to stabilizing social
109
relationships than the famous social contract.
Heinz von Foerster arrives at the notion of objects as eigen-behaviors of
110
recursive calculations.
cursive self-application of communication contribute more than any other kinds of norms and sanctions to supplying the social system with necessary redundancies. This may be even more true of objects that have been invented for the sake of this specific function, such as kings or soc-
111
112
Presumably, the objects that emerge from the re-
cer balls.
to this function. They assume a sufficient amount of variance, of recog- nizability in different situations, to keep up with changing social constel- lations. In contrast to concepts that are determined by specified ant- onyms, such quasi objects maintain their concreteness as objects--in the sense of excluding the unmarked space of all other events and conditions --throughout changing situations.
Works of art are quasi objects in this sense. They individualize them- selves by excluding the sum total of everything else; not because they are construed as given but because their significance as objects implies a realm of social regulation. One must scrutinize works of art as intensely and with as close attention to the object as one does when watching kings and soccer balls; in this way--and in the more complex case, where one ob- serves other observers by focusing on the same object--the socially regu- lative reveals itself. The object relation thus helps differentiate recursive connections between observations--the court, the soccer game, the art scene--which subsequently construct their own guiding objects.
This is how the exclusion of the unmarked space is carried along--and forgotten. It might as well be left to religion.
X
Consciousness cannot communicate, communication cannot perceive --this is where we began. Correspondences between modes of operation
Such "quasi objects"
can be comprehended only in relation
Via a different route,
48 Perception and Communication: The Reproduction ofForms
as disparate as these must be formulated in a highly abstract manner. Af- ter all, we are dealing with two entirely different, operatively closed sys- tems that do not interfere with each other. With concepts such as the self- referential event, distinction, form, and paradox, we have gained the necessary level of abstraction. This conceptual background, as we sug- gested earlier, permits conclusions concerning the specificity of art. Art
j makes perception available for communication, and it does so outside the standardized forms of a language (that, for its part, is perceptible). Art cannot overcome the separation between psychic and social systems. Both types of system remain operatively inaccessible to each other. And this ac- countsfor the significance of art. Art integrates perception and communica- tion without merging or confusing their respective operations. Integration means nothing more than that disparate systems operate simultaneously (are synchronized) and constrain one another's freedom. By participating in communication via perception, the psychic system might generate in- tensities of experience that remain incommunicable as such. In order to produce such experiences, it must be able to perceive differences between the forms created in the social system of art for the purpose of communi- cation. Communication through art must present sensuously perceptible objects without being able to reproduce itself within individually encap- sulated psychic systems. The need for structural coupling and the oppor- tunity to establish such coupling impose rigorous demands upon the forms that specify and determine a work of art at the boundary between psychic and social systems.
A kind of quantum mechanical solution to the problem of integration
is conceivable if one thinks of forms in terms of a distinction between two sides. One can always count on the other, operatively inaccessible system to operate in a binary manner, as a system that indicates one side of an ac- tualized form and excludes (for the time being) the other. That much one can expect from communication with regard to perception, and from per- ception with regard to communication, although the rich, reference-filled inner horizons of these systems remain mutually inaccessible. Forms, in other words, ensure identity and difference at once: identity through their fixed outlines, and difference through the recursive system reference of the op- erations that actualize these outlines--in the form of a contrasting per- ception or intuition, or of a place where communication can continue by understanding and reconstructing its connective possibilities.
Since works of art are objects that bind time, such an integration can be
Perception and Communication: The Reproduction of Forms 49
synchronized. It outlives the sheer eventfulness of system operations for as long as a consciousness is occupied with the work of art. Since the focus is on an object, this can happen recursively, that is, by recapitulating and anticipating other perceptions of forms. Recursions of this type generate so-called "aha" experiences, sudden flashes of insight that reveal the or- dered nexus of the artwork. In this case, the coupling is unique in that it does not require a merging of psychic and social systems. Consciousness
remains entirely within itself.
Our analysis has reached a point where it becomes apparent that, and in what ways, communication through art tends toward system formation and eventually differentiates a social system of art. In the following chap- ters, we shall deal both with the historicity of this inner-social process and with its consequences. For the time being, our concern is to show, on the basis of our analysis of artistic communication, that, and in what ways, system formation is possible--much as the spectacular individual exis- tence of artworks and the diffuse heterogeneity of observer perspectives may speak against such an approach.
The problem of system formation is connectivity, the recursive reusabil- ity of events. Operations (conscious perceptions as well as communica- tions) are nothing more than events. They cannot persist, nor can they be altered. They emerge and vanish in the same instant, taking no more time than is needed to fulfill the function of an element that cannot be decom- posed any further. The art system has no reality except at the level of ele- mental events. It rests, one might say, on the ongoing dissolution of its el- ements, on the transitory nature of its communications, on an all-pervasive entropy against which anything that persists must organize itself. Concepts such as connectivity or recursive reusability indicate this process, but they do not explain it. They only show that the stability of a system based on time-sensitive events must be a dynamic stability, a stability that depends on the continual change of the system's resources.
We shall call this state of affairs an "autopoietic system. " This means that the elements of the system are produced within the network of the System's elements, that is, through recursions. A communication cannot occur as an isolated phenomenon, as a singular event brought about by a combination of physical, chemical, living, and psychic causes. Nor can it proceed through simple replication, merely by substituting disappearing elements for one another. It is not enough--and it wouldn't work, any- way--to repeat what has been said (shown, perceived, or thought) once it
50 Perception and Communication: The Reproduction ofForms
dies away. Something else, something new must follow, for the informa- tional component of communication presupposes surprise and gets lost in
113
repetition.
art--once it is differentiated as an autopoietic system--must always pre- sent something new, something artistically new; otherwise its communi- cation breaks down or turns into general social communication about artistic quality, prices, the private life of artists, their successes and failures. Operative closure, in other words, requires information in order to move from operation to operation. This is why artworks must convey informa- tion not only in themselves but also in relation to other works--by their novelty or by virtue of the fact that the viewer's observations are not de-
114
Consequently, as we should note in view of what follows,
termined unambiguously and can vary with each observation.
ultimate reward for the complexity of formal arrangements that they pro- vide an opportunity to discover something new, something that strikes us as more astonishing every time we look at them. Conversely, a work that lacks complexity is compelled to offer more conspicuous or, to put it bluntly, more scandalous forms of novelty. Another feature of autopoietic systems is that they are limited to one type of operation, which they must employ for the dual purpose of producing further operations and of creat- ing structures that serve as programs for this production and allow the sys- tem to distinguish between system-immanent/system-external events. An autopoietic system reproduces both its reproduction and the conditions for its reproduction. The environment cannot participate in the reproduc- tion of systems; the manner in which it affects the systems reproduction is never instructive, only destructive. Of course, structural couplings be-
115
tween system and environment are presupposed.
plings, the system could not exist. Works of arts need a material existence. Artists must breathe if communication through art is to be possible. But the effects of dissolving these structural couplings can only be obstructive or destructive. The persistence of such couplings does nothing more than prevent these effects from obstructing the continuation of autopoietic re- production. The evolution of complex systems of this sort displays the complicated structure of a surplus production, of a structure that inhibits
116
and de-inhibits possibilities. We shall take up this point subsequently,
speaking of medium and form.
Taken by itself, the concept of autopoiesis explains very little. It implies that any specification of structures (here, any determination of artistic form) is produced by the system itself; it cannot be imported from the
Without such cou-
It is the
Perception and Communication: The Reproduction of Forms 51
outside. Moreover, the explanation of certain structural developments de- mands further analyses that must take into account structural couplings between autopoietic systems. One cannot derive the existence of worms, birds, and human beings from the singular invention of die autopoiesis of life. The autopoiesis of communication does not tell what kinds of social formations will emerge in the course of evolution. Nor can one infer from die autopoiesis of art which works will be created. The concept's lack of explanatory power stands in disproportional relation to its revolutionizing effect. (This insight could have avoided much controversy. ) Instead of pursuing an ontology or a theory of essences, we follow the injunction: "Indicate the system from which you want to observe the world, draw a distinction, and distinguish yourself from what you observe while ac- knowledging the autological implication diat the same holds when you observe yourself (rather than the external world). "
Needless to say, the autopoiesis of life and the autopoiesis of conscious- ness come about without art, although they may be influenced by art (the brain, for example, or the fingers of a piano player). Neither life nor con- sciousness depends on art in the sense of being unable to reproduce itself without art. The same is true for the communication system of society. We can certainly consider the structural consequences of a society without art. The regeneration of art is autopoietically necessary only for art itself. This is a matter of general agreement, even in aesthetic theories of quite
117
different orientation.
come about without society, without consciousness, without life or mate- rial. But in order to determine how die autopoiesis of art is possible, one must observe the art system and treat everything else as environment.
In the following, we adopt the communication system of art as our sys- tem reference. When using compact concepts such as "observer," "be- holder," "artist," "artwork," and so forth, we are always referring to con- densations of the communication system of art--to the sediments, so to speak, of an ongoing communication that moves from one condensation to the next by means of circumscribed recursions. Artists, artworks, and so on serve a structuring function within the autopoiesis of art. They bun- dle expectations. This is why they are less ephemeral than the basal events
118
of artistic communication.
which operate as events--can reach back and forth while remaining fo- cused on the same thing--on the same work, the same artist, or on the educational qualities of an informed observer. This is not to say that com-
Of course, artistic communication could never
They make sure that communications--
52. Perception and Communication: The Reproduction ofForms
munication reaches down to the physical substratum, to life or conscious- ness, or to the sum total of structural couplings that make such a con- struction of order possible. One can certainly speak of artists as human beings or of artworks as material artifacts; indeed, one would need to the- matize such matters if one's ambition were a complete description of the object. Such a description, however, would have to deploy another system reference or switch back and forth between different system references.
Against this theoretical background, the question of how art communi-
119
cates is no longer trivial. The answer is obvious: by means of artworks. Artistic communication distinguishes itself both from communication that relies exclusively on language and from indirect communications that are either analogous to language or unable to secure the autopoiesis of com- munication, because the communicative intent of conveying information can always be denied. Artistic communication, by contrast, employs per- ceptions that it prepares exclusively for its own use. In so doing, artistic communication realizes specific forms of structural coupling between con- sciousness and society. It communicates by means of distinctions located within the work or by means of forms, for the concept of form, in the sense we use it here, implies a two-sided form, a distinction that can be dis- tinguished. The work of art, then, is anything but an "end in itself. " Nor does it serve external purposes--for example, as ornament. It determines
forms which yield the dual insight that (i) distinctions make possible indi- cations that enter into a play of nonarbitrary combination with other dis- tinctions and indications, and that (a) along with this recognition, it be- comes apparent that this order contains information that is meant to be communicated and understood. Without the fixation of forms in the work and without the opportunity of further actualizations by different ob- servers, a communication of this kind would never occur. It is important that communication can be preserved, just as language is preserved in writ- ing. This does not mean that artistic communication aims at an identical reproduction of the work (consensus and the like! ). The mere fact that the observational sequences that accompany the work's production necessarily differ from those that occur in the perception of the finished work ensures that there can never be a genuine agreement between the two--and yet communication does take place! The artwork makes sure that the observa- tion of observations--a second-order observation--continues, on the side of the producer as well as on the side of the observer of art.
Until now we have focused on how a work of art mediates communica-
Perception and Communication: The Reproduction of Forms 53
tion. An isolated work of art, however, is not yet a communication system of art. We therefore need to ask how and what the individual artwork con-
120
tributes to the social system of art.
The question we raised in conjunc-
tion with the basal elements of artistic communication thus poses itself
again: How do we get beyond the compact communication condensed in
die individual artwork? And how are individual works of art possible
within the reproductive network of art? To be sure, the artwork, in con-
trast to a communicative event negatable in the moment of its occurrence,
is not a basal element of the system; nevertheless, it only comes into being
by virtue of recursive networking with other works of art, with widely dis-
tributed verbal communications about art, with technically reproducible
copies, exhibitions, museums, theaters, buildings, and so forth. This ap-
pears to be beyond dispute. A work of art without other works is as im-
possible as an isolated communication without further communications.
The same holds for different types and genres of art--for sonatas and son-
nets, for statues and still lifes, for novellas as much as for comedies and
121
tragedies.
Works of art, broadly speaking, lead the autopoiesis of art in two direc-
tions and, in so doing, expand and secure its continuation. On. the one hand, one can learn from works of art how to observe and subsequently reintegrate what one has learned into the work's form. One can realize new variants of certain ideas--perhaps in better, more convincing ways and with less of an investment--or one can derive suggestions for new be- ginnings from works that are exhausted. An observer might then confront die task of understanding all of this as being communicated along with die artwork--Manet's black, for example, as a color. On the other hand, these and other issues can become a topic for art-related conversation and writing. One turns to the medium of language while retaining art and its works as a topic. It is well known that the romantics celebrated art criti- cism as the perfection of art, as the production of its history, even as its "medium of reflection" (Benjamin). Whatever one might think of this to- day, the fact that people write and speak about art contributes signifi- candy to the stabilization and destabilization of its autopoiesis up to the remarkable point where the search for a concept of art and the probing of its limits begins to influence the avant-garde and their exploration of forms at the level of the artworks themselves.
? 2 Observation of the First and of the Second Order
I
The activity of observing establishes a distinction in a space that re-
mains unmarked, the space from which the observer executes the distinc-
tion. The observer must employ a distinction in order to generate the dif-
ference between unmarked and marked space, and between himself and
what he indicates. The whole point of this distinction (its intention) is to
mark something as distinct from something else. At the same time, the
observer--in drawing a distinction--makes himself visible to others. He
betrays his presence--even if a further distinction is required to distin-
1
guish him. Once a distinction is established as a form, it points back to
the observer, thus generating both the form's self-reference and its hetero- reference.
The self-referential closure of the form includes the question of the observer as the excluded third.
The forms of possible distinctions are innumerable. But when several observers select a certain distinction, their operations are attuned to one another. What they have in common is generated outside of the form in a manner that remains unspecified. (To call this shared space "consensus" as opposed to "dissent" would require another observer who employs just this distinction. ) This common ground entails the prospect of a formal calculus that leads all participating observers to the same result. We can therefore say that the form is the observer. This notion presupposes the reduplication of the unobservable world in the imaginary space of math- ematical forms. The same procedure--which makes communication pos- sible by creating a shared space--is used in art. Here, too, there is no pres-
54
Observation of the First and of the Second Order 5 5
sure to select a certain distinction and to transform the unobservable world into an imaginary space--now a space of art--by cutting up die world. But when this selection occurs by means of certain artistically de- termined forms, then the observers employing these forms all observe in the same vein. This is how the artist manages to bind the free-floating
23
awareness of other observers. As we indicated earlier, a distinct "object"
is thus secured as an "eigenvalue" of communication. Judgments might still diverge on the basis of different standards of artistic quality, but to deal with such differences one must focus on observing how different ob- servers observe--which is certainly possible.
Each observation immediately observes something that can be distin-
4
guished--objects or events, movements or signs. We cannot get rid of the
immediately given world, although the philosopher may doubt whether it exists at all or whether it exists as it appears and may express his doubt by withholding judgment (Husserl's epoche). Nor can we completely divorce ourselves from the intuitively apprehended world, not even in our imagi- nation; we can only simulate what we might observe under different cir- cumstances. When reading novels, we need to have the text in front of our eyes, even though our "inner eye" can furnish the text with lively details and recall its fictional world once the text is no longer at hand. We know perfectly well tliat no real world corresponds to our imagination, just as we "know away," so to speak, an optical illusion while seeing it nonetheless. Yet, we still follow an experience that accepts the world just the way it could be. Nothing can change that.
We remind the reader of this basic situation in order to introduce the-- not quite simple--distinction between first- and second-order observa- tion. Every observation--this holds for second-order observations as well --uses a distinction to mark one side (but not the other). No procedure can get around that. Even negations presuppose the prior distinction and
5
indication of what one wants to negate. One cannot start from an im-
mediately given nondetermination--an unmarked space, a primordial en- tropy or chaos, an empty canvas or a white sheet of paper--without dis- tinguishing this state from what is being done to it. Even when moving toward fictionality, away from the real world in which we exist, we need this distinction in order to indicate the "whence? " and "where? " This is how we construct reality as reality.
We shall call observations of observations second-order observations. Con- sidered as an operation, the second-order observation is also a first-order
$6 Observation of the First and of the Second Order
observation. It observes something that can be distinguished as observa- tion. Accordingly, there must be structuralcouplingsbetween first- and sec- ond-order observations, which guarantee that something is observed at all in the mode of second-order observation. As usual, there are two sides to the concept of structural coupling: the second-order observer is more irri- table as a result of his first-order observations (of textual features, for ex- ample, or of the characteristics of another observer's observations), while at the same time, he is equipped with a higher degree of indifference against all other conceivable influences.
As first-order observer, the second-order observer remains anchored in the world (and, accordingly, observable). And he sees only what he can distinguish. If he wants to observe from a second-order perspective, then he must be able to distinguish observations from other things (objects, for example).
According to a certain tradition, which, however, we shall not be bound by here, one might say that he must be able to distinguish subjects from objects. But this linguistic usage itself needs explanation and unduly re- stricts the topics we seek to approach. Hence our attempt to use a more formal terminology and to speak of second-order observations only in terms of observations that observe observations. In so doing, we remain at the level of operations. Whether we are dealing with an observation of other observers is a different question. It certainly facilitates the observa- tion of observations when we can hold on to an observer to whom we can attribute these observations. In the case of art, however, this is true only with certain reservations. We may decide to observe a work of art solely in view of its intrinsic observations without observing the artist; it is enough to know or to recognize that we are dealing with an artificial, rather than a natural, object.
The proposition that a second-order observer is always also a first-order
observer rephrases the familiar insight that the world cannot be observed
from the outside. There is no "extramundane subject. " Whoever employs
this figure of thought or wonders how a transcendental subject becomes an
6
empirical subject is thinking in the long shadow of theology or is led
astray by a philosophical theory. As we know from operative epistemol- ogy--which is widely accepted today--the activity of observing occurs in the world and can be observed in turn. It presupposes the drawing of a boundary across which the observer can observe something (or himself as an other), and it accounts for the incompleteness of observations by virtue
Observation of the First and of the Second Order 57
of the fact that the act of observing, along with the difference of the ob- servation that constitutes it, escapes observation. Observation therefore re- lies on a blind spot that enables it to see something (but not everything).
7
A world equipped to observe itself withdraws into unobservability. Or, to
use a more traditional formulation, the unobservability of the observing operation is the transcendental condition of its possibility. The condition of possibility for observation is not a subject (let alone a subject endowed with reason), but rather a paradox that condemns to failure those who want the world to be transparent. Many an artist may have dreamed of making it in another world, while being bound to reproduce the unob- servability of this one.
In order to observe the world as an object, one would have to indicate the world as distinguished from something else; one would have to pre- suppose a metaworld containing the world and its other. What functions as world in each case resists observation--as does the observing operation. The retreat into unobservability leaves nothing behind in the world; it erases, to speak with Jacques Derrida, its own traces. At best metaphysics
8 (or theology? or the rhetorical theory of how to use rhetorical forms? or
a second-order observer? ) may just barely catch a glimpse of "the trace of
9
the erasure of the trace. "
These considerations are meant to irritate philosophers. What matters
to us is deriving foundations (which are no foundations) for an operative concept of observation, so that we can describe more accurately what is going on and what we have to expect when society encourages observers 10 observe observations or even demands that the conditions of social ra- tionality be met at the level of second-order observation.
II
Once the conditions for second-order observation are established, so- ciocultural evolution embarks on a detour that--like the detour of capital according to Bohm-Bawerk--proves to be extraordinarily productive. One restricts one's observation to observing other observers, which opens up possibilities (social psychologists would speak of "vicarious learning") diat are unavailable when the world is observed directly and on the basis of the belief that it is as it appears. Second-order observing maintains a distance from the world until it dismisses the world as unity (wholeness, totality) altogether, henceforth relying entirely on the "eigenvalues" that
58 Observation of the First and of the Second Order
emerge from the dynamic-recursive process of ongoing observations of observations.
This is both true in general terms and a typical trend of all modern functional systems (and of their self-reflection as well). If one looks for further specifications within the larger framework of an operative episte- mology, one discovers a variety of starting points. This is due pardy to the multitude of participating disciplines or research fields, partly to the con- cept of observation itself, which refers to a number of quite disparate em- pirical facts. The operative execution of observations can be described in physical, biological, or sociological terms, whereby in each case potentially disruptive realities come into view as well. As a physicist and mathemati- cian, Heinz von Foerster employs the notion of computing a reality. Humberto Maturana proceeds from a very general, biologically grounded concept of cognition. George Spencer Brown develops a formal calculus that builds on the concept of "indication. " Indication presupposes a dis- tinction, but it can use only one side of this distinction as the starting point of further operations. In semiotics, one would describe the basal op- eration in terms of a sign use that makes the difference between signifier
{signifiant) and signified (signifie) available for operations (primarily, but not only, linguistic ones). Gotthard Giinther investigates logical structures that can describe in adequately complex terms what happens when a sub- ject observes another subject, not as an object but as another subject, that is, as an observer. For others, the problem is how to attribute observations to observers, and diey tend to think of psychological processes that are in- vestigated in attribution research. In the social sciences, too, one typically thinks of psychological manifestations when methodological problems arise. This is because an observer who is doing research is observed in turn and can absorb only what is presented to him as a consequence of the ob- servation of his being observed. Cybernetics, which must be mentioned here, concerns itself with operations of regulation and control, whatever
10
All of these starting points remain capable of dialogue and mutually ac- cessible--although mainly because of an extreme formalization of the con- cept of observation, which is beginning to announce itself in the literature without yet assuming the form of an integrated interdisciplinary theory. By formalization, we mean the coinage of an operative concept with em- pirical reference, which leaves open at what level of reality a given opera- tion occurs and what kinds of realities must be presumed in order to guar-
the apparatus might be that carries out these operations.
Observation of the First and of the Second Order 59
antee that the operation proceeds smoothly and without destructive inter- ference from the outside.
For our purposes, it suffices to define observation with Spencer Brown as the use of a distinction for the purpose of indicating one side and not the other. We disregard any reference, in the concept of observation, to the
11
material conditions that make observation possible.
conditions would explode the concept's unity and lead us into quite dis- parate fields of reality. Moreover, the concept embraces--this distinguishes
12
it from standard usage--experience and action,
both of which depend
on distinguishing and indication (as opposed to mere behavior). As noted
earlier, this conceptual disposition helps describe how artists and observers
participate in communication. We further assume that observation does
not merely happen (in the manner in which an avalanche tears down one
portion of a snow field and leaves the other half intact); if it did, any oper-
ation that yields an effect would be an observation. Instead, the concept
entails that the other side of the distinction is co-presented along with it,
so that the indication of one side is decoded by the system as information
according to the general pattern of "this-and-not-something-else," "this-
and-not-that. " That is to say, one can also observe operations that are not
13
observations.
ating), distinction and indication are executed simultaneously (rather than consecutively, that is, in the sense that a distinction is selected first and then followed by an indication). The observing operation--this is its dis- tinctive feature--realizes the unity of the distinction between distinction and indication. The motivation for this unity resides in observation itself; it does not depend on the prior existence of corresponding objects in a world that exists separately from this unity. Finally, we shall speak of ob- servations only when the indication of one side of a distinction is moti- vated by recursive interconnections--partly by prior observations, hence memory, and pardy through connectivity, that is, by anticipating what one can do with the distinction, where one can go with it, what kinds of pos- sibilities the observation discloses or forecloses. In this sense, observation is always the operation of an observing system (even if this fact remains un-
I observed). Observation cannot happen as a singular event; more precisely, ! when singular events occur, they cannot be observed as observations.
We shall apply this concept consistently whenever its features are given (to an observer), especially when dealing with second-order observation. Perhaps (we leave the question open) a capacity for observation can be at-
In the act of observing (as distinguished from simply oper-
Considering such
60 Observation of the First and of the Second Order
tributed even to chemical processes in living systems, if the necessary "si- multaneity" of distinction and indication can be represented in chemical terms. At the level of nervous systems or immune systems, one might con- sider the capacity for discrimination--which such systems undoubtedly possess--to be a form of observation. Animals certainly observe, and the same is true for systems of human consciousness that process meaning psychologically. One may equally well attribute a capacity for observation to communication systems on the grounds that such systems, in using language, simultaneously handle distinctions and indications. This raises the question as to whether the shaping of material by an artist's hand rep- resents an instance of observation, in that it creates a difference not just for its own sake but in view of a two-sided form, in view of a difference that produces meaning. We shall return to this question below.
What interests us here are the consequences for the notion of second- order observation. We shall speak of second-order observation when two observations are coupled in such a way that both fully realize the features of first-order observation while the second-order observer, in indicating his object, refers to an observer of the first order and, in so doing, distin- guishes and indicates an observation as observation. This raises the fol- lowing question: What needs to be observed from a first-order perspective so that second-order observation becomes possible and can, so to speak, unfold what first-order observation observes directly? Or, what tells us that the activity of observing and indication is going on somewhere? Does it suffice to say that an "observer" must be observed? Or should we prefer formulations that, rather than aiming at compact, self-organizing realities, speak of the materialities involved in the process of observation? Wouldn't this lure us onto dangerous ground in the vain attempt to distinguish mat- ter from "Spirit"?
In the face of such thorny questions, we retreat to a constructivist start- ing point: an observation of the second order is present whenever the fo- cus is on distinctions or, to use a more pointed formulation, when one's own activity of distinguishing and indicating refers to further distinctions and indications. To observe in the mode of second-order observation is to distinguish distinctions--however, not simply by placing distinctions side by side in the manner of "there are large and small objects, pleasant and unpleasant things, theologians and other academics" and so forth ad in- finitum. Rather, what is observed as a distinction--a distinction that both distinguishes and observes--must be observed in operation so that the
Observation of the First and of the Second Order 61
features we identified above as constitutive of die concept of observation come into view: the simultaneity of distinction and indication (keeping an eye on the other side) and their recursive networking with prior and subsequent observations, which, for their part, must also be distinguish- ing indications.
First-order observation is an indication of something in opposition to everything that is not indicated. In this kind of observation, the distinc- tion between distinction and indication is not thematized. The gaze re- mains fixed on the object. The observer and his observing activity remain unobserved. With the occurrence of second-order observation, the situa- tion changes--whether or not the observer is the same. Now the observa- tion indicates that the observation occurs as observation, that it must use a distinction, and perhaps even what kind of distinction it must use. The second-order observer encounters the distinction between distinction and indication. He treats the instrument of observation as the form of obser- vation, which implies that there might be other forms (and observers). The form of observation (there is no need to elaborate this point) also im- plies a reentry of the form into the form, since the distinction employed in the observation presupposes a distinction between distinction and in- dication. This distinction is always already copied into itself as a distinc-
14
tion that distinguishes itself from the indication that it makes possible. The second-order observer does not need to be as complex as to be able to observe this reentry, but he presupposes it as implied in the form he ob- serves as the form of an observation.
The unobservability of first-order observation thus becomes observable in an observation of the second order--on the condition that the second- order observer, considered as first-order observer, can now observe neither his own observing nor himself as observer. A third-order observer can point this out and draw the autological conclusion that all this applies to himself as well. Focusing one's observation on the means of observation-- on artistic means (such as the twelve-tone technique)--excludes a total view of the world. No further reflection can get around that. Nor is there a dialectical Aufhebung that would elevate the blindness of distinguishing into a form of "Spirit" for which the world, including Spirit, becomes fully transparent. Rather, second- and third-order observations explicate the world s unobservability as an unmarked space carried along in all ob- servations. Transparency is paid for with opacity, and this is what ensures the (autopoietic) continuation of the operations, the displacement, the
6 2 Observation of the First and of the Second Order
diffirance (Derrida) of the difference between what is observed and what remains unobserved.
Second-order observation observes only how others observe. Once the question "How? " is posed, a characteristic difference between first- and second-order observations comes into view. The first-order observer con- centrates on what he observes, experiences, and acts out within a horizon of relatively sparse information. He may be puzzled by some things and look for explanations when his expectations go unfulfilled; this, however, is the exception rather than the rule and happens in accordance with his capacity to process information. The first-order observer lives in a world that seems both probable and true [wahr-scheinlich]. By contrast, the sec- ond-order observer notices the improbability [Unwahrscheinlichkeit] of first-order observation. A movement of the hand, a sentence spoken-- every such act is extremely improbable when considered as a selection among all other possibilities. But since this improbability characterizes every operation, it is also quite normal and unproblematic. It remains la- tent in the operation itself, even the operation of first-order observing. There is no need to thematize this improbability, nor can it be thematized. One could never begin if one were to consider all possible beginnings. This holds for second-order observation as well, to the extent that it is an oper- ation. It cannot think through all possible ways of observing an observer before setding on observing a certain observer. Second-order observation recognizes (and experiences in observing itself) that the total information contained in the world cannot be concentrated in one point--unless one assumes a God. But as second-order observation, it can at least thematize the improbability of first-order observation (including its own). It can comprehend more extended realms of selectivity and identify contingen- cies where the first-order observer believes he is following a necessary path
15
or is acting entirely naturally.
say that only the second-order observer notices that the first-order observer "reduces complexity"; it makes no sense to ask a first-order observer to re- duce complexity. Or to put it in yet another way, the world of possibility is an invention of the second-order observer which, for the first-order ob- server, remains necessarily latent.
From the viewpoint of an observer whose reality is guaranteed by his own observations, there are two ways of engaging in second-order obser- vation: self-observation and hetero-observation. Such abstract terminol- ogy has certain advantages. It permits both types of observation to be con-
Simplifying matters somewhat, one might
Observation of the First and of the Second Order 63
sidered as parallel possibilities, as two instances of the same principle, and it calls attention to the necessity of a system reference for the self/hetero distinction. Above all, we begin to sense that there may be a link between the possibility for self-observation, on the one hand, and of second-order observation of others, on the other. If one can see others as observers, then why not oneself, too?
But can observation turn reflexive in this manner? How does such a hyper-formation emerge on top of first-order observation?
It is important to preserve our astonishment in view of this phenomenon, because our goal is to combine second-order observation with a theory of modern society which claims that second-order observation is at once a highly improbable evolutionary fact and an entirely normal occurrence.
Ill
There is nothing extraordinary in the assertion that the functional sys- tems of modern society should establish themselves at the level of second- order observation. Take the system of science [Wissenschaft], for example. Recent investigations of scientific laboratories amply demonstrate that first-order observation plays a role in science and that the behavior of sci-
16
entists can hardly be explained in terms of a "striving for Truth. "
trary to what such research suggests, however, scientific practice*does not preclude second-order observation. The instrument that mediates between first- and second-order observation and ensures their structural coupling is die publication of articles. From a first-order perspective, these are pro- duced and read as texts, but they acquire genuine scientific significance by
17
providing a window onto the observational mode of other scientists. Publishing a text (including summarizing the current state of research and quoting other publications) becomes the basis of scientific produc-
18
tion, the operation of the autopoiesis of science.
meory of science, the code true/untrue along with its own supplementary semantics, the special programs containing theoretical and methodologi- cal directives that rule over the code values true/untrue--all of these be- come meaningful only in relation to texts that are published for the sake of communication. This is how publications secure the continuity of the differentiated system of science at the level of second-order observation.
We can identify similar features in the economic system. Assisted by the
19
market, the economy, too, has shifted toward second-order observation.
The semantics of the
Con-
64 Observation of the First and of the Second Order
This shift requires a focus on a medium of first-order observation. One observes payments in the context of transactions: How much for what? This requires and makes possible variable prices that indicate someone's
20
readiness to sell or buy.
bring about) a temporary stability of prices. This makes second-order ob- servation possible. A player in the market watches others (and himself), asking whether they will buy/sell at a certain price; whether or not it may be worthwhile to produce and invest in view of expectations concerning future market prices. At the same time, the markets for products, raw ma- terials, labor, and money generate situations at the level of second-order observation that are subject to change and need to be monitored con- stantly. Without market-dependent prices, there can be no second-order observation and thus (as socialist state planning learned the hard way) no specific economic rationality. This is why economic theory must distin- guish values and prices, depending on whether it observes an observer of the first or of the second order. And it makes sense to transform values (ecological harmlessness, for example) into prices--not to make sure that such values are implemented, but for the sake of observing how an obser- vation of observations establishes itself under these structural conditions.
Our third example is taken from the political system. Not surprisingly, the same structure is realized in this system as well, albeit in an entirely dif- ferent context. Politics employs power to enable collectively binding deci- sions. At the level where power is exercised, this can be observed directly in institutions specifically designed for this purpose. Classical political the- ory insisted that the ruler ought not to be indifferent to public opinion; speaking with Machiavelli, he must have his fortress in the hearts of his
21
22
The transactions assume (and simultaneously
people.
slaves must observe (whether and) how the master observes them, whereas the master is master only to the extent that he can be content with first- order observation--to the master, the slaves are merely objects who carry
23
out orders or fail to do so.
ment, an asymmetry must be preserved wherever second-order observa- tion occurs--whether it looks down from above or up from below.
In the wake of the so-called democratization of politics and its depen- dence on the media of public opinion, this situation has changed. Today, hierarchy plays a role only at the organizational level. Those participating in politics--politicians and voters alike--observe one another in the mir- ror of public opinion, and behavior is "political" when participants react
In Hegel's dialectic of master and slave, on the other hand,
the
According to the political concept of govern-
Observation of the First and of the Second Order 65
to how they are being observed. The level of first-order observation is guaranteed by the continuous reports of the mass media. Initially, the ef- fects of such media are information and entertainment. Second-order ob- servation occurs via the inferences one can draw about oneself or others, if one assumes that those who wish to participate politically encounter one another in the judgment of public opinion, and that this is sufficient. Public opinion is not an aggregate of psychic system-states, but rather a product of a specific communication that provides the starting point for
24
further communications.
We could supply further examples. In the system of religion God has al-
ways been conceived as an observer, and this is why observing this ob- server became a problem that surfaced in the Devil's fate or the theolo- gian's own, and for some courageous theologians even in the very notion of God. Under the logic of intimacy, modern families (for which there was no concept in the Old World) are hot cells of observation, burdened with a pressure to observe that makes spontaneous behavior difficult and gen-
25
erates either communication routines or pathologies.
tem, the relationship between legislation and jurisdiction is considered in terms of mutual observation, and in the so-called realistic legal doctrine, die law boils down to a mechanism for predicting the decisions of judges (instead of securing the implementation of norms that are recognized as correct). These analyses, evoked for the sake of comparison, cannot be pursued any further here. Our question is, rather, whether art--at least since it differentiated itself as the fine arts against the artes in general--un- folds its own dynamic at the level of second-order observation and distin- guishes itself at this level as a social system from other social systems.
